INTERPLAY OF HOST AND PATHOGEN FACTORS DETERMINES IMMUNITY AND CLINICAL OUTCOME FOLLOWING CAMPYLOBACTER JEJUNI INFECTION IN MICE By Jean Marie Brudvig A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of Comparative Medicine and Integrative Biology—Doctor of Philosophy 2018 INTERPLAY OF HOST AND PATHOGEN FACTORS DETERMINES IMMUNITY AND CLINICAL OUTCOME FOLLOWING CAMPYLOBACTER JEJUNI INFECTION IN MICE ABSTRACT By Jean Marie Brudvig Campylobacter jejuni is a common cause of bacterial enteritis worldwide, and is associated with the post-infectious neuropathy Guillain-Barré syndrome (GBS). The currently accepted pathogenesis of C. jejuni-associated GBS involves generation of cross-reactive antibodies to the C. jejuni LOS and structurally similar gangliosides enriched in peripheral nerves. Complex host-pathogen interactions determine innate and adaptive immunity and corresponding disease outcomes. The overarching aim of this study was to determine how these interactions underlie contrasting immune responses to C. jejuni in mice, beginning with the initial interaction of C. jejuni with dendritic cells (DCs) in vitro and assessing the impact on development of adaptive immunity and disease outcome in vivo. The model system was developed to exploit both C. jejuni strain differences and immunogenetic biases reported for C57BL/6 and BALB/c mice. C. jejuni strains associated with colitis (C. jejuni 11168) and GBS (C. jejuni 260.94) were used in combination with C57BL/6 mice, a reportedly Th1-biased strain, and BALB/c mice, a reportedly Th2-biased strain. C. jejuni strains were evaluated for invasion efficiency, intracellular survival, and elicitation of pro-inflammatory and Th-polarizing cytokine production in vitro using bone marrow-derived DCs (BMDCs) from C57BL/6 and BALB/c mice. In vivo models were used to assess how local and systemic adaptive immunity, leading to disease outcomes including colitis and production of anti-ganglioside antibodies, vary by both C. jejuni strain and mouse genetic background. BALB/c wild-type and IL-10-/- mice infected with C. jejuni 260.94 were first assessed as a potential GBS model. C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 were previously shown to mount Th2-mediated immunity and produce anti-ganglioside antibodies, and BALB/c mice are reportedly Th2-biased. BALB/c mice were therefore expected to mount strong Th2-mediated responses, produce anti-ganglioside antibodies, and develop neurological deficits. Instead, infected BALB/c mice developed systemic Th1/Th17-mediated immunity, and did not develop anti-ganglioside antibodies or neurological disease. The contrasting immune response to C. jejuni 260.94 demonstrated by mice of two different genetic backgrounds highlights the important and incompletely understood role of host factors in determining immunity to C. jejuni. The second in vivo study was designed to further explore these contrasting immune responses. Both C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were orally infected with colitogenic C. jejuni 11168 or GBS-associated C. jejuni 260.94. Immunity and disease outcome were C. jejuni strain-specific in BALB/c IL-10-/- mice: C. jejuni 11168 stimulated strong Th1/Th17-mediated immunity and colitis with greater transmigration of C. jejuni through intestinal layers and higher numbers of intracellular organisms. C. jejuni 260.94 was less invasive and induced less robust Th1/Th17 responses, without colitis or anti- ganglioside antibody production. These findings were mirrored in vitro, as C. jejuni 11168 demonstrated higher invasion efficiency, intracellular survival ability, and stimulated more robust production of IL-6 by BALB/c IL-10-/- BMDCs than C. jejuni 260.94. The findings in BALB/c IL-10-/- mice indicate that outcome of infection is dependent upon the infecting C. jejuni strain, and that DCs have a potential early role in driving the immune response. Interestingly, infected C57BL/6 IL-10-/- mice in this study colonized with C. jejuni but developed neither the previously reported severe colitis following C. jejuni 11168 infection nor the Th2-mediated immunity and anti-ganglioside antibody production after C. jejuni 260.94 infection. Lactobacillus murinus was cultured from all C57BL/6 IL-10-/- mice, but no BALB/c IL-10-/- mice, at the end of the study. Further studies evaluating L. murinus as a potential probiotic candidate are warranted. Results of the current studies demonstrate that immunity and disease outcome following C. jejuni infection depend upon the characteristics of both the infecting C. jejuni strain and host genetic background, and implicate DCs as early contributors to adaptive immunity. Copyright by JEAN MARIE BRUDVIG 2018 ACKNOWLEDGEMENTS I would like to acknowledge my advisor, Dr. Linda Mansfield, for her expertise, guidance, and encouragement throughout this study. I am grateful to Dr. Julia Bell for her assistance in multiple aspects of this study. I appreciate her thoroughness and efficiency, and am indebted to her for the excellent technical assistance with laboratory procedures, her statistical advice, and for our productive discussions. My co-hort of graduate students (Dr. Phil Brooks, Dr. Barbie Gadsden, Dr. Ankit Malik) provided a supportive and team-based environment that was both enjoyable and mutually productive. The incoming graduate students (Dan Claiborne, Zack Johnson, Hinako Terauchi) also contributed to an enjoyable laboratory and office environment. I have had the pleasure of working with and in some cases mentoring several undergraduate and veterinary students (Jenna Baker, Xavier Brandon, James Chen, Matt Cluett, Alex Ethridge, Elizabeth Gensterblum, and Keenan O’Dea) and I thank them for their hard work and contributions to my research. My sincere thanks go to my committee members: Dr. Kathleen Hoag, Dr. Shannon Manning, Dr. Adam Moeser, and Dr. Jon Patterson. Each member of my committee provided thoughtful feedback and guidance throughout this study, and I am grateful for their encouragement and expertise. I thank Robert Crawford for his excellent expertise and technical assistance with flow cytometry, and for our productive discussions. I am indebted to my family. Their love, support, patience, encouragement, belief in me, and understanding made this work possible. v TABLE OF CONTENTS LIST OF TABLES…………………………………………………………………………………………………………………………….....…..viii LIST OF FIGURES………………………………………………………………………………………………………………………………………x KEY TO ABBREVIATIONS………………………………………………………………………………………………………………………..xii CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW……………………………………………………………………...…1 CAMPYLOBACTER JEJUNI: HISTORY AND CLASSIFICATION……………………………………………………..…1 ENTERIC CAMPYLOBACTERIOSIS: EPIDEMIOLOGY AND DISEASE MANIFESTATIONS………………...2 EXTRA-INTESTINAL COMPLICATIONS AND POST-INFECTIOUS SEQUELAE OF CAMPYLOBACTER JEJUNI INFECTION: EMPHASIS ON GUILLAIN-BARRÉ SYNDROME…………………………………………………………………………………………………………………...7 HOST RESPONSE IN CAMPYLOBACTERIOSIS: INITIAL HOST-MICROBE INTERACTION AND INNATE IMMUNITY……………………………………………………………………………………………..15 ADAPTIVE IMMUNITY TO CAMPYLOBACTER JEJUNI………………………………………………………………..24 RATIONALE FOR STUDY……………………………………………………………………………………………………………26 REFERENCES……………………………………………………………………………………………………………………………..31 CHAPTER 2: BALB/C MICE INFECTED WITH GUILLAIN-BARRÉ SYNDROME-ASSOCIATED CAMPYLOBACTER JEJUNI STRAIN 260.94 EXHIBIT TH1/TH17-MEDIATED IMMUNITY.............................43 ABSTRACT…………………………………………………………………………………………………………………………………43 INTRODUCTION………………………………………………………………………………………………………………………..45 MATERIALS AND METHODS……………………………………………………………………………………………………..49 RESULTS……………………………………………………………………………………………………………………………………60 DISCUSSION……………………………………………………………………………………………………………………………..66 APPENDIX………………………………………………………………………………………………………………………………...76 REFERENCES……………………………………………………………………………………………………………………………..83 CHAPTER 3: INFECTING CAMPYLOBACTER JEJUNI STRAIN DETERMINES TH1/TH17-MEDIATED IMMUNITY AND COLITIS IN INTERLEUKIN-10-DEFICIENT BALB/C MICE………………………………………………..90 ABSTRACT………………………………………………………………………………………………………………………………...90 INTRODUCTION………………………………………………………………………………………………………………………..92 MATERIALS AND METHODS……………………………………………………………………………………………………..96 RESULTS………………………………………………………………………………………………………………………………….108 DISCUSSION…………………………………………………………………………………………………………………………...121 APPENDIX……………………………………………………………………………………………………………………………....137 REFERENCES…………………………………………………………………………………………………………………………...151 CHAPTER 4: INVASION EFFICIENCY, INTRACELLULAR SURVIVAL, AND ELICITATION OF CYTOKINE PRODUCTION IN MURINE DENDRITIC CELLS IS DETERMINED BY BOTH CAMPYLOBACTER JEJUNI STRAIN CHARACTERISTICS AND MOUSE GENOTYPE………………………............158 ABSTRACT……………………………………………………………………………………………………………………………….158 INTRODUCTION……………………………………………………………………………………………………………………...160 MATERIALS AND METHODS…………………………………………………………………………………………………...166 vi RESULTS………………………………………………………………………………………………………………………………….172 DISCUSSION…………………………………………………………………………………………………………………………...180 APPENDICES……………………………………………………………………………………………………………………………195 APPENDIX A: TABLES AND FIGURES…………………………………………………………………………..196 APPENDIX B: SUPPLEMENTAL DATA………………………………………………………………………….213 REFERENCES…………………………………………………………………………………………………………………………...214 CHAPTER 5: SUMMARY AND FUTURE DIRECTIONS…………………………………………………………………………….221 SUMMARY……………………………………………………………………………………………………………………………..221 FUTURE DIRECTIONS……………………………………………………………………………………………………………...228 REFERENCES…………………………………………………………………………………………………………………………...233 vii LIST OF TABLES Table 2.1. Results of DigiGait analysis…………………………………………………………………………………………………..77 Table 3.1. Experimental design……………………………………………………………………………………………….............138 Table 4.1. Mean ± Standard Deviation of parameters determined by flow cytometric analysis of three independent experiments, BMDCs……………………………………………………………..196 Table 4.2A. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 1 hour invasion time followed by 1 hour exposure to gentamicin and immediate lysis………………………………………………………………………...197 Table 4.2B. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 2 hours invasion time followed by 1 hour exposure to gentamicin and immediate lysis…………………………………………………………………..…….198 Table 4.2C. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 3 hours invasion time, followed by 1 hour exposure to gentamicin and immediate lysis…………………………………………………………………..…….199 Table 4.2D. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 23 hours invasion time, followed by 1 hour exposure to gentamicin and immediate lysis…………………………………………………………………..…….200 Table 4.3A. Results of gentamicin killing assays evaluating intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of wild-type (WT) mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; one hour of invasion followed by gentamicin treatment, with lysis at 24 hours post-infection…………………………………………………………………………………………………………………..…….201 Table 4.3B. Results of gentamicin killing assays evaluating intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; one hour of invasion followed by gentamicin treatment, with lysis at viii 24 hours post-infection…………………………………………………………………………………………………….……202 Table 4.4. Cytokine analysis, first independent experiment……………………………………………………………….203 Table 4.5. Cytokine analysis, second independent experiment………………………………………………………….204 Table 4.6. Cytokine analysis, third independent experiment……………………………………………………….…….206 Table A.1. Results of additional GKAs assessing intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of wild-type or interleukin-10-/- mice on C57BL/6 and BALB/c genetic backgrounds…………………………..…….213 ix LIST OF FIGURES Figure 2.1. Culture results of Campylobacter jejuni 260.94 colonization of the cecum at the time of necropsy………………………………………………………………………………………………………………………78 Figure 2.2. Gross pathology and ileocecocolic junction histopathology……………………………………….………79 Figure 2.3. Assessment of plasma anti-C. jejuni and anti-ganglioside IgG isotypes………………………………80 Figure 2.4. Th cytokine production in the proximal colon…………………………………………………………….………81 Figure 2.5. Quantification of macrophages in dorsal root ganglia………………………………………………..………82 Figure 3.1. Survival of C57BL/6 interleukin (IL)-10-/- and BALB/c IL-10-/- mice inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB)………………………………..…….139 Figure 3.2. Culture results, cecum.………………………………………………………………………………………………..……140 Figure 3.3. Gross pathology noted in the cecum, colon, mesenteric lymph nodes (MLN), or spleen at the time of necropsy in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice following inoculation with C. jejuni 11168, C. jejuni 260.94, or sham inoculation (tryptic soy broth; TSB)...................................................................................................................................141 Figure 3.4. Colon histopathology………………………………………………………………………………………………….…….142 Figure 3.5. Plasma anti-C. jejuni IgG1, IgG2b, IgG3 for all mice, IgG2c (C57BL/6 IL-10-/- only), and IgG2a (BALB/c IL-10-/- only) antibodies measured by ELISA in samples taken at necropsy…………………………………………………………………………………………………………………………..…….143 Figure 3.6. Plasma anti-GM1 IgG1, IgG2b, IgG3 for all mice, and IgG2c (C57BL/6 IL-10-/- only), and IgG2a (BALB/c IL-10-/- only) antibodies measured by ELISA in samples taken at necropsy…………………………………………………………………………………………………………………………..…….144 Figure 3.7. Plasma anti-GD1a IgG1, IgG2b, IgG3 in all mice, and IgG2c (C57BL/6 IL-10-/- only), and IgG2a (BALB/c IL-10-/- only) antibodies measured by ELISA in samples taken at necropsy…………………………………………………………………………………………………………………………..…….145 Figure 3.8. Measurement of anti-C. jejuni specific IgA in supernatants of feces collected at necropsy from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice inoculated with C. jejuni 260.94, C. jejuni 11168, or sham inoculated (tryptic soy broth; TSB)……………………………….……146 Figure 3.9. Assessment of colon cytokine production in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice inoculated with C. jejuni 11168, C. jejuni 260.94, or sham (tryptic soy broth, x TSB) by flow cytometry-based multiplexed bead assay……………………………………………..............147 Figure 3.10. Assessment of colon cytokine production in C57BL/6 IL-10-/- mice inoculated with C. jejuni 11168, C. jejuni 260.94, or sham (tryptic soy broth, TSB) by flow cytometry-based multiplexed bead assay……………………………………………………………………….…….148 Figure 3.11. Assessment of colon cytokine production in BALB/c IL-10-/- mice inoculated with C. jejuni 11168, C. jejuni 260.94, or sham (tryptic soy broth, TSB) by flow cytometry-based multiplexed bead assay……………………………………………………………………….…….149 Figure 3.12. Assessment of cells positively labeled with the F4/80 macrophage marker in lumbar dorsal root ganglia (DRG) of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice at the time of necropsy…………………………………………………………………………………………………………………….150 Figure 4.1. Analysis of CD11c and MHC II expression by flow cytometry……………………………………..…….208 Figure 4.2. Analysis of CD11c, MHC II, and F4/80 expression by flow cytometry……………………………….209 Figure 4.3. Differences in invasion efficiency and intracellular survival of two Campylobacter jejuni strains in bone marrow-derived dendritic cells (BMDCs) from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice assessed by gentamicin killing assay……………………………………………….210 Figure 4.4. Differences in intracellular survival of two C. jejuni strains in bone marrow- derived dendritic cells (BMDCs) from wild-type (WT) and IL-10-/- mice on both C57BL/6 and BALB/c backgrounds as assessed by gentamicin killing assay…………………….…….211 Figure 4.5. Cytokine production in C. jejuni-infected bone marrow-derived dendritic cells (BMDCs) derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice……………………………………..……212 xi KEY TO ABBREVIATIONS µg: Microgram µL: Microliter µm: Micrometer µM: Micromolar AIDP: Acute Inflammatory Demyelinating Polyneuropathy AMAN: Acute Motor Axonal Neuropathy AMSAN: Acute Motor Sensory Axonal Neuropathy ANOVA: Analysis of Variance APC: Antigen Presenting Cells ATCC: American Type Culture Collection BA: Bolton Agar BMDC: Bone Marrow-Derived Dendritic Cell BSA: Bovine Serum Albumin CCR: CC Chemokine Receptor CCV: Campylobacter jejuni-Containing Vacuole CD: Cluster of Differentiation cDC: Conventional Dendritic Cells CDT: Cytolethal Distending Toxin CFA: Freund’s Complete Adjuvant CFU: Colony Forming Unit Cia: Campylobacter Invasion Antigen CIDP: Chronic Inflammatory Demyelinating Polyneuropathy xii cm: Centimeter CV%: Coefficient of Variation CXCR: C-X-C Motif Chemokine Receptor DC: Dendritic Cell(s) deg: Degrees DNA: Deoxyribonucleic Acid DRG: Dorsal Root Ganglion/Ganglia DSS: Dextran Sulphate Sodium EAN: Experimental Autoimmune Neuritis EDTA: Ethylenediaminetetraacetic Acid ELISA: Enzyme-Linked Immunosorbent Assay FBS: Fetal Bovine Serum Fiji: Fiji Is Just ImageJ FITC: Fluorescein Isothiocyanate Foxp3: Forkhead Box P3 Fr: French g: Gauge g: Gram g: Gravity GBS: Guillain-Barré Syndrome GI: Gastrointestinal GKA: Gentamicin Killing Assay GM-CSF: Granulocyte-Macrophage Colony Stimulating Factor H&E: Hematoxylin and Eosin xiii HBSS: Hank’s Balanced Salt Solution HEPES: 4-(2-hydroxyethyl)-1-piperazineethanesulfonic Acid Hg: Mercury HLA: Human Leukocyte Antigen HMW: High Molecular Weight IBS: Irritable Bowel Syndrome ICJ: Ileocecocolic Junction IFN: Interferon Ig: Immunoglobulin IHC: Immunohistochemistry IL: Interleukin iNOS: Inducible Nitric Oxide Synthetase IP-10: IFN-Gamma-Inducible Protein 10 KO: Knock-Out LF: Left Front LH: Left Hind LMW: Low Molecular Weight LOS: Lipooligosaccharide LPS: Lipopolysaccharide LTβ: Lymphotoxin-Beta MALDI-TOF: Matrix-Assisted Laser Desorption/Ionization Time of Flight MALP-2: Macrophage-Activating Lipopeptide 2 MAP: Mitogen Activated Protein Mbp: Megabase Pair xiv MCP-1: Monocyte Chemoattractant Protein 1 Med: Medium MFI: Median Fluorescence Intensity MFS: Miller Fisher Syndrome mg: Milligram MHC: Major Histocompatibility Complex mL: Milliliter MLN: Mesenteric Lymph Node mM: Millimolar MMP: Matrix Metalloproteinase MOI: Multiplicity of Infection mRNA: Messenger Ribonucleic Acid MSU: Michigan State University MyD88: Myeloid Differentiation Primary Response 88 N: Normal NAD: Normal Antibody Diluent NCTC: National Collection of Type Cultures NF-κB: Nuclear Factor Kappa B ng: Nanogram NLRP3: Nucleotide-Binding Domain, Leucine-Rich-Containing Family, Pyrin Domain-Containing-3 nm: Nanometer NMR: Nuclear Magnetic Resonance NO: Nitric Oxide NOD: Nonobese Diabetic xv NOD: Nucleotide-Binding Oligomerization Domain ns: Non-significant OD: Optical Density OFT: Open Field Test p.i.: Post Infection PBS: Phosphate Buffered Saline PCR: Polymerase Chain Reaction PE: R-Phycoerythrin pg: Picogram PI 3: Phosphoinositide 3 PRR: Pattern Recognition Receptor rcf: Relative Centrifugal Force RF: Right Front RH: Right Hind RPMI: Roswell Park Memorial Institute s: Seconds SAPP: Spontaneous Autoimmune Peripheral Polyneuropathy SCID: Severe Combined Immunodeficiency SD: Standard Deviation SEM: Standard Error of the Mean Siglec: Sialic Acid-Binding Ig-Like Lectins spp: Species ssp: Subspecies T3SS: Type 3 Secretion System xvi TBS: Tris Buffered Saline Tfh: T Follicular Helper TGF: Transforming Growth Factor Th: T helper TLR: Toll-Like Receptors TMB: Tetramethylbenzidine TNF: Tumor Necrosis Factor Treg: T Regulatory TRIF: Toll-Interleukin 1 (IL-1) Receptor Domain-Containing Adaptor-Inducing Beta Interferon (IFN-β) PAMP: Pathogen-Associated Molecular Pattern TSAB: Tryptic Soy Agar with Blood TSAB-CVA: Tryptic Soy Agar with Blood, containing Cefoperazone, Vancomycin, Amphotericin B TSB: Tryptic Soy Broth Tx Grp: Treatment Group u: Unit URCF: University Research Containment Facility WT: Wild-Type YAMC: Young Adult Mouse Colon xvii CHAPTER 1: INTRODUCTION AND LITERATURE REVIEW CAMPYLOBACTER JEJUNI: HISTORY AND CLASSIFICATION Campylobacter jejuni is a leading cause of human food-borne bacterial diarrhea worldwide, and is associated with numerous post-infectious sequelae. C. jejuni is a member of the class Epsilonproteobacteria and family Campylobacteraceae. The genus Campylobacter was first described in 1886, became a separate genus from Vibrio in 1963,27 and, in 1973, C. fetus, C. coli, C. jejuni, and C. sputorum species were classified within the genus.27 Characteristics of Campylobacter spp. include a slender rod shape which may vary from straight, curved, or spiral; cells of most species (excepting C. gracilis) are motile.27 Campylobacters are Gram-negative and non-spore-forming. Campylobacter spp., C. jejuni and C. coli in particular, are currently considered one of the most common causes of human bacterial gastroenteritis in the United States and worldwide.22; 102; 121 However, difficulties in isolating the organism prevented successful identification and classification for many years. Two major clinical and laboratory contributions furthering our understanding of this organism occurred in the 1970s: use of a filtration technique allowing successful isolation of “related vibrio” from human stool in addition to blood samples28 and development of selective culture medium containing vancomycin, polymyxin B, and trimethoprim.107 These two advancements contributed to the identification of Campylobacter as a hitherto unrecognized common cause of bacterial enteritis. Since its initial discovery, classification, and rise to prominence as an important etiologic agent of diarrhea worldwide, an additional milestone in understanding the biology of C. jejuni was the publication of the first C. jejuni genome by Parkhill et al. in 2000.93 Sequencing of the enteritis- associated C. jejuni NCTC11168 strain revealed a small (approximately 1.6 Mbp) genome with several interesting features, including hypervariable sequences found in genes involved in surface structure synthesis or modification.93 The genome sequences of numerous other C. jejuni strains, including some associated with various clinical manifestations such as the human post-infectious peripheral neuropathy 1 Guillain-Barré syndrome,92 abortion in livestock,119 that of the highly virulent and commonly studied C. jejuni 81-176,47 and many others have since become available as complete or draft genome sequences. ENTERIC CAMPYLOBACTERIOSIS: EPIDEMIOLOGY AND DISEASE MANIFESTATIONS Animal Infection. Campylobacter spp. are well recognized causes of bovine diarrhea, reproductive disease including abortion and infertility in cattle and sheep, and gastroenteritis in dogs and cats.67; 89 However, many mammals and birds may be asymptomatically colonized with C. jejuni or C. coli.114 Sources of infection to humans thus include contaminated meat (especially poultry), unpasteurized milk, and inadequately treated water. Poultry in particular are important reservoirs for human infection with C. jejuni, with high numbers of C. jejuni (104-107 CFU/g) detected in the ceca in natural90 and experimental9 infections. Interestingly, in chickens C. jejuni heavily colonizes the crypts of the cecum, large intestine, and cloaca, but appears to preferentially colonize the mucus layer, without attachment to microvilli and incites little to no pathology experimentally.9 Domestic poultry are essentially asymptomatic carriers,69 and the high pathogen load poses a public health risk from contamination during processing at the abattoir. The infectious dose of C. jejuni is as low as 800 organisms, determined in healthy human volunteers.15 Human Campylobacteriosis Epidemiology. The vast majority of human campylobacteriosis has historically been attributed to infection with C. jejuni or C. coli.67 Campylobacter is considered by the World Health Organization as one of four key global causes of diarrheal disease, and the most common etiology of human bacterial gastroenteritis worldwide.121 Recent estimates from surveillance data suggest that more than 800,000 cases of domestically acquired foodborne illness, resulting in over 8,000 hospitalizations and more than 70 deaths, are caused by Campylobacter spp. each year in the United States.102 2 Human Clinical Manifestations. Different clinical manifestations of enteric campylobacteriosis are recognized in developing and industrialized countries.42 Repeated infections are common in children in developing countries, with clinical severity and duration of colonization and excretion decreasing with age.16; 114 Fewer symptomatic infections occur in older children and adults, consistent with protective immunity resulting after multiple exposures.42 In contrast, infections in patients in developed countries, while generally self-limiting, are clinically severe, with fever, diarrhea, and abdominal cramping.42 The diarrhea is often bloody; leukocytes in stool aid in clinical diagnosis. The most common clinical features reported in community outbreaks included diarrhea, abdominal pain, fever, myalgia, and headache.16 Histological changes are consistent with acute infectious colitis and may vary depending upon stage of infection.112 Predominant histopathology includes neutrophilic infiltration, ulceration, presence of exudate, cryptitis and crypt abscesses, edema and inflammatory cell infiltration of the lamina propria.112 Animal Models of Gastrointestinal C. jejuni infection. The relatively recent recognition of C. jejuni as an important human pathogen and the lack of robust, reproducible small animal models of C. jejuni-induced colitis hindered our understanding of the pathogenesis and immune responses of campylobacteriosis until recently. Though chickens are natural hosts of C. jejuni and a significant reservoir for human infection, outcomes of infection and colonization in domestic poultry do not parallel the disease induced in humans. Several studies of C. jejuni infection in immunocompetent adult mice of different genetic backgrounds with conventional murine microbiotas have resulted in colonization, either transient or stable, but without induction of disease as seen in humans.13; 18; 23; 76-78 Possible reasons immunocompetent mice appear resistant to C. jejuni-induced enteritis include both murine factors (cross reactive immunity against other enteric bacteria, inherent characteristics of the murine microbiome), and C. jejuni factors (differences in infecting strain, number of in vitro passages).78 Building on early investigations, ferret, rabbit, and murine models were subsequently developed that produce enteric symptoms and pathology mimicking human disease. Ferrets infected with C. jejuni 3 strain CG8421 developed diarrhea and colon histopathology resembling human disease, including epithelial cell loss and infiltration of neutrophils and macrophages.86 Cimetidine-treated infant rabbits infected with C. jejuni NCTC 11168 developed diarrhea and swollen, red, distended intestines; histopathological and ultrastructural changes included edema, capillary congestion in cecal villi, epithelial damage, and infiltration of intraepithelial lymphocytes.104 Mouse Models of Human Colitis. Four landmark mouse models exploiting alterations in the immune system, microbiota, or both have furthered our understanding of C. jejuni colonization and pathogenesis. Mansfield et al (2007) showed that both wild-type (WT) C57BL/6 and congenic interleukin (IL)-10-/- mice infected with C. jejuni 11168 stably colonized; the WT mice did not develop clinical signs or pathology, while the IL-10-/- mice exhibited severe typhlocolitis.76 Histopathology resembled that seen in human disease, including marked polymorphonuclear and mononuclear inflammation in the lamina propria and epithelial erosion and ulceration.76 Fox and colleagues (2004) infected both WT and NF-κB- deficient (3X) C57BL/129 mice with C. jejuni 81-176 (WT and cdtB knockout (KO) strains).33 They found that both WT and KO C. jejuni 81-176 persistently infected 3X, but not WT, mice at 2—4 months p.i. Histopathology, primarily characterized by gastritis and duodenitis, was most severe in 3X mice infected with the WT C. jejuni 11168.33 Chang and Miller (2006) observed that following oral inoculation with C. jejuni 81-176, WT C3H mice with normal flora did not colonize as efficiently as C3H mice harboring a defined, limited enteric flora consisting of Clostridial species, Lactobacillus, and Acinetobacter. Furthermore, SCID mice (effectively lacking mature B and T lymphocytes) also harboring the limited flora remained persistently colonized, and in contrast to WT mice with the limited flora, exhibited severe inflammation, particularly in the cecum.23 Notable lesions included inflammatory cell infiltration in the lamina propria and submucosa, accompanied by edema, ulceration, and hemorrhage.23 Finally, mice with a “humanized” microbiota perorally infected with C. jejuni ATCC 43431 exhibited stable colonization and pronounced inflammatory responses in the colon including increased apoptotic cells, 4 neutrophils, T and B lymphocytes, and T regulatory (Treg) cells, while mice with a conventional murine microbiota cleared C. jejuni in 2-3 days.12 These studies have greatly enhanced our understanding of the role of the immune system, specifically IL-10, NF-κB (transcription factor involved in immune cell activation) and mature T and B lymphocytes, and the mouse intestinal microbiota in relative protection of mice following C. jejuni infection. C. jejuni Virulence Factors. Campylobacter jejuni has numerous virulence factors contributing to pathogenicity, survival, and modulation of the host response. Among factors shown to be important for host-pathogen interactions are motility, secretion, adhesion, a polysaccharide capsule, cytolethal distending toxin (CDT), and lipooligosaccharide (LOS). Flagella. C. jejuni, through a combination of flagellum-mediated movement and spiral morphology, retains superior motility compared to other bacterial rods (E. coli, V. cholerae, and S. enteriditis) in solutions of high viscosity.32 Retention of motility in highly viscous environments, such as mucus in the gastrointestinal tract, likely contributes to C. jejuni pathogenesis through colonization of the mucus layer. In addition to its critical role in motility, the flagellum also functions in attachment to the epithelium81 and in secretion. Sequencing of the C. jejuni NCTC11168 genome did not identify type III secretion systems (T3SS) other than the flagellin export apparatus;93 secretion of Campylobacter invasion antigen (Cia) proteins necessary for invasion, production of IL-8 by epithelial cells, and induction of disease occurs through the flagellar export apparatus.65; 101 The hook, basal body, and at least one of two filaments of the apparatus must be functional for secretion.65 Adherence and Capsule. Adherence to host cells protects bacteria from elimination from the gastrointestinal tract by normal peristalsis, and is important for entry into host cells. C. jejuni are able to adhere to fibronectin, a component of the extracellular matrix, through the outer membrane protein CadF.63 The polysaccharide capsule of C. jejuni also influences virulence by contributing to adherence and invasion, colonization, and modulation of the host immune response. A C. jejuni 81-176 mutant 5 lacking high molecular weight (HMW) capsular glycan (kpsM mutant) showed reduced invasion and adherence to INT407 (human embryo intestinal epithelial) cells in vitro compared to WT C. jejuni 81- 176.7 This same mutant strain exhibited reduced colonization in BALB/c mice79 and lower virulence in a ferret diarrhea model7 compared to WT C. jejuni 81-176. In addition, WT C. jejuni 81-176 induced less IL- 17 expression in small intestine lamina propria lymphocytes and reduced activation of Toll-like receptor (TLR) 2 and TLR4 in HEK cells than the mutant strain,79 suggesting the capsule is also important in mediating the host immune response to C. jejuni. Finally, both flagellin and “lipopolysaccharide (LPS)” (discussed below) were found to be important in adherence: LPS was able to adhere to both intestinal mucus gel and cultured INT407 epithelial cells, and flagellin could also bind epithelial cells.81 Cytolethal Distending Toxin. C. jejuni and other Campylobacter spp. produce CDT. When HeLa cells are treated with CDT extract from C. jejuni 81-176, cells exhibit distension, rapid and irreversible cell cycle arrest in the G2 phase, and cell death.120 CDT is also important for colonization and development of gastritis in mice,33 stimulation of production of pro-inflammatory IL-8, an important chemotactic factor for neutrophils and antigen presenting cells such as macrophages and dendritic cells, from INT407 cells,46 and mediation of apoptotic killing in cultured human monocytic cells through induction of caspases.45 Lipooligosaccharide. High molecular weight LPS in the outer membrane of Gram-negative bacteria is a virulence factor. LPS comprises the O-polysaccharide chain linked to a core oligosaccharide anchored in the bacterial outer membrane by lipid A.36 Low molecular weight (LMW) LPS, lacking the O- polysaccharide chain, is referred to as LOS. Karlyshev et al (2000)58 identified genes in the NCTC 11168 genome highly similar in sequence and organization to genes in E. coli and other Gram-negative bacteria involved in capsular polysaccharide biosynthesis and transport, and determined that what was previously thought to be HMW LPS in C. jejuni is actually capsular polysaccharide. It had been previously thought that all C. jejuni strains produce LOS, with some also producing HMW LPS.58; 124 The current 6 thought is that C. jejuni produces both LOS (LMW) and capsular polysaccharide (HMW).36 While the C. jejuni LOS is important in ganglioside mimicry and likely also the development of Guillain-Barré syndrome (discussed below), LOS also impacts C. jejuni invasiveness into host cells. C. jejuni strains bearing sialylated LOS (class A, B, or C) displayed higher invasion into Caco-2 cells than strains with non- sialylated LOS (classes D, E).73 Interestingly, the LOS outer core of C. jejuni 81-176 was shown to convert between GM2 and GM3 ganglioside mimics due to variation in the cgtA gene.39 Furthermore, site- specific cgtA mutation resulted in an LOS with a GM3 but not GM2 mimic, and enhanced invasion into INT407 epithelial cells.39 EXTRA-INTESTINAL COMPLICATIONS AND POST-INFECTIOUS SEQUELAE OF CAMPYLOBACTER JEJUNI INFECTION: EMPHASIS ON GUILLAIN-BARRÉ SYNDROME Extraintestinal Complications and Post-Infectious Sequelae of Campylobacteriosis. In addition to enteric symptoms, extraintestinal complications including pancreatitis, hepatitis, bacteremia, nephritis, and miscarriage also occur.16; 121 Infection with C. jejuni is also associated with numerous post- infectious complications. Irritable bowel syndrome (IBS), reactive arthritis and Reiter’s syndrome, and Guillain-Barré syndrome (GBS) are associated with Campylobacter infection.16; 121 The authors of a recent (2014) systematic literature review and meta-analysis estimated the proportion of Campylobacter infections associated with chronic sequelae: IBS (4.01%), reactive arthritis (2.86%), and GBS (0.07%).59 Guillain-Barré Syndrome. Following the near eradication of poliomyelitis, GBS is the most common cause of acute neuromuscular flaccid paralysis worldwide.125 The authors of a recent systematic literature review estimated the overall incidence of GBS as between 1.1/100,000/year and 1.8/100,000/year.80 This review also identified increased incidence with age, but was not able to confirm bimodality in incidence by age.80 Up to 70% of GBS cases were associated with antecedent infection, mainly upper respiratory and gastrointestinal infections.80 Another recent systematic literature review with meta-analysis also identified an increase in incidence with age, and reported an increased risk of 7 GBS in males than females, an unusual finding compared to many other autoimmune diseases.103 Reported mortality rates vary from 3-7%.111 Subtypes of GBS. GBS is recognized as a heterogeneous post-infectious disorder, with variable antecedent infection, clinical course, severity, and outcome. Several subtypes of GBS are recognized; the most widely recognized forms include acute inflammatory demyelinating polyneuropathy (AIDP), acute motor axonal neuropathy (AMAN), acute motor sensory axonal neuropathy (AMSAN), and Miller Fisher syndrome (MFS). Two features required for a diagnosis of GBS include progressive weakening of legs and arms, and areflexia or decreased tendon reflexes in the weakened limbs.111 Beyond these two features, additional clinical symptoms and nerve conduction studies can be used to distinguish between subtypes.111 Clinical Manifestations of GBS. Main clinical features of AIDP include sensorimotor deficits, which may be accompanied by cranial nerve and autonomic dysfunction, and pain.111 Physical examination findings include flaccid paralysis, reflex deficits, and variable sensory loss.85 Nerve conduction studies reveal demyelinating polyneuropathy.111 Histological findings include macrophage- mediated segmental demyelination and involvement of lymphocytes; T cells are presumed to be most important, with macrophage invasion occurring later.85; 125 The axonal forms, AMAN and AMSAN, exhibit some clinical overlap and AMAN often precedes AMSAN.85 Patients with AMAN present with paresis or paralysis, but with intact sensory function,85 and pain is sometimes reported.111 There is selective loss of motor fibers with preservation of sensory fibers and normal sensory action potential.85; 111 Histological changes include lesions at the nodes of Ranvier, including lengthening of the nodal gap corresponding to binding of IgG and complement activation, and entrance of macrophages into the periaxonal internode space.85 In contrast to AIDP, lymphocytes are typically rare or absent and demyelination is not a feature.85; 125 AMSAN resembles and can be considered a severe form of AMAN, with both sensory and motor deficits, abnormal sensory action 8 potential, and Wallerian-like degeneration in both sensory and motor fibers.85; 111 Clinically, AMSAN patients develop widespread paralysis with a protracted recovery.85 Finally, MFS is characterized clinically by the typical triad of ataxia, ophthalmoplegia, and areflexia. Although clinical outcome of MFS is generally good111 and weakness is not a prominent feature,85 “MFS-GBS overlap syndrome” can occur, characterized clinically by additional limb weakness and respiratory insufficiency.111 GBS Epidemiology. Interestingly, frequency of the clinical subtypes varies by geographic region. AIDP predominates in Europe and North America, accounting for between 60-90% of GBS cases.85; 111; 125 Conversely, the axonal forms are common in China, Japan, Mexico, and Bangladesh,85; 125 accounting for 30-65% of GBS cases in Asia and Central and South America.111 The basis of geographical diversity is not known, but may involve varying exposure to certain types of infections, or genetic differences between populations.111 Antecedent Infections. Antecedent infections are most commonly reported less than four weeks prior to the onset of GBS and are frequently grouped by “upper respiratory” or “gastrointestinal” manifestations.80 Several infectious agents have been associated with development of GBS, including viral (cytomegalovirus, and Epstein-Barr, varicella-zoster, and influenza A viruses) and bacterial (Haemophilus influenzae, Mycoplasma pneumoniae, C. jejuni) agents.54; 111; 125 A case-control study involving 154 GBS patients determined that C. jejuni was reported most frequently as the recent infection (32%), followed in frequency by cytomegalovirus, Epstein-Barr virus, and M. pneumoniae.54 The presence of anti -GM1 and -GD1b antibodies was also associated with C. jejuni infection.54 One large multicenter study found that C. jejuni was the most common cause of antecedent infection, followed by cytomegalovirus and Epstein-Barr virus.40 Additionally, these investigators found that patients with preceding C. jejuni infection were more likely to have axonal neuropathy, pure motor GBS, and anti-GM1 antibodies.40 Among other factors, C. jejuni infection, diarrhea, older age, severe weakness, and 9 abnormal motor action potentials were all associated with poor outcomes.40 In a case-control study, 27/103 GBS or MFS patients (26%) were C. jejuni positive.99 A recent systematic literature review aiming to quantify the association between Campylobacter and GBS determined that 31% of 2,502 GBS patients described in case-control studies were attributable to Campylobacter infection.94 Pathogenesis of GBS: Molecular Mimicry and Anti-Ganglioside Antibodies. The pathogenesis of GBS following C. jejuni infection is incompletely understood. Molecular mimicry-induced nerve damage initiated by cross-reactive antibodies to LOS in the C. jejuni outer membrane and structurally similar peripheral nerve gangliosides, including GM1 and GD1a, is accepted as a likely mechanism.111 Gangliosides are sialic-acid containing glycosphingolipids organized in microdomains in cell membranes, playing a role in cell growth, differentiation, and signaling.55 Gangliosides are found on most nucleated cells throughout the body but are enriched in the nervous system. The term “molecular mimicry” is used to indicate “sharing of antigens between hosts and microorganisms” and also broadly describes cross- reactivity in an immunological context.4 Ang, Jacobs, and Laman define molecular mimicry as “dual recognition of structures of a microbe and host by a single B- or T-cell receptor” and “the mechanism by which infections trigger cross-reactive antibodies or T cells that cause symptoms of autoimmune disease”.4 GBS patients often have serum antibodies to gangliosides including GD1a, GM1a, GM1b, GQ1b, GT1a, or ganglioside complexes.95; 111 In 1993, Yuki and colleagues extracted LPS from C. jejuni strain 90-26 isolated from a GBS patient with anti-GM1 serum antibodies. Using gas-liquid chromatography mass-spectrometry and protein NMR spectroscopy, they showed that the LPS contained sugar components of GM1, the oligosaccharide structure protrudes from the LPS core, potentially exposing the GM1-oligosaccharide moiety, and that the terminal structure was identical to the GM1 terminal tetrasaccharide.127 This study provided the first conclusive evidence of structural similarity between a ganglioside and C. jejuni LOS, and others have followed.37; 39 10 Several studies have reported the prevalence and type of anti-ganglioside antibodies in GBS and identified associations between these autoantibodies and preceding infection or clinical outcome. In a multicenter study of 229 GBS patients, presence of anti-GM1 IgA or IgG, but not IgM, was associated with a worse outcome.40 Additionally, patients with preceding C. jejuni infection had higher frequencies of anti-GM1 IgA, IgG, and IgM, and worse outcomes than the study population as a whole.40 A prospective case-control study of 96 GBS patients determined that of the 25 C. jejuni positive patients, 52% had anti-GM1 antibodies, while 15% of C. jejuni negative GBS patients had these antibodies.99 C. jejuni positive patients with anti-GM1 antibodies were also found to have slower recoveries and greater disability than C. jejuni positive patients without anti-GM1 antibodies.99 Subtype of anti-ganglioside IgG also has been associated with type of antecedent infection and clinical outcome of GBS. Koga et al (2003) found that, in 42 GBS patients with anti-GM1 antibodies, the IgG1 subtype was most common and was associated with previous gastroenteritis, positive C. jejuni serology, and a slower clinical recovery reflected by return to independent locomotion.62 In contrast, anti-GM1 IgG3 was associated with antecedent respiratory infection and more rapid clinical recovery.62 In a study of 176 GBS patients, Jacobs and colleagues (2008) corroborated these findings. They determined that presence of only IgG1 subtype antibodies to motor gangliosides, without IgG3, was associated with diarrhea, positive C. jejuni serology and LOS cross-reactivity, and worse neurological outcomes; in contrast, anti-ganglioside antibodies of both IgG1 and IgG3 types were associated with previous upper respiratory infection, cross-reactivity with H. influenzae LOS, and better neurological outcomes.55 Role of Host Factors in GBS. Some C. jejuni strains isolated from patients with uncomplicated enteritis also expressed ganglioside mimics, although less frequently than strains isolated from GBS or MFS patients.5 Furthermore, although ganglioside mimics were present in C. jejuni strains from some enteritis patients, antibody reactivity of enteritis patients to neural glycolipids was significantly lower 11 than in GBS/MFS patients, supporting a role for host-related factors in addition to C. jejuni ganglioside mimicry in development of neurological disease.5 Expression of ganglioside mimics in C. jejuni strains isolated from both GBS and uncomplicated enteritis patients has been shown previously.105 Moreover, anti-GM1 antibodies derived from the serum of a GBS patient did not react with autologous C. jejuni, suggesting a change in expression of the ganglioside mimic or involvement of a mechanism additional to molecular mimicry; binding of anti-GM1 antibodies by LPS in C. jejuni strains from both GBS and uncomplicated enteritis patients was also observed.52 Associations between several human genetic factors, including MMP9, TNFα, Fc gamma receptor III, class I or II HLA types, TLR4, IL-10 and GBS incidence, severity, and outcome have been studied with variable results.51; 55; 56 The geographic distribution of various subtypes and the rarity of GBS outbreaks also support a role of variable host susceptibility. Animal Models of GBS. The immunopathogenesis of GBS is complex, and a robust, reproducible animal model mimicking human GBS would greatly further our understanding. As our understanding of GBS evolves, there is increasing support for contributions of both characteristics of C. jejuni, such as the presence of ganglioside mimics in the LOS, and host susceptibility factors. Exploration of several animal models focusing on pathogenesis and histopathological findings have been attempted. Two published reports suggest chickens could be useful models for GBS.70; 87 In both of these studies, chickens between 4 weeks and 6 months of age were orally infected with C. jejuni strains isolated from GBS patients. One study also included a spontaneous disease group, in which chickens from flocks owned by families having a member with GBS exhibited weakness and developed paralysis.70 Two of the five chickens with spontaneous neurological disease had histological lesions resembling AMAN. In this same study, some chickens experimentally fed C. jejuni isolated from a patient with AMAN developed diarrhea 2—4 days later and/or paralytic neuropathy after approximately 12 days.70 In the second study, 73% of infected chickens developed diarrhea after 3—7 days, and 17/22 of these 12 chickens developed paralytic neuropathy resembling GBS after 5—14 days.87 Sciatic nerve pathology in diseased chickens included Wallerian-like degeneration, nodal lengthening, paranodal demyelination, and infiltration of lymphocytes and macrophages.70; 87 Rabbits have also been investigated as models for C. jejuni-induced GBS. Rabbits injected repeatedly for months with a mixture containing LOS isolated from GBS patient-derived C. jejuni strain CF 90-26 bearing a GM1 mimic, along with keyhole limpet hemocyanin and Freund’s complete adjuvant (CFA) developed neurological signs sooner than control rabbits.126 Neurological lesions in sciatic nerves of paralyzed rabbits included Wallerian-like degeneration and macrophages within periaxonal spaces.126 Several mouse models have been produced that reflect different forms of GBS or various aspects of pathogenesis. These include both spontaneous and experimentally induced models resembling the AIDP form of GBS. Nonobese diabetic (NOD) mice lacking the costimulatory molecule B7- 2 (CD86) do not develop diabetes, but instead exhibit spontaneous autoimmune peripheral polyneuropathy (SAPP) arising at 20 weeks of age in a proportion of mice.100 SAPP resembles chronic inflammatory demyelinating polyneuropathy (CIDP) in humans with symmetrical limb paralysis. Histopathologic features also resemble the AIDP form of GBS, including lesions with T cell infiltration into dorsal root ganglia (DRG) and peripheral nerves.100 Another model is experimental autoimmune neuritis (EAN) that has been induced in Lewis rats and mice. SJL/J mice injected with bovine-derived peripheral nerve myelin with CFA develop a constellation of signs and lesions termed severe murine EAN, including neuromuscular weakness, nerve electrophysiology abnormalities, and demyelination and infiltration of macrophages and lymphocytes in sciatic nerves.122 One BALB/c mouse model of MFS involves induction of respiratory paralysis through intraperitoneal injection of antibody reacting against multiple gangliosides along with normal human serum as a complement source.41 Anti-GQ1b antibody binding led to complement activation and subsequent transmission block at diaphragmatic neuromuscular junctions, resulting in respiratory 13 paralysis. In support of a role of complement in MFS pathogenesis, respiratory abnormalities were abrogated by treatment with eculizumab, a human monoclonal antibody that prevents production of terminal complement components by blocking the intermediary C5 component.41 This model elucidates some aspects of MFS pathogenesis including molecular mimicry, but it is not a complete model of post- infectious GBS. There are drawbacks to each of these animal models. The chicken models produce similar symptoms of diarrhea and GBS as in people, but the fundamental anatomic and physiological differences between humans and birds limit the utility of avian models. Rabbits and mice may prove to be more relevant models of human disease. While the models described above provide insight into various aspects of GBS, none of these models are induced following oral infection of C. jejuni, as in people; with the exception of the SAPP model, disease is induced through subcutaneous or intraperitoneal injection of LOS, anti-ganglioside antibody, or myelin and requires additional factors such as CFA or human serum for disease induction. Mouse Models of GBS Following Oral C. jejuni Infection. Promising progress in developing mouse models resembling GBS through oral infection with C. jejuni has been made recently. Over several experiments, WT, IL-10-/-, and CD86-/- NOD mice orally infected with C. jejuni strains associated with GBS (strains HB93-13, 260.94) or MFS (CF93-6) developed combinations of neurological abnormalities determined by phenotyping tests including DigiGait treadmill analysis and open field testing, anti- ganglioside antibodies, and nerve lesions characterized by infiltration of macrophages or T-cells into sciatic nerves and dorsal root ganglia.108 Interestingly, treatment with the antibiotic chloramphenicol modulated the immune response in NOD IL-10-/- mice, reflected by significant decreases or increases in anti-C. jejuni or anti-ganglioside antibodies relative to C. jejuni- or sham-inoculated groups not receiving chloramphenicol.108 This suggests that antibiotic-induced changes in the microbiome may be important in the immune response to C. jejuni infection. The importance of the microbiome in immunity to C. jejuni 14 is supported by the finding of increased anti -C. jejuni and -GM1 antibodies in mice with a humanized compared to a conventional mouse microbiota.21 These two studies provide promising murine models of GBS development following oral C. jejuni infection. All of these described animal models have drawbacks, including methods of induction that do not resemble post-infectious GBS and induction of only mild lesions. Both C. jejuni factors, such as the presence of ganglioside mimics in the LOS, and host factors, such as propensity for developing anti- ganglioside antibodies, likely contribute to susceptibility of an individual to development of GBS following C. jejuni infection. Therefore, a mouse model exploring both C. jejuni factors, such as invasiveness and pathogenicity in previous mouse models, and host factors, including reported immunological biases toward T helper (Th) 1- or Th2-mediated responses in C57BL/6 and BALB/c mice respectively, would be valuable to determine factors contributing to GBS susceptibility following oral C. jejuni infection. Understanding of disease outcome in C. jejuni infection requires understanding of the initial host-microbe interaction and development of both innate and adaptive immunity. HOST RESPONSE IN CAMPYLOBACTERIOSIS: INITIAL HOST-MICROBE INTERACTION AND INNATE IMMUNITY The initial host-C. jejuni interaction influences both non-specific innate immunity and subsequent adaptive immunity.75 Therefore, especially in infections in which the immune response is not fully understood, knowledge of the initial interaction between the pathogen and innate barriers such as mucus in the gastrointestinal tract, and subsequent interaction with the epithelium and resident immune cells, is crucial. Studies have shown that C. jejuni can be found inside non-phagocytic gut epithelial cells and also within phagocytic immune cells such as macrophages and dendritic cells. Contact with Mucus Layer. Following ingestion, C. jejuni is chemotactically attracted to mucins and glycoproteins present in the mucus layer, an important mechanism facilitating colonization.19 The flagella and spiral morphology of C. jejuni contribute to retained motility in viscous environments.32 15 Furthermore, both C. jejuni flagella and LPS act as adhesins to INT 407 epithelial cells, while the LPS also bound to intestinal mucus derived from rabbits.81 The ability of C. jejuni to retain motility in and adhere to mucus likely facilitates its attachment to epithelial cells and eventual colonization. Immunohistochemical and ultrastructural studies in infected humans showed that C. jejuni can be found not only in the glycocalyx covering surface epithelium, but also within surface epithelial cells, including goblet cells, and within histiocytes in the lamina propria.112 Interestingly, these findings did not correlate with severity of inflammation.112 Interaction with Host Epithelium. Expression of the outer membrane protein CadF allows C. jejuni to bind to fibronectin. This is an interaction of particular importance prior to internalization into epithelial cells, as fibronectin is abundant in the basement membrane and at points of cell contact in the epithelium.63 Further work with polarized cells, exhibiting apical and basolateral surfaces and tight intercellular junctions, of the T84 line resembling those in colonic crypts, corroborated the importance of the binding of C. jejuni to fibronectin.83 This study showed that C. jejuni translocation occurs paracellularly, rather than intracellularly, and that invasion occurs primarily through interaction with fibronectin at the basolateral surface.83 While there is consensus that C. jejuni can transmigrate across epithelium, whether the primary mechanism is transcellular, paracellular, or both, is still debated.6 In vitro models utilizing cultured INT407 epithelial cells and enteric C. jejuni strain 81-176 have helped elucidate various host cell events in invasion.50; 88 In contrast to the microfilament-dependent invasion of other enteric pathogens including E. coli and Salmonella, invasion of C. jejuni 81-176 into epithelial cells instead depends primarily upon microtubules.88 Production of bacterial, but not eukaryotic, proteins was necessary for C. jejuni invasion.88 Hu et al (2006) proposed a model in which C. jejuni interaction with G-protein-coupled receptors within caveolae, small pits in the plasma membrane of many different cell types, are an early event in host cell signal transduction during invasion.50 This 16 interaction leads to activation of host cell PI 3-kinase and MAP kinases, shown through inhibition studies to be important in C. jejuni invasion.50 The fate of C. jejuni within intestinal epithelial cells and innate immune cells such as macrophages may be different. Watson and Galán (2008) showed that C. jejuni 81-176 could survive within human intestinal epithelial T84 cells for at least 24 hours, although C. jejuni was efficiently killed by macrophages derived from mouse bone marrow.117 This difference in survival between different cell types was mediated through different intracellular trafficking: in epithelial cells, the internalized C. jejuni resided in a “C. jejuni-containing vacuole” (CCV) which avoided delivery to lysosomes through the endocytic pathway, while in macrophages, C. jejuni was delivered to lysosomes and killed.117 In human and mouse macrophages, C. jejuni activates the NLRP3 inflammasome, cytosolic multiprotein complexes that induce activation of caspase-1, IL-1β, and IL-18 following bacterial invasion.20 Interestingly, cell death by pyroptosis typically follows inflammasome activation by other bacterial pathogens, but C. jejuni induces activation of the inflammasome without cytotoxicity.20 Correlating Invasion and Pathogenicity. Several studies have used models of C. jejuni association/adherence, invasion, and intracellular survival in cultured cells to assess correlations between these features and pathogenicity. Law et al (2009) examined 59 C. jejuni strains isolated from poultry and chose 5 with variable ability to invade HEp-2 or survive in murine macrophages (J cells). When piglets were inoculated with these strains, little correlation was found between the in vitro findings and development of lesions in vivo.68 Everest et al (1992) showed that a higher proportion of colitis-associated C. jejuni strains were able to transcytose polarized Caco-2 cells and invade HeLa and Caco-2 cells than non-inflammatory strains.30 These findings were difficult to interpret, as C. jejuni association with HeLa cells varied with time. Additionally, even though more colitis-associated strains were able to invade, similar numbers of viable intracellular C. jejuni from non-inflammatory strains capable of invading were seen relative to colitis strains.30 However, Fauchere et al (1986) observed that 17 ability of C. jejuni or C. coli strains to associate with HeLa cells correlated with clinical symptoms of diarrhea and fever (although not blood in the feces), and concluded that in vitro association assays could be used to predict pathogenicity.31 Similarly, Konkel and Joens (1989) showed that although C. jejuni strains varied considerably in ability to invade HEp-2 cells, isolates from patients with clinical campylobacteriosis were generally more invasive than non-clinical isolates.64 Interestingly, a C. jejuni mutant strain deficient in the formic acid receptor Tlp7 demonstrated markedly reduced invasion into Caco-2 cells. This mutant was able to colonize mice as well as the parental strain, but did not induce immunopathology as seen by the parental strain, indicating that invasiveness and not just colonization contributed to the host immune response.12 Thus, in vitro invasion may reflect pathogenicity in some strains, although technical factors such as type of cultured cells, assay used, number of laboratory passages of C. jejuni, and other experimental conditions should be considered. Additionally, in vitro assays by nature do not consider additional host factors such as microbiome and immune response, which are integral contributors to the host-microbe interaction and consequences of infection. Interaction with innate immune cells in the lamina propria necessitates passage of C. jejuni through the epithelium. C. jejuni was shown to traverse polarized T84 human intestinal epithelial cells through the paracellular, rather than intracellular route.83 C. jejuni also produces CDT, which stimulates chemotactic IL-8 production in INT 407 cells46 but also induces cell cycle arrest.120 These studies suggest that C. jejuni could interact with innate cells present in the intestinal subepithelium either after crossing the epithelium between cells, or following liberation from dying epithelial cells following invasion and CDT production. Upon entrance to the subepithelial lamina propria, C. jejuni subsequently interacts with resident innate immune cells, such as neutrophils and antigen presenting cells including macrophages and dendritic cells. Dendritic Cells. In 1973, Steinman and Cohn described a novel type of cell, distinct from lymphocytes, granulocytes, and mononuclear phagocytes, isolated from murine spleen and peripheral 18 lymphoid tissues.109 This cell type comprised approximately 1.0-1.6% of the total splenic nucleated cell population, exhibited numerous branching forms, and was observed to extend and retract cellular processes, leading the investigators to apply the name “dendritic cell.”109 Subsequent in vitro studies further distinguished dendritic cells (DC) from macrophages, including by characterization of the relatively poorer endocytic capacity of DC compared to macrophages using a variety of test substances.110 Current knowledge indicates that DC originate from hematopoietic precursors and comprise diverse subtypes with heterogeneous intermediate precursors, location, form, function, and expression of cell-surface antigen.71; 106 Understanding of these different DC types is a new and evolving area of research. DC development during hematopoiesis is complex, with production of both DC and macrophages possible through both myeloid and lymphoid pathways;106 however, myeloid- or lymphoid-origin distinction is not typically relevant even when referring to DC in lymphoid tissues, as even DC in the spleen and thymus are mostly myeloid-derived.71 Broad categories of DC presented by Shortman include: pre-dendritic cells, such as steady state plasmacytoid DC or monocytes, capable of stimulus-induced development into DC; and conventional DC (cDC), including migratory DC and lymphoid-tissue-resident DC.106 Migratory cDC sample the periphery, migrate to the lymph node bearing antigen, and interact with T cells; lymphoid-tissue-resident DC include cDC restricted to the thymus or spleen, where they sample and present antigen locally.106 Inflammatory DC are an additional category, not usually present in steady state without stimulus, but can develop from cells such as inflammatory monocytes.106 A summary of the current understanding follows: cDC are important sentinel immune cells, positioned throughout the periphery in places such as the lamina propria of the gut to sample the local environment for potential antigens. They are extremely efficient antigen presenting cells (APC) and connect the innate and adaptive immune systems following the initial host-pathogen interaction. In the 19 gut, cDC surveil the environment for antigens and in the steady state, and are considered tolerogenic of innocuous antigens, maintaining symbiosis with the microbiota.57 Uptake of antigens by DC can occur by pinocytosis, phagocytosis, or receptor-mediated endocytosis.113 Once a differentiated but immunologically immature DC acquires an antigen, it migrates to lymph nodes and during migration, upregulates molecules important in stimulation of T cells: costimulatory molecules, cytokines, and MHC complexes.2 As DC are considered the most important APC for the priming of naïve T cells and subsequent direction of Th differentiation into Th1, Th2, Th17, and Treg subsets,113 DC are crucial components of the initiation of an effective and pathogen-appropriate adaptive immune response. In a landmark 1986 study, Mosmann and Coffman described two distinct groups of Th cells based upon surface antigen expression, B cell stimulating activity, and cytokine production, designated as Th1 and Th2 subsets.84 It is now known that Th1 cells are characterized by production of IFN-γ and control of intracellular pathogens, while Th2 cells produce IL-4, IL-5, IL-10, and IL-13 and are important in parasite control and allergic responses.96 Additional Th subsets described to date include Treg and Th17, among others.96 A subset of “suppressive” T cells mediating tolerance was originally described in 1970 by Gershon and Kondo35 and since this discovery, immune suppression by Treg cells is known to be mediated in part by the anti-inflammatory actions of TGF-β and IL-10.96 Th17 cells were relatively recently (2005) designated a distinct subset of CD4+ T cells based on characteristics of differentiation and the production of IL-17.91 Cells producing IL-17, the signature cytokine of Th17 cells, are now understood to have a critical role in defense against extracellular bacterial and other types of pathogens at mucosal surfaces.60; 96; 118 While differentiation of Th17 cells is closely related to that of Treg cells,14 the production of IL-17 leads to inflammation propagated by neutrophil recruitment, enhancement of B cell function, and additional release of pro-inflammatory cytokines.96 Several studies with C. jejuni have utilized in vitro assays, including the use of cultured monocytes, macrophages or dendritic cells from humans or mice derived by different methods. Critical 20 events in the host-microbe interaction have been evaluated, including initial internalization of C. jejuni, intracellular survival or killing, and stimulation of cytokine production and T cell priming. C. jejuni Interaction with Monocytes/Macrophages. Evaluation of internalization and intracellular survival of C. jejuni within monocytes and macrophages are relevant for understanding the response to C. jejuni during bacteremia or following invasion into the intestinal subepithelium. Studies have shown that C. jejuni is capable of both survival and replication in human or mouse monocytes or macrophages. Kiehlbach et al showed that C. jejuni 2964 from a diarrheic patient continued to be taken up into cultured J cells (BALB/c macrophage cell line), resident peritoneal macrophages from BALB/c mice, and human peripheral blood monocytes over the 8-hour infection period; C. jejuni then remained viable for 6-7 days intracellularly.61 These observations were subsequently confirmed, as C. jejuni 81-176 remained viable inside human monocytic cells (28SC line) for up to 7 days.45 CDT-mediated apoptosis was observed, and live C. jejuni could be visualized inside vacuoles within dead monocytes.45 Interestingly, phagocytes from some individuals may exhibit relatively decreased ability to kill phagocytosed C. jejuni compared to other individuals. Macrophages derived from peripheral monocytes in human blood by cytokine stimulation were used to assess uptake and intracellular survival of C. jejuni by gentamycin protection assay. The investigators showed that the 16 C. jejuni strains tested were killed by macrophages from most donors within 48 hours, but macrophages from approximately 10% of donors were incapable of killing C. jejuni despite adequate uptake, instead allowing intracellular replication of C. jejuni.115 C. jejuni Interaction with Dendritic Cells. Despite the importance of C. jejuni as an enteric pathogen and the critical role of dendritic cells in bridging innate and adaptive immunity, relatively few studies in recent years have specifically evaluated C. jejuni-DC interaction. Both human DC derived from blood monocytes8; 29; 48 or mouse DC derived from bone marrow97; 98 have been used to study C. jejuni invasion, DC activation, and subsequent Th polarization. C. jejuni 81-176 was found to invade human DC 21 in a dose-dependent manner in up to 4 hours of infection time, although marked reduction in viable intracellular C. jejuni with limited DC cytotoxicity were seen by 24 hours.48 C. jejuni also induced DC maturation (reflected by increased costimulatory molecule expression), activated NF-κB, and stimulated production of cytokines including IL-1β, IL-6, IL-8, IL-10, IL-12, IFN-γ, and TNF-α.48 Similarly, human monocyte-derived DC infected with C. jejuni 11168 produced IL-12, IL-23, IL-1β, and IL-6, and stimulated production of IL-17A and IFN-γ in T cells.29 Together, these studies support C. jejuni-induced cytokine production and Th1/Th17 polarization by infected DC. Sialylation of the C. jejuni LOS enhances invasion of epithelial cells in vitro,73 and has recently been shown to impact DC function as well. Interestingly, human DC-mediated T cell polarization can vary depending upon the infecting C. jejuni strain, and specifically by the sialylation of the LOS: α2-3-linked sialylated LOS drove a predominant Th2 response, while α2-8-linked sialylation resulted in Th1 differentiation.8 Expression of siglecs (sialic acid-binding Ig-like lectins) on the DC was implicated in this response, as different siglecs preferentially bind glycans with specific sialic acid linkage.8 C. jejuni strains with sialylated LOS also stimulated enhanced DC activation and release of soluble factors resulting in proliferation of mucosal B cells, compared to strains with non-sialylated LOS.66 As with in vitro studies with human DC, murine bone marrow-derived DC (BMDC) internalize C. jejuni, stimulating DC maturation, cytokine production, and ability to differentiate T cells in co-culture systems. BMDC derived from C57BL/6 WT mice infected with C. jejuni 11168 showed efficient internalization and complete killing, with no viable intracellular C. jejuni enumerated 8 hours post infection (p.i.).98 Furthermore, similar to human DC, C. jejuni-infected DC underwent maturation, reflected by increased MHC II and costimulatory molecule expression. Cytokine production and T cell polarization also mirrored that found in human DC studies: infected DC produced pro-inflammatory cytokines and induced Th1 polarization.98 Further studies showed that TLR2, TLR4, and subsequent signaling through adaptor molecules MyD88 and TRIF are involved in events including DC maturation, 22 production of IL-12, and maximum Th1 polarization reflected by IFN-γ production induced by C. jejuni infection.97 Stimulation of Pattern Recognition Receptors and Downstream Events. Binding of pathogen- associated molecular patterns (PAMPs) by cell surface or intracellular pattern recognition receptors (PRRs) of host innate cells is a critical initial event in the host-pathogen interaction. TLRs are expressed by epithelial cells and immune cells including macrophages and DC, and can be located either inside or on the surface of the cell.1 Cytoplasmic PRRs have also been identified, including nucleotide-binding oligomerization domain (NOD) proteins. Recognition of specific PAMPs by PRRs leads to initiation of the host defense by activation of complex cellular signaling pathways and stimulation of cytokine production.1 C. jejuni has been shown to interact with host cells through stimulation of TLR2 and TLR4.34; 97 Interestingly, the flagella on C. jejuni are able to avoid recognition by TLR5 on host cells, while retaining motility important in pathogenesis through colonization of mucus.3 C. jejuni also does not efficiently stimulate TLR9, correlating with relatively low frequency of [CG] dinucleotides, compared to other bacteria.26 The cytoplasmic NOD1 PRR in human intestinal epithelial (Caco-2) cells also recognizes C. jejuni, while NOD2 is apparently not as important in this system128 or in a secondary abiotic C57BL/6 IL-10-/- mouse infection model.43 Downstream events following initial interaction of C. jejuni with host cells, including epithelial or antigen presenting cells, includes production of chemotactic and pro-inflammatory cytokines and subsequent initiation of the adaptive immune response. C. jejuni elicits secretion of chemoattractant IL- 8 from cultured intestinal epithelial cells (INT407), and amount of IL-8 released was correlated with C. jejuni strain invasiveness.44 Hu and Hickey (2005) subsequently showed that INT407 cells also produced pro-inflammatory chemokines, including macrophage inflammatory protein 1 and monocyte chemoattractant protein 1 (MCP-1), at least partially mediated by NF-κB, following exposure to C. jejuni.49 Cultured polarized Caco-2 epithelial cells produce IL-6, a mediator of the acute phase 23 inflammatory response, following exposure to 8 different C. jejuni strains.34 Induction of cytokines from C. jejuni-infected human gut explant tissue included IFN-γ, IL-12, IL-23, and IL-6.29 In addition to chemoattractant and pro-inflammatory cytokine production by infected epithelial cells, both human and murine DC have been shown to produce pro-inflammatory and T-cell polarizing cytokines following C. jejuni infection. Infected human monocyte-derived DC produced increased IL-1β, IL-6, IL-8, IL-10, TNF-α, IFN-γ, and IL-12.48 C. jejuni-infected murine BMDC also produced a similar pro- inflammatory profile, including increased TNF-α, IL-6, and IL-12p70.98 Furthermore, the Th1 polarization predicted by Hu in response to increased IL-12 production was confirmed by this murine study, in which naïve Th cells co-cultured with infected BMDC produced IFN-γ, but not IL-4 or IL-10, indicating Th1 polarization.98 Collectively, these studies show that C. jejuni is proficient in navigating the gastrointestinal mucus layer, and subsequent colonization and invasion of the gut leads to innate immune responses by both epithelial and antigen presenting cells. Production of chemoattractant IL-8 and pro-inflammatory and T-cell polarizing cytokines leads to initiation of the adaptive immune response. Both C. jejuni strain characteristics and host immune responses can contribute to differences in adaptive immunity and variable disease outcomes. ADAPTIVE IMMUNITY TO CAMPYLOBACTER JEJUNI Antibody responses to C. jejuni infection in people have been studied in natural and experimental infection. Sera from people suffering from Campylobacter enteritis showed increased specific IgM, IgG, and IgA antibodies; IgM and IgG rose simultaneously and remained elevated longer than IgA, which was more transient.17 In an experimental infection study, IgA and IgM increased more than did IgG levels, and all three Ig classes peaked at 11 days before declining.15 Furthermore, the magnitude of serological response correlated with clinical illness.15 24 Ex vivo studies of infected human intestinal tissue and in vivo mouse models have helped to further characterize systemic immune responses to C. jejuni infection. Human intestinal biopsies were cultured and infected with C. jejuni in vitro; the mucosal cytokine response included increased IFN-γ, along with more modest increases in IL-23, IL-12, IL-6, and proportionally lower increases in IL-17 and IL- 1β.29 Supernatants from infected monocyte-derived DC also induced production of IFN-γ and IL-17A positive or double positive T cells.29 Collectively these data support Th1/Th17-mediated adaptive immune responses. In vivo mouse studies have helped further delineate systemic immune responses to C. jejuni infection, reflected by increases in various Th1-, Th2-, or Th17-mediated isotypes of C. jejuni-specific plasma IgG. Fox et al (2004) measured C. jejuni-specific plasma IgG2a (Th1-mediated) and IgG1 (Th2- mediated) in infected mice, and found that IgG2a was more frequently detected than IgG1 within the treatment groups and correlated with C. jejuni cdtB mutant strain clearance.33 Similar results were found in other mouse models. C57BL/6 WT and IL-10-/- mice infected with colitogenic C. jejuni 11168 produced robust Th1/Th17-mediated plasma IgG2b, but not Th2-mediated IgG1.76 Increased Th1-mediated plasma IgG2c and IgG3, but not IgG1, was observed in C57BL/6 IL-10-/- mice infected with C. jejuni 11168; interestingly, in this study, C57BL/6 IL-10-/- infected with GBS patient-derived C. jejuni 260.94 responded with elevated IgG1 as well as IgG2c.75 St. Charles et al (2017) also reported robust anti-C. jejuni IgG2b responses in NOD WT, IL-10-/-, and B7-2(CD86)-/- mice; the magnitude of the response varied by both mouse strain and infecting C. jejuni strain (GBS-associated 260.94 and HB93-13 strains).108 Considering these data and the pro-inflammatory and Th1-polarizing responses produced by infected DC,48; 98 it appears a Th1-mediated adaptive immune response is commonly elicited in C. jejuni infection. However, Th2-mediated responses can also be produced, and responses vary by both infecting C. jejuni strain and genotype of mouse used. 25 The concept that development of GBS results from an aberrant adaptive immune response to C. jejuni antigens, involving molecular mimicry between C. jejuni LOS and peripheral nerve gangliosides, is a central hypothesis for the immunopathogenesis of GBS.4; 111 It is generally thought that C. jejuni infection leads to Th1-mediated adaptive immunity, which leads clinically to inflammation and diarrhea, but also clearance.124 However, antibodies specific for the LOS of C. jejuni can be cross-reactive with peripheral nerve gangliosides, potentially leading to antibody- and complement-mediated nerve damage. The isotype of the cross-reactive antibody also has implications, with Th2-mediated IgG1 anti- ganglioside antibodies in GBS correlating with previous C. jejuni infection and worse clinical outcome.53; 62 Thus it appears that a deviation in typical immune response to C. jejuni, likely mediated by both C. jejuni and host immune factors, predisposes an individual to development of GBS following C. jejuni infection. RATIONALE FOR STUDY Overarching Aim: Examine host-microbe interaction and clinical outcome in mice of different genetic backgrounds infected with different C. jejuni strains to assess interplay of both microbe and host factors In vitro, C. jejuni strains vary considerably in virulence related to specific mechanisms in pathogenesis. Adherence to intestinal mucus substrate and adherence and invasiveness in cultured epithelial cells varies by C. jejuni strain.73; 81 Variation in LOS structure, such as sialylation, between strains can impact Th-polarization following interaction with DC8 and contributes to differences in epithelial cell invasion.73 In vivo, numerous C. jejuni strains have been used in mice of the same genetic background (C57BL/6 IL-10-/-) in models established by Mansfield and colleagues.10; 11; 75 In our models, C. jejuni strains have varied in ability to colonize and produce colitis, and also in magnitude and type of systemic immune response reflected by plasma anti-C. jejuni and anti-ganglioside antibodies. 26 In addition to C. jejuni strain differences, host genetic background is postulated to contribute to disease outcome, including subsequent development of GBS. Evidence supporting a potential role of host factors include geographic clustering of GBS subtypes,111 association of specific genetic factors with development of GBS,56 and the fact that while ganglioside mimics on the C. jejuni LOS are more frequent in GBS than uncomplicated enteritis, alone they are not sufficient to cause GBS following C. jejuni enteritis.5; 105 Choice of C. jejuni Strains. For the current study, two Campylobacter jejuni strains used previously in mouse models in our laboratory were chosen based upon production of different disease outcomes and immune responses. C. jejuni 260.94 was originally isolated from a GBS patient in South Africa. It possesses a GM1a ganglioside mimic, is LOS class A and encodes cst-II sialyl transferase.74 In our mouse models, encompassing different mouse strains, it has colonized well but is not colitogenic.10; 21; 75; 108 Malik et al (2014) showed that C. jejuni 260.94 drove a predominantly Th2-mediated immune response, including production of anti-ganglioside antibodies, in C57BL/6 IL-10-/- mice.75 In contrast, C. jejuni 11168 was originally isolated from an enteritis patient in the United Kingdom. It harbors both GM1 and GM2 mimics, but is not reportedly associated with GBS in people. It is LOS class C and encodes cst-III sialyl transferase.74 In our mouse models, C. jejuni 11168 has colonized well and produced moderate to severe colitis and Th1/Th17-associated anti-C. jejuni plasma antibodies in C57BL/6 IL-10-/- mice.10; 11; 75-77; 101 A few studies have contained experiments with less striking results: a low proportion of mice exhibited colitis on a first passage, but subsequent passages resulted in marked colitis;11 the severity of lesions was not as marked despite most mice displaying colitis in dose-response experiments;76 and a more recent study in which infected mice exhibited lower colonization levels and less gross pathology.21 Aside from these exceptions, infection of C57BL/6 IL-10-/- mice with C. jejuni 11168 has produced repeatable colitis in our models. 27 Choice of Mouse Genetic Background. While referring to Th1 or Th2 “prototypic” mouse strains is likely an oversimplification, several reports in the literature suggest that C57BL/6 mice exhibit a Th1 bias, while BALB/c mice tend to produce Th2 responses. Differences have also been seen in relative protection or susceptibility to different pathogens or disease between these mice, or higher or lower propensities for inflammatory responses. Differences in chemokine, chemokine receptor, and cytokine mRNA expression in spleen of immunologically naïve mice have been found. C57BL/6 mice expressed higher Th1-associated IFN-γ and IP-10, while BALB/c mice expressed more regulatory and Th2- associated LTβ, IFNβ, TGFβ1, CCR3, and CXCR4.24 When stimulated with LPS or synthetic TLR2 ligand in vitro, C57BL/6 peritoneal macrophages produced more TNF-α and IL-12, and had higher bacteriocidal activity than did those from BALB/c mice.116 Similarly, when spleen cells were stimulated with concanavalin A, C57BL/6 cells produced higher levels of IFN-γ and less IL-4, while the opposite was true in BALB/c cells.82 Studies of splenic-derived DC showed that, upon stimulation with TLR ligands LPS, lipoprotein, and CpG, DC from C57BL/6 mice produced higher IL-12 while BALB/c DC produced more MCP-1.72 Mucosal immunity also varies between these two mouse strains. Compared to C57BL/6 mice, BALB/c mice exhibited enhanced vitamin A metabolism, leading to greater IgA production.38 BALB/c mice also had higher percentages of Treg cells in the small intestine and were more resistant to colitis induced by dextran sulphate sodium (DSS).38 Another study demonstrated that resistance to DSS- induced colitis by BALB/c mice was associated with a Th2/Th17/Treg-mediated response, characterized by increased production of IL-4, IL-6, IL-10, and IL-17 and more Treg cells in the regional lymph node, while susceptible C57BL/6 mice produced more TNF-α and IFN-γ.123 Finally, Chen et al (2005) showed that while Treg cells were phenotypically similar between C57BL/6 and BALB/c mice, the proportion of Treg cells was higher in the spleen, lymph node, and thymus of BALB/c mice. Furthermore, C57BL/6 CD4+ “T responders” were more resistant to Treg-mediated suppression than were BALB/c CD4+ T 28 cells.25 Collectively, these studies support a systemic anti-inflammatory, Th2 bias in BALB/c mice and a more pro-inflammatory, Th1 systemic bias in C57BL/6 mice. Choice of C57BL/6 and BALB/c mouse strains for the following study was made with consideration of the differences described above between these two strains. There is a reported association between Th2-mediated anti-ganglioside IgG1 antibodies and worse clinical outcome in GBS patients,53; 62 and Malik et al (2014) reported that C57BL/6 IL-10-/- mice mounted a Th2-mediated response to GBS patient-derived C. jejuni 260.94.75 As “Th2” biased mice, BALB/c are potential candidates for studying GBS immunopathogenesis. Furthermore, comparing C57BL/6 and BALB/c mice infected with C. jejuni strains known to produce different clinical outcomes will allow elucidation of the contributions of both host and C. jejuni factors in determining the immune response and disease outcome. Objectives of this study included 1) investigating WT and IL-10-/- BALB/c mice as a model of C. jejuni-induced GBS, 2) comparing immune response and disease outcome in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with colitogenic and GBS-associated C. jejuni strains, and 3) performing parallel in vitro assays with dendritic cells from C57BL/6 and BALB/c mice to determine how host-pathogen interactions impacting C. jejuni invasion, intracellular survival, and elicitation of cytokine production contribute to immune responses in vivo. Specific Aim 1: Investigate use of BALB/c (WT and IL-10-/-) mice infected with GBS-associated C. jejuni strain 260.94 as model of GBS Hypotheses: BALB/c mice will develop Th2-mediated responses to C. jejuni infection, and will produce Th2-driven anti-ganglioside antibodies. Neurological disease will manifest as gait abnormalities and neuromuscular weakness observed during neurological phenotyping tests and macrophage infiltration into dorsal root ganglia. Colitis will be mild, if present. Responses will be exacerbated in IL-10-/- compared to WT mice. 29 Specific Aim 2: Describe differences in immune responses and enteric and neurological outcomes in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with colitogenic C. jejuni 11168 and GBS-associated C. jejuni 260.94 Hypotheses (note: formulated after consideration of results from first Specific Aim): Both C57BL/6 IL- 10-/- and BALB/c IL-10-/- mice infected with C. jejuni 11168 will develop Th1/Th17 immune responses and colitis. C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 will develop Th2-mediated responses, including anti-ganglioside antibodies, and nerve lesions manifested by increased macrophage infiltration into dorsal root ganglia. Conversely, BALB/c IL-10-/- mice infected with C. jejuni 260.94 will develop Th1/Th17 responses and be protected from neurological disease, relative to C57BL/6 IL-10-/- mice. Specific Aim 3: Use parallel in vitro assays mirroring the in vivo model of Specific Aim 2 to assess the initial host-microbe interaction, considering both C. jejuni strain differences and host genetic background Hypotheses: Bone marrow-derived dendritic cells (BMDC) from both C57BL/6 IL-10-/- and BALB/c IL-10-/- mice will efficiently internalize and kill both colitogenic and GBS-associated C. jejuni strains. Cytokines produced by infected BMDC derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice will mirror immunogenicity and pathogenicity previously observed with these C. jejuni strains: C. jejuni 11168- infected IL-10-/- BMDC from both mouse strains and C. jejuni 260.94-infected BALB/c IL-10-/- BMDC will produce pro-inflammatory and Th1/Th17-polarizing cytokines. C57BL/6 IL-10-/- BMDC infected with C. jejuni 260.94 will produce Th2-polarizing cytokines. 30 REFERENCES 31 REFERENCES Akira S, Uematsu S, Takeuchi O. 2006. Pathogen recognition and innate immunity. Cell 124:783- 801. Alvarez D, Vollmann EH, von Andrian UH. 2008. Mechanisms and consequences of dendritic cell migration. Immunity 29:325-342. Andersen-Nissen E, Smith KD, Strobe KL, Barrett SL, Cookson BT, Logan SM, Aderem A. 2005. Evasion of Toll-like receptor 5 by flagellated bacteria. Proceedings of the National Academy of Sciences of the United States of America 102:9247-9252. Ang CW, Jacobs BC, Laman JD. 2004. The Guillain-Barre syndrome: a true case of molecular mimicry. Trends in immunology 25:61-66. Ang CW, Laman JD, Willison HJ, Wagner ER, Endtz HP, De Klerk MA, Tio-Gillen AP, Van den Braak N, Jacobs BC, Van Doorn PA. 2002. Structure of Campylobacter jejuni lipopolysaccharides determines antiganglioside specificity and clinical features of Guillain-Barre and Miller Fisher patients. Infection and immunity 70:1202-1208. Backert S, Boehm M, Wessler S, Tegtmeyer N. 2013. Transmigration route of Campylobacter jejuni across polarized intestinal epithelial cells: paracellular, transcellular or both? Cell communication and signaling : CCS 11:72. Bacon DJ, Szymanski CM, Burr DH, Silver RP, Alm RA, Guerry P. 2001. A phase-variable capsule is involved in virulence of Campylobacter jejuni 81-176. Molecular microbiology 40:769-777. Bax M, Kuijf ML, Heikema AP, van Rijs W, Bruijns SC, Garcia-Vallejo JJ, Crocker PR, Jacobs BC, van Vliet SJ, van Kooyk Y. 2011. Campylobacter jejuni lipooligosaccharides modulate dendritic cell-mediated T cell polarization in a sialic acid linkage-dependent manner. Infection and immunity 79:2681-2689. Beery JT, Hugdahl MB, Doyle MP. 1988. Colonization of gastrointestinal tracts of chicks by Campylobacter jejuni. Applied and environmental microbiology 54:2365-2370. Bell JA, Jerome JP, Plovanich-Jones AE, Smith EJ, Gettings JR, Kim HY, Landgraf JR, Lefebure T, Kopper JJ, Rathinam VA, St Charles JL, Buffa BA, Brooks AP, Poe SA, Eaton KA, Stanhope MJ, Mansfield LS. 2013. Outcome of infection of C57BL/6 IL-10(-/-) mice with Campylobacter jejuni strains is correlated with genome content of open reading frames up- and down-regulated in vivo. Microbial pathogenesis 54:1-19. Bell JA, St Charles JL, Murphy AJ, Rathinam VA, Plovanich-Jones AE, Stanley EL, Wolf JE, Gettings JR, Whittam TS, Mansfield LS. 2009. Multiple factors interact to produce responses resembling spectrum of human disease in Campylobacter jejuni infected C57BL/6 IL-10-/- mice. BMC microbiology 9:57. 32 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. Bereswill S, Fischer A, Plickert R, Haag LM, Otto B, Kuhl AA, Dasti JI, Zautner AE, Munoz M, Loddenkemper C, Gross U, Gobel UB, Heimesaat MM. 2011. Novel murine infection models provide deep insights into the "menage a trois" of Campylobacter jejuni, microbiota and host innate immunity. PloS one 6:e20953. Berndtson E, Danielsson-Tham ML, Engvall A. 1994. Experimental colonization of mice with Campylobacter jejuni. Veterinary microbiology 41:183-188. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, Weiner HL, Kuchroo VK. 2006. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441:235-238. Black RE, Levine MM, Clements ML, Hughes TP, Blaser MJ. 1988. Experimental Campylobacter jejuni infection in humans. The Journal of infectious diseases 157:472-479. Blaser M, Engberg J. 2008. Clinical Aspects of Campylobacter jejuni and Campylobacter coli Infections. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Blaser MJ, Duncan DJ. 1984. Human serum antibody response to Campylobacter jejuni infection as measured in an enzyme-linked immunosorbent assay. Infection and immunity 44:292-298. Blaser MJ, Duncan DJ, Warren GH, Wang WL. 1983. Experimental Campylobacter jejuni infection of adult mice. Infection and immunity 39:908-916. Bolton DJ. 2015. Campylobacter virulence and survival factors. Food microbiology 48:99-108. Bouwman LI, de Zoete MR, Bleumink-Pluym NM, Flavell RA, van Putten JP. 2014. Inflammasome activation by Campylobacter jejuni. Journal of immunology 193:4548-4557. Brooks PT, Brakel KA, Bell JA, Bejcek CE, Gilpin T, Brudvig JM, Mansfield LS. 2017. Transplanted human fecal microbiota enhanced Guillain Barre syndrome autoantibody responses after Campylobacter jejuni infection in C57BL/6 mice. Microbiome 5:92. CDC. [Internet]. 2017. Campylobacter (Campylobacteriosis). Available at: https://www.cdc.gov/campylobacter/faq.html. Chang C, Miller JF. 2006. Campylobacter jejuni colonization of mice with limited enteric flora. Infection and immunity 74:5261-5271. Charles PC, Weber KS, Cipriani B, Brosnan CF. 1999. Cytokine, chemokine and chemokine receptor mRNA expression in different strains of normal mice: implications for establishment of a Th1/Th2 bias. Journal of neuroimmunology 100:64-73. Chen X, Oppenheim JJ, Howard OM. 2005. BALB/c mice have more CD4+CD25+ T regulatory cells and show greater susceptibility to suppression of their CD4+CD25- responder T cells than C57BL/6 mice. Journal of leukocyte biology 78:114-121. 33 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Dalpke A, Frank J, Peter M, Heeg K. 2006. Activation of toll-like receptor 9 by DNA from different bacterial species. Infection and immunity 74:940-946. Debruyne L., Gevers D, Vandamme P, 2008. Taxonomy of the Family Campylobacteraceae. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Dekeyser P, Gossuin-Detrain M, Butzler JP, Sternon J. 1972. Acute enteritis due to related vibrio: first positive stool cultures. The Journal of infectious diseases 125:390-392. Edwards LA, Nistala K, Mills DC, Stephenson HN, Zilbauer M, Wren BW, Dorrell N, Lindley KJ, Wedderburn LR, Bajaj-Elliott M. 2010. Delineation of the innate and adaptive T-cell immune outcome in the human host in response to Campylobacter jejuni infection. PloS one 5:e15398. Everest PH, Goossens H, Butzler JP, Lloyd D, Knutton S, Ketley JM, Williams PH. 1992. Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. Journal of medical microbiology 37:319-325. Fauchere JL, Rosenau A, Veron M, Moyen EN, Richard S, Pfister A. 1986. Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infection and immunity 54:283-287. Ferrero RL, Lee A. 1988. Motility of Campylobacter jejuni in a viscous environment: comparison with conventional rod-shaped bacteria. Journal of general microbiology 134:53-59. Fox JG, Rogers AB, Whary MT, Ge Z, Taylor NS, Xu S, Horwitz BH, Erdman SE. 2004. Gastroenteritis in NF-kappaB-deficient mice is produced with wild-type Camplyobacter jejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent colonization with both strains. Infection and immunity 72:1116-1125. Friis LM, Keelan M, Taylor DE. 2009. Campylobacter jejuni drives MyD88-independent interleukin-6 secretion via Toll-like receptor 2. Infection and immunity 77:1553-1560. Gershon RK, Kondo K. 1970. Cell interactions in the induction of tolerance: the role of thymic lymphocytes. Immunology 18:723-737. Gilbert M, Parker CT, Moran AP, 2008. Campylobacter jejuni Lipooligosaccharides: Structures and Biosynthesis. Campylobacter, 3rd ed. Washinton, D.C.: ASM Press. Godschalk PC, Kuijf ML, Li J, St Michael F, Ang CW, Jacobs BC, Karwaski MF, Brochu D, Moterassed A, Endtz HP, van Belkum A, Gilbert M. 2007. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barre and Miller Fisher syndromes. Infection and immunity 75:1245-1254. Goverse G, Olivier BJ, Molenaar R, Knippenberg M, Greuter M, Konijn T, Cook EC, Beijer MR, Fedor DM, den Haan JM, Napoli JL, Bouma G, Mebius RE. 2015. Vitamin A metabolism and mucosal immune function are distinct between BALB/c and C57BL/6 mice. European journal of immunology 45:89-100. 34 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Guerry P, Szymanski CM, Prendergast MM, Hickey TE, Ewing CP, Pattarini DL, Moran AP. 2002. Phase variation of Campylobacter jejuni 81-176 lipooligosaccharide affects ganglioside mimicry and invasiveness in vitro. Infection and immunity 70:787-793. Hadden RD, Karch H, Hartung HP, Zielasek J, Weissbrich B, Schubert J, Weishaupt A, Cornblath DR, Swan AV, Hughes RA, Toyka KV, Plasma Exchange/Sandoglobulin Guillain-Barre Syndrome Trial G. 2001. Preceding infections, immune factors, and outcome in Guillain-Barre syndrome. Neurology 56:758-765. Halstead SK, Zitman FM, Humphreys PD, Greenshields K, Verschuuren JJ, Jacobs BC, Rother RP, Plomp JJ, Willison HJ. 2008. Eculizumab prevents anti-ganglioside antibody-mediated neuropathy in a murine model. Brain : a journal of neurology 131:1197-1208. Havelaar AH, van Pelt W, Ang CW, Wagenaar JA, van Putten JP, Gross U, Newell DG. 2009. Immunity to Campylobacter: its role in risk assessment and epidemiology. Critical reviews in microbiology 35:1-22. Heimesaat MM, Grundmann U, Alutis ME, Fischer A, Bereswill S. 2017. Absence of Nucleotide- Oligomerization-Domain-2 Is Associated with Less Distinct Disease in Campylobacter jejuni Infected Secondary Abiotic IL-10 Deficient Mice. Frontiers in cellular and infection microbiology 7:322. Hickey TE, Baqar S, Bourgeois AL, Ewing CP, Guerry P. 1999. Campylobacter jejuni-stimulated secretion of interleukin-8 by INT407 cells. Infection and immunity 67:88-93. Hickey TE, Majam G, Guerry P. 2005. Intracellular survival of Campylobacter jejuni in human monocytic cells and induction of apoptotic death by cytholethal distending toxin. Infection and immunity 73:5194-5197. Hickey TE, McVeigh AL, Scott DA, Michielutti RE, Bixby A, Carroll SA, Bourgeois AL, Guerry P. 2000. Campylobacter jejuni cytolethal distending toxin mediates release of interleukin-8 from intestinal epithelial cells. Infection and immunity 68:6535-6541. Hofreuter D, Tsai J, Watson RO, Novik V, Altman B, Benitez M, Clark C, Perbost C, Jarvie T, Du L, Galan JE. 2006. Unique features of a highly pathogenic Campylobacter jejuni strain. Infection and immunity 74:4694-4707. Hu L, Bray MD, Osorio M, Kopecko DJ. 2006. Campylobacter jejuni induces maturation and cytokine production in human dendritic cells. Infection and immunity 74:2697-2705. Hu L, Hickey TE. 2005. Campylobacter jejuni induces secretion of proinflammatory chemokines from human intestinal epithelial cells. Infection and immunity 73:4437-4440. Hu L, McDaniel JP, Kopecko DJ. 2006. Signal transduction events involved in human epithelial cell invasion by Campylobacter jejuni 81-176. Microbial pathogenesis 40:91-100. Hughes RA, Cornblath DR. 2005. Guillain-Barre syndrome. Lancet 366:1653-1666. 35 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. Jacobs BC, Hazenberg MP, van Doorn PA, Endtz HP, van der Meche FG. 1997. Cross-reactive antibodies against gangliosides and Campylobacter jejuni lipopolysaccharides in patients with Guillain-Barre or Miller Fisher syndrome. The Journal of infectious diseases 175:729-733. Jacobs BC, Koga M, van Rijs W, Geleijns K, van Doorn PA, Willison HJ, Yuki N. 2008. Subclass IgG to motor gangliosides related to infection and clinical course in Guillain-Barre syndrome. Journal of neuroimmunology 194:181-190. Jacobs BC, Rothbarth PH, van der Meche FG, Herbrink P, Schmitz PI, de Klerk MA, van Doorn PA. 1998. The spectrum of antecedent infections in Guillain-Barre syndrome: a case-control study. Neurology 51:1110-1115. Jacobs BC, van Belkum A, Endtz HP, 2008. Guillain-Barre Syndrome and Campylobacter Infection. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Janssen R, Krogfelt KA, Cawthraw SA, van Pelt W, Wagenaar JA, Owen RJ. 2008. Host-pathogen interactions in Campylobacter infections: the host perspective. Clinical microbiology reviews 21:505-518. Joeris T, Muller-Luda K, Agace WW, Mowat AM. 2017. Diversity and functions of intestinal mononuclear phagocytes. Mucosal immunology 10:845-864. Karlyshev AV, Linton D, Gregson NA, Lastovica AJ, Wren BW. 2000. Genetic and biochemical evidence of a Campylobacter jejuni capsular polysaccharide that accounts for Penner serotype specificity. Molecular microbiology 35:529-541. Keithlin J, Sargeant J, Thomas MK, Fazil A. 2014. Systematic review and meta-analysis of the proportion of Campylobacter cases that develop chronic sequelae. BMC public health 14:1203. Khader SA, Gaffen SL, Kolls JK. 2009. Th17 cells at the crossroads of innate and adaptive immunity against infectious diseases at the mucosa. Mucosal immunology 2:403-411. Kiehlbauch JA, Albach RA, Baum LL, Chang KP. 1985. Phagocytosis of Campylobacter jejuni and its intracellular survival in mononuclear phagocytes. Infection and immunity 48:446-451. Koga M, Yuki N, Hirata K, Morimatsu M, Mori M, Kuwabara S. 2003. Anti-GM1 antibody IgG subclass: a clinical recovery predictor in Guillain-Barre syndrome. Neurology 60:1514-1518. Konkel ME, Garvis SG, Tipton SL, Anderson DE, Jr., Cieplak W, Jr. 1997. Identification and molecular cloning of a gene encoding a fibronectin-binding protein (CadF) from Campylobacter jejuni. Molecular microbiology 24:953-963. Konkel ME, Joens LA. 1989. Adhesion to and invasion of HEp-2 cells by Campylobacter spp. Infection and immunity 57:2984-2990. Konkel ME, Klena JD, Rivera-Amill V, Monteville MR, Biswas D, Raphael B, Mickelson J. 2004. Secretion of virulence proteins from Campylobacter jejuni is dependent on a functional flagellar export apparatus. Journal of bacteriology 186:3296-3303. 36 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. Kuijf ML, Samsom JN, van Rijs W, Bax M, Huizinga R, Heikema AP, van Doorn PA, van Belkum A, van Kooyk Y, Burgers PC, Luider TM, Endtz HP, Nieuwenhuis EE, Jacobs BC. 2010. TLR4- mediated sensing of Campylobacter jejuni by dendritic cells is determined by sialylation. Journal of immunology 185:748-755. Lastovica A, Allos BM, 2008. Clinical Significance of Campylobacter and Related Species Other Than Campylobacter jejuni and Campylobacter coli. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Law BF, Adriance SM, Joens LA. 2009. Comparison of in vitro virulence factors of Campylobacter jejuni to in vivo lesion production. Foodborne pathogens and disease 6:377-385. Lee MD, Newell DG. 2006. Campylobacter in poultry: filling an ecological niche. Avian diseases 50:1-9. Li CY, Xue P, Tian WQ, Liu RC, Yang C. 1996. Experimental Campylobacter jejuni infection in the chicken: an animal model of axonal Guillain-Barre syndrome. Journal of neurology, neurosurgery, and psychiatry 61:279-284. Liu K, Nussenzweig MC. 2010. Origin and development of dendritic cells. Immunological reviews 234:45-54. Liu T, Matsuguchi T, Tsuboi N, Yajima T, Yoshikai Y. 2002. Differences in expression of toll-like receptors and their reactivities in dendritic cells in BALB/c and C57BL/6 mice. Infection and immunity 70:6638-6645. Louwen R, Heikema A, van Belkum A, Ott A, Gilbert M, Ang W, Endtz HP, Bergman MP, Nieuwenhuis EE. 2008. The sialylated lipooligosaccharide outer core in Campylobacter jejuni is an important determinant for epithelial cell invasion. Infection and immunity 76:4431-4438. Louwen R, Horst-Kreft D, de Boer AG, van der Graaf L, de Knegt G, Hamersma M, Heikema AP, Timms AR, Jacobs BC, Wagenaar JA, Endtz HP, van der Oost J, Wells JM, Nieuwenhuis EE, van Vliet AH, Willemsen PT, van Baarlen P, van Belkum A. 2013. A novel link between Campylobacter jejuni bacteriophage defence, virulence and Guillain-Barre syndrome. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 32:207-226. Malik A, Sharma D, St Charles J, Dybas LA, Mansfield LS. 2014. Contrasting immune responses mediate Campylobacter jejuni-induced colitis and autoimmunity. Mucosal immunology 7:802- 817. Mansfield LS, Bell JA, Wilson DL, Murphy AJ, Elsheikha HM, Rathinam VA, Fierro BR, Linz JE, Young VB. 2007. C57BL/6 and congenic interleukin-10-deficient mice can serve as models of Campylobacter jejuni colonization and enteritis. Infection and immunity 75:1099-1115. 37 Mansfield LS, Patterson JS, Fierro BR, Murphy AJ, Rathinam VA, Kopper JJ, Barbu NI, Onifade TJ, Bell JA. 2008. Genetic background of IL-10(-/-) mice alters host-pathogen interactions with Campylobacter jejuni and influences disease phenotype. Microbial pathogenesis 45:241-257. Mansfield LS, Schauer DB, Fox JG, 2008. Animal Models of Campylobacter jejuni Infection. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Maue AC, Mohawk KL, Giles DK, Poly F, Ewing CP, Jiao Y, Lee G, Ma Z, Monteiro MA, Hill CL, Ferderber JS, Porter CK, Trent MS, Guerry P. 2013. The polysaccharide capsule of Campylobacter jejuni modulates the host immune response. Infection and immunity 81:665- 672. McGrogan A, Madle GC, Seaman HE, de Vries CS. 2009. The epidemiology of Guillain-Barre syndrome worldwide. A systematic literature review. Neuroepidemiology 32:150-163. McSweegan E, Walker RI. 1986. Identification and characterization of two Campylobacter jejuni adhesins for cellular and mucous substrates. Infection and immunity 53:141-148. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. 2000. M-1/M-2 macrophages and the Th1/Th2 paradigm. Journal of immunology 164:6166-6173. Monteville MR, Konkel ME. 2002. Fibronectin-facilitated invasion of T84 eukaryotic cells by Campylobacter jejuni occurs preferentially at the basolateral cell surface. Infection and immunity 70:6665-6671. Mosmann TR, Cherwinski H, Bond MW, Giedlin MA, Coffman RL. 1986. Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. Journal of immunology 136:2348-2357. Nachamkin I, Allos BM, Ho T. 1998. Campylobacter species and Guillain-Barre syndrome. Clinical microbiology reviews 11:555-567. Nemelka KW, Brown AW, Wallace SM, Jones E, Asher LV, Pattarini D, Applebee L, Gilliland TC, Jr., Guerry P, Baqar S. 2009. Immune response to and histopathology of Campylobacter jejuni infection in ferrets (Mustela putorius furo). Comparative medicine 59:363-371. Nyati KK, Prasad KN, Kharwar NK, Soni P, Husain N, Agrawal V, Jain AK. 2012. Immunopathology and Th1/Th2 immune response of Campylobacter jejuni-induced paralysis resembling Guillain-Barre syndrome in chicken. Medical microbiology and immunology 201:177- 187. Oelschlaeger TA, Guerry P, Kopecko DJ. 1993. Unusual microtubule-dependent endocytosis mechanisms triggered by Campylobacter jejuni and Citrobacter freundii. Proceedings of the National Academy of Sciences of the United States of America 90:6884-6888. Olson CK, Ethelberg S, van Pelt W, Tauxe RV, 2008. Epidemiology of Campylobacter jejuni Infections in Industrialized Nations. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. 38 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. Oosterom J, Notermans S, Karman H, Engels GB. 1983. Origin and Prevalence of Campylobacter jejuni in Poultry Processing. Journal of food protection 46:339-344. Park H, Li Z, Yang XO, Chang SH, Nurieva R, Wang YH, Wang Y, Hood L, Zhu Z, Tian Q, Dong C. 2005. A distinct lineage of CD4 T cells regulates tissue inflammation by producing interleukin 17. Nat Immunol 6:1133-1141. Parker CT, Huynh S, Heikema AP, Cooper KK, Miller WG. 2015. Complete Genome Sequences of Campylobacter jejuni Strains RM3196 (233.94) and RM3197 (308.95) Isolated from Patients with Guillain-Barre Syndrome. Genome announcements 3. Parkhill J, Wren BW, Mungall K, Ketley JM, Churcher C, Basham D, Chillingworth T, Davies RM, Feltwell T, Holroyd S, Jagels K, Karlyshev AV, Moule S, Pallen MJ, Penn CW, Quail MA, Rajandream MA, Rutherford KM, van Vliet AH, Whitehead S, Barrell BG. 2000. The genome sequence of the food-borne pathogen Campylobacter jejuni reveals hypervariable sequences. Nature 403:665-668. Poropatich KO, Walker CL, Black RE. 2010. Quantifying the association between Campylobacter infection and Guillain-Barre syndrome: a systematic review. Journal of health, population, and nutrition 28:545-552. Prendergast MM, Moran AP. 2000. Lipopolysaccharides in the development of the Guillain- Barre syndrome and Miller Fisher syndrome forms of acute inflammatory peripheral neuropathies. Journal of endotoxin research 6:341-359. Raphael I, Nalawade S, Eagar TN, Forsthuber TG. 2015. T cell subsets and their signature cytokines in autoimmune and inflammatory diseases. Cytokine 74:5-17. Rathinam VA, Appledorn DM, Hoag KA, Amalfitano A, Mansfield LS. 2009. Campylobacter jejuni-induced activation of dendritic cells involves cooperative signaling through Toll-like receptor 4 (TLR4)-MyD88 and TLR4-TRIF axes. Infection and immunity 77:2499-2507. Rathinam VA, Hoag KA, Mansfield LS. 2008. Dendritic cells from C57BL/6 mice undergo activation and induce Th1-effector cell responses against Campylobacter jejuni. Microbes and infection 10:1316-1324. Rees JH, Gregson NA, Hughes RA. 1995. Anti-ganglioside GM1 antibodies in Guillain-Barre syndrome and their relationship to Campylobacter jejuni infection. Annals of neurology 38:809- 816. Salomon B, Rhee L, Bour-Jordan H, Hsin H, Montag A, Soliven B, Arcella J, Girvin AM, Padilla J, Miller SD, Bluestone JA. 2001. Development of spontaneous autoimmune peripheral polyneuropathy in B7-2-deficient NOD mice. The Journal of experimental medicine 194:677-684. Samuelson DR, Eucker TP, Bell JA, Dybas L, Mansfield LS, Konkel ME. 2013. The Campylobacter jejuni CiaD effector protein activates MAP kinase signaling pathways and is required for the development of disease. Cell communication and signaling : CCS 11:79. 39 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson MA, Roy SL, Jones JL, Griffin PM. 2011. Foodborne illness acquired in the United States--major pathogens. Emerging infectious diseases 17:7-15. Sejvar JJ, Baughman AL, Wise M, Morgan OW. 2011. Population incidence of Guillain-Barre syndrome: a systematic review and meta-analysis. Neuroepidemiology 36:123-133. Shang Y, Ren F, Song Z, Li Q, Zhou X, Wang X, Xu Z, Bao G, Wan T, Lei T, Wang N, Jiao XA, Huang J. 2016. Insights into Campylobacter jejuni colonization and enteritis using a novel infant rabbit model. Scientific reports 6:28737. Sheikh KA, Nachamkin I, Ho TW, Willison HJ, Veitch J, Ung H, Nicholson M, Li CY, Wu HS, Shen BQ, Cornblath DR, Asbury AK, McKhann GM, Griffin JW. 1998. Campylobacter jejuni lipopolysaccharides in Guillain-Barre syndrome: molecular mimicry and host susceptibility. Neurology 51:371-378. Shortman K, Naik SH. 2007. Steady-state and inflammatory dendritic-cell development. Nature reviews Immunology 7:19-30. Skirrow MB. 1977. Campylobacter enteritis: a "new" disease. British medical journal 2:9-11. St Charles JL, Bell JA, Gadsden BJ, Malik A, Cooke H, Van de Grift LK, Kim HY, Smith EJ, Mansfield LS. 2017. Guillain Barre Syndrome is induced in Non-Obese Diabetic (NOD) mice following Campylobacter jejuni infection and is exacerbated by antibiotics. Journal of autoimmunity 77:11-38. Steinman RM, Cohn ZA. 1973. Identification of a novel cell type in peripheral lymphoid organs of mice. I. Morphology, quantitation, tissue distribution. The Journal of experimental medicine 137:1142-1162. Steinman RM, Cohn ZA. 1974. Identification of a novel cell type in peripheral lymphoid organs of mice. II. Functional properties in vitro. The Journal of experimental medicine 139:380-397. van den Berg B, Walgaard C, Drenthen J, Fokke C, Jacobs BC, van Doorn PA. 2014. Guillain- Barre syndrome: pathogenesis, diagnosis, treatment and prognosis. Nature reviews Neurology 10:469-482. van Spreeuwel JP, Duursma GC, Meijer CJ, Bax R, Rosekrans PC, Lindeman J. 1985. Campylobacter colitis: histological immunohistochemical and ultrastructural findings. Gut 26:945-951. 113. Walsh KP, Mills KH. 2013. Dendritic cells and other innate determinants of T helper cell polarisation. Trends in immunology 34:521-530. 114. Wassenaar TM, Blaser MJ. 1999. Pathophysiology of Campylobacter jejuni infections of humans. Microbes and infection 1:1023-1033. 40 115. Wassenaar TM, Engelskirchen M, Park S, Lastovica A. 1997. Differential uptake and killing potential of Campylobacter jejuni by human peripheral monocytes/macrophages. Medical microbiology and immunology 186:139-144. 116. Watanabe H, Numata K, Ito T, Takagi K, Matsukawa A. 2004. Innate immune response in Th1- and Th2-dominant mouse strains. Shock 22:460-466. 117. Watson RO, Galan JE. 2008. Campylobacter jejuni survives within epithelial cells by avoiding delivery to lysosomes. PLoS pathogens 4:e14. 118. Weaver CT, Harrington LE, Mangan PR, Gavrieli M, Murphy KM. 2006. Th17: an effector CD4 T cell lineage with regulatory T cell ties. Immunity 24:677-688. 119. Weis AM, Clothier KA, Huang BC, Kong N, Weimer BC. 2016. Draft Genome Sequences of Campylobacter jejuni Strains That Cause Abortion in Livestock. Genome announcements 4. 120. Whitehouse CA, Balbo PB, Pesci EC, Cottle DL, Mirabito PM, Pickett CL. 1998. Campylobacter jejuni cytolethal distending toxin causes a G2-phase cell cycle block. Infection and immunity 66:1934-1940. 121. WHO. [Internet]. 2016. Campylobacter Fact Sheet. Available at: http://www.who.int/mediacentre/factsheets/fs255/en/. 122. 123. 124. 125. 126. 127. 128. Xia RH, Yosef N, Ubogu EE. 2010. Clinical, electrophysiological and pathologic correlations in a severe murine experimental autoimmune neuritis model of Guillain-Barre syndrome. Journal of neuroimmunology 219:54-63. Yang F, Wang D, Li Y, Sang L, Zhu J, Wang J, Wei B, Lu C, Sun X. 2017. Th1/Th2 Balance and Th17/Treg-Mediated Immunity in relation to Murine Resistance to Dextran Sulfate-Induced Colitis. Journal of immunology research 2017:7047201. Young KT, Davis LM, Dirita VJ. 2007. Campylobacter jejuni: molecular biology and pathogenesis. Nature reviews Microbiology 5:665-679. Yuki N, Hartung HP. 2012. Guillain-Barre syndrome. The New England journal of medicine 366:2294-2304. Yuki N, Susuki K, Koga M, Nishimoto Y, Odaka M, Hirata K, Taguchi K, Miyatake T, Furukawa K, Kobata T, Yamada M. 2004. Carbohydrate mimicry between human ganglioside GM1 and Campylobacter jejuni lipooligosaccharide causes Guillain-Barre syndrome. Proceedings of the National Academy of Sciences of the United States of America 101:11404-11409. Yuki N, Taki T, Inagaki F, Kasama T, Takahashi M, Saito K, Handa S, Miyatake T. 1993. A bacterium lipopolysaccharide that elicits Guillain-Barre syndrome has a GM1 ganglioside-like structure. The Journal of experimental medicine 178:1771-1775. Zilbauer M, Dorrell N, Elmi A, Lindley KJ, Schuller S, Jones HE, Klein NJ, Nunez G, Wren BW, Bajaj-Elliott M. 2007. A major role for intestinal epithelial nucleotide oligomerization domain 1 41 (NOD1) in eliciting host bactericidal immune responses to Campylobacter jejuni. Cellular microbiology 9:2404-2416. 42 CHAPTER 2: BALB/C MICE INFECTED WITH GUILLAIN-BARRÉ SYNDROME-ASSOCIATED CAMPYLOBACTER JEJUNI STRAIN 260.94 EXHIBIT TH1/TH17-MEDIATED IMMUNITY To be submitted to: Comparative Medicine ABSTRACT Campylobacter jejuni is an important cause of bacterial gastroenteritis worldwide. A small proportion of C. jejuni infections are associated with subsequent development of Guillain-Barré syndrome (GBS), a debilitating polyneuropathy. The immunopathogenesis of GBS is poorly understood, but is thought to involve generation of antibodies cross-reactive to lipooligosaccharide on the C. jejuni outer membrane and structurally similar peripheral nerve gangliosides. Both infecting C. jejuni strain characteristics and host factors have been implicated in development of GBS as opposed to uncomplicated enteritis. Induction of anti-ganglioside antibodies in association with T helper 2 (Th2)-mediated immune responses was reported previously in C57BL/6 interleukin (IL)-10-/- mice infected with GBS patient- derived C. jejuni strain 260.94. In the current study, we investigated BALB/c mice, a strain reported to exhibit a Th2 immunological bias, as candidates for development of a GBS model. Wild-type (WT) and IL- 10-/- BALB/c mice were orally inoculated with C. jejuni 260.94 and humanely euthanized after 5 weeks. Immune response was assessed by evaluation of C. jejuni-specific and anti-ganglioside plasma antibodies, in addition to histological lesions and cytokine production in the proximal colon. Mice were tested for neurological deficits by weekly phenotyping tests. Peripheral nerve lesions were assessed by scoring the number of macrophages in dorsal root ganglia. C. jejuni 260.94 stably colonized both WT and IL-10-/- mice and induced a systemic Th1/Th17-mediated immune response, as reflected by significant increases in C. jejuni-specific IgG2a, IgG2b, and IgG3 plasma antibodies. However, C. jejuni 260.94 did not induce production of anti-ganglioside antibodies, colitis, or marked neurological deficits or peripheral nerve lesions in either WT or IL-10-/- mice. In the current study, both WT and IL-10-/- BALB/c mice exhibited relative protection from development of Th2-mediated immunity and anti-ganglioside antibodies as seen previously in C57BL/6 IL-10-/- mice. These results indicate that BALB/c mice are a 43 useful model for studying the immune response to C. jejuni and provide insight into the role of host genetic background in determining susceptibility to GBS following C. jejuni infection. 44 INTRODUCTION Campylobacter jejuni, a spiral, motile, Gram-negative, microaerophilic bacterium, is a common cause of human bacterial gastroenteritis worldwide, with the World Health Organization considering members of the genus Campylobacter as one of the four key global causes of diarrheal disease.58 While C. jejuni can colonize chickens in high numbers without causing apparent disease, symptoms in infected people typically include diarrhea, abdominal pain, and fever.58; 62 Although usually self-limiting, human campylobacteriosis is linked to multiple post-infectious complications including irritable bowel syndrome, inflammatory bowel disease, reactive arthritis, and Guillain-Barré syndrome (GBS).26; 58 GBS is debilitating, with many patients experiencing protracted recovery and residual neurologic deficits.55 This syndrome is clinically heterogeneous, but common characteristic symptoms include rapidly progressing limb weakness and reflex deficits. Pain, weakness, ataxia, and respiratory insufficiency are variably seen depending upon the specific GBS disease subtype present.55 Current reported estimates of GBS incidence following C. jejuni infection vary from 1/1,000- 5,000 to 7/10,000.26; 55 Although several infectious agents have been associated with GBS, C. jejuni has been identified as the most common cause of antecedent infection in GBS in epidemiological studies.19; 25 The author of a recent literature review estimated that of >2,500 GBS cases, 31% were attributable to previous Campylobacter infection.44 Structural similarity between the lipooligosaccharide (LOS) of some C. jejuni isolates and peripheral nerve gangliosides, such as GM1, has been implicated in development of anti-ganglioside antibodies and GBS.2; 40; 64 Antibodies to gangliosides, particularly to GM1 and often of the IgG isotype, are frequently detected in GBS patient sera.45 Furthermore, presence of anti-ganglioside antibodies is correlated with slower patient clinical recovery and increased disability, and clinical improvement was seen with decreasing anti-ganglioside antibodies in patients with C. jejuni-associated GBS and the Miller- Fisher syndrome (MFS) GBS subtype.23; 47 45 Ganglioside-like epitopes were found more frequently in LOS from C. jejuni isolates from GBS or MFS patients than in isolates from those with uncomplicated enteritis.2 However, presence of ganglioside-like moieties alone appears insufficient to elicit GBS, as expression of ganglioside mimics can be found in C. jejuni isolates from enteritis patients without GBS as well.2; 53 These findings suggest a potential role of both C. jejuni-specific factors such as presence of ganglioside mimics on the LOS and host factors such as immunogenetic background in susceptibility to GBS following campylobacteriosis. Murine models resembling C. jejuni-induced colitis seen in humans have been developed in recent years. These models,5; 10; 16; 35; 36 which exploit alterations in the immune system, microbiota, or both, provide robust models of disease and overcome the transient or persistent asymptomatic colonization observed in previously studied murine models.7; 8 Of particular relevance to the current study was the development of a colitis model in C57BL/6 mice lacking anti-inflammatory interleukin (IL)- 10 following infection with C. jejuni 11168.36 Development of spontaneous enterocolitis with age in IL- 10-/- mice on different genetic backgrounds is well documented, and results primarily from unchecked T helper 1 (Th1)-mediated immunological responses to intestinal microbiota in the absence of the regulatory action of IL-10.6; 28 This cytokine has important immunosuppressive action, especially upon monocytes and macrophages, and inhibits release of pro-inflammatory mediators.49 In our studies, absence of IL-10 has proven critical for induction of inflammatory disease following C. jejuni infection, as wild-type (WT) mice infected with colitogenic C. jejuni strain 11168 are stably colonized, but do not develop the severe colitis seen in infected IL-10-/- mice.9; 36; 37 Significant progress has been made in developing murine colitis models, but robust animal models of human GBS have proven difficult to develop. Models in species including mice,20; 50; 60 rabbits,63 and chicken30; 42 bear some similarity to various aspects of human disease. However, anatomic and physiologic differences between birds and humans limit the utility of the chicken model. Aside from the spontaneous autoimmune peripheral polyneuropathy occurring in mice,50 the aforementioned 46 mammalian models are experimentally induced and require injection of LOS,63 myelin,60 or anti- ganglioside antibody20 with additional factors such as Freund’s complete adjuvant60; 63 or normal human serum20 for disease induction. Importantly, mouse models of GBS induced following oral infection with C. jejuni have been developed that mimic the typical route of human infection.35 9; 54 C57BL/6 IL-10-/- mice infected with C. jejuni 260.94, a strain isolated from a GBS patient, developed Th2-mediated responses including production of IgG1 antibodies to gangliosides GM1 and GD1a.35 WT non-obese diabetic (NOD) mice and congenic IL-10-/- and costimulatory molecule B7-2-/- mice infected with C. jejuni 260.94 showed increased generation of anti-ganglioside antibodies and macrophage and T cell infiltration into peripheral nerves, including the sciatic nerve and dorsal root ganglia (DRG), compared to controls.54 The microbiota also contributes to anti-ganglioside antibody production following C. jejuni 260.94 infection in WT C57BL/6 mice, as increased anti-GM1 IgG1 antibodies were seen in mice with a human-derived microbiota compared to mice with a conventional mouse microbiota.9 These models show that expression of ganglioside mimics in an infecting C. jejuni LOS elicits the production of antiganglioside antibodies and likely contributes to peripheral nerve damage observed, yet other factors may also contribute. The importance of host factors in determining susceptibility to GBS is poorly understood. Mouse strains including C57BL/6 and NOD backgrounds have been studied, and development of additional mouse models of different genetic backgrounds would facilitate understanding of the impact of host immune response in mediating the neuropathology in GBS. Another prerequisite of such studies would be that comparisons of mice of different genetic backgrounds be carried out with a rigorously similar study design. The choice of mouse strains to compare is critical. C57BL/6 IL-10-/- mice infected with colitogenic C. jejuni 11168 developed colitis and Th1/Th17-mediated immune responses, whereas mice of the same genotype infected with GBS patient-derived C. jejuni 260.94 mounted Th2-mediated responses, including IgG1 reacting with GM1 and GD1a gangliosides, but did not develop colitis.35 47 Several reports in the literature indicate immunological biases between BALB/c and C57BL/6 mice. One classic example is infection with Leishmania major, in which susceptible BALB/c mice produce high amounts of IL-4, while resistant C57BL/6 mice down-regulate IL-4 production and instead produce IFN- γ.48 Multiple studies have demonstrated that upon in vitro stimulation, peritoneal macrophages, spleen cells, and dendritic cells derived from BALB/c mice exhibit a relative Th2 bias when exposed to various stimuli compared to C57BL/6 mice, which exhibit a more inflammatory, Th1-mediated bias.32; 33; 39; 56 In two studies of GBS patients, presence of IgG1 anti-ganglioside antibodies alone was associated with previous C. jejuni infection and worse clinical outcomes, while IgG3 antibodies, either alone or in combination with IgG1 anti-ganglioside antibodies, were associated with preceding respiratory infection and better outcomes.24; 27 Thus, we sought to develop a GBS mouse model characterized by Th2-biased responses after enteric infection, resulting in strong IgG1 responses. Considering the production of Th2-mediated responses in C57BL/6 IL-10-/- mice infected with GBS patient-derived C. jejuni 260.94, the reported relative Th2-bias in BALB/c compared to C57BL/6 mice, and the increased severity of GBS associated with anti-ganglioside IgG1 in people, we reasoned that BALB/c mice are potential candidates for a C. jejuni-induced GBS model. We hypothesized that WT and IL-10-/- BALB/c mice infected with C. jejuni 260.94 would 1) mount systemic and local (gastrointestinal (GI) tract) Th2-driven immune responses, including production of anti-ganglioside antibodies, without development of colitis, 2) develop gait abnormalities, neuromuscular weakness, and macrophage infiltration into peripheral nerves, reflecting C. jejuni-associated neuropathology; and finally, 3) that immune responses and neuropathology would be exacerbated in mice lacking anti- inflammatory IL-10. To test these hypotheses, BALB/c mice (WT and IL-10-/-) were orally infected with C. jejuni 260.94. This strain was originally isolated from a GBS patient in South Africa and harbors a GM1a ganglioside mimic. In our GBS models including mice of NOD and C57BL/6 genetic backgrounds, it 48 colonized well without inducing colitis9 4; 35; 54 and has induced anti-ganglioside antibodies35; 54 and nerve lesions.54 ELISAs were used to measure plasma anti -C. jejuni, -GM1, and -GD1a IgG antibody subtypes. A flow-cytometry based multiplexed bead assay was used to measure cytokines reflecting local Th adaptive immune responses in the proximal colon. Histological lesions in the gastrointestinal tract were graded according to a previously published scale.36 Neurological phenotyping tests included DigiGait, open field test (OFT), and hang testing. Morphometry was used to quantitate number of macrophages labeled immunohistochemically with F4/80 in lumbar dorsal root ganglia (DRG). Results from this study indicate that C. jejuni 260.94 colonized infected BALB/c mice well and induced a Th1/Th17 response that was exacerbated in IL-10-/- mice, but did not produce colitis. Significant differences in anti-ganglioside antibody isotypes were related to mouse genotype, but not C. jejuni infection status, and did not correlate with increased macrophage numbers in DRG. No overt neurological phenotype was observed in any experimental mouse. Thus, surprisingly this reportedly Th2- biased mouse strain, when infected with a C. jejuni strain expressing ganglioside mimics in the outer core that previously induced a Th2 response in C57BL/6 IL-10-/- mice, mounted predominantly Th1/Th17- mediated responses. The contrasting immune responses between mice of different genetic backgrounds to the same C. jejuni strain seen in this study and previously35 offer insight into the role of host factors in determining susceptibility to GBS following C. jejuni infection. Additionally, these results support the use of WT and IL-10-/- mice on the BALB/c genetic background as an additional model for studying immune responses to C. jejuni infection. MATERIALS AND METHODS Mice. All mouse experiments were performed according to recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All procedures were approved by the Michigan State University (MSU) Institutional Animal Care and Use Committee (approval numbers 49 06/12-107-00 and 06/15-101-00). BALB/cJ (referred to as BALB/c WT) and C.129P2(B6)-Il10tm1Cgn/J (referred to as BALB/c IL-10-/-) mice were purchased from the Jackson Laboratory (Bar Harbor, ME) and bred in house. Age matched male and female mice used in the experiment were bred and maintained in-house. Husbandry has been previously described.36 Wild-type or IL-10-deficient genotype was verified using a PCR assay offered by the Jackson Laboratory (https://www2.jax.org/protocolsdb/f?p=116:5:0::NO:5:P5_MASTER_PROTOCOL_ID,P5_JRS_CODE:23475 ,004333). DNA extracted from feces or cecal tissue collected from experimental mice used in this study was screened for enteric pathogens including Citrobacter rodentium, Enterococcus faecium and E. faecalis, Helicobacter spp., and Campylobacter spp. by PCR as previously described.36 Prior to inoculation, mice were transported to the MSU’s University Research Containment Facility. The mice were housed individually following acclimation. At the humane endpoint or at the end of the experiment, all mice were humanely euthanized using an overdose of CO2 according to the guidelines of the American Veterinary Medical Association (https://www.avma.org/KB/Policies/Pages/Euthanasia- Guidelines.aspx). Experimental Design. Prior to this study, the ability of C. jejuni 260.94 to colonize BALB/c WT mice was verified by inoculation of 5 mice with C. jejuni 260.94 (dose = 3.7 × 109 colony forming units, CFU) by gastric gavage. Three mice inoculated with tryptic soy broth (TSB; sham/vehicle) served as controls. The inoculum was prepared and confirmation of colonization of C. jejuni 260.94 by culture and PCR for this pilot study was performed as described below. Forty mice, including 20 BALB/c WT and 20 BALB/c IL-10-/- mice, were used in this study. Mice of each genotype were randomized regarding sex, litter, treatment group, and cage position upon the rack. Mice were inoculated at 14─15 weeks of age. Ten BALB/c WT and 10 BALB/c IL-10-/- mice received TSB; likewise, 10 BALB/c WT and 10 BALB/c IL-10-/- mice received C. jejuni 260.94. BALB/c WT mice were 50 inoculated one week before BALB/c IL-10-/- mice. Both groups were sacrificed 5 weeks following their respective inoculation. Campylobacter jejuni Inoculum Preparation and Inoculation. Campylobacter jejuni strain 260.94 originally obtained from ATCC (strain BAA-1234) and stored as glycerol stock cultures at -80°C was used in this study. The inocula for the BALB/c WT and BALB/c IL-10-/- groups were prepared as previously described.36 Briefly, for each preparation, inoculum from frozen stock was streaked onto tryptic soy agar plates supplemented with 5% defibrinated sheep’s blood (Cleveland Scientific, Bath, OH) (TSAB plates). The plates were incubated for approximately 24 hours at 37°C in vented anaerobic jars. Generation of the microaerobic environment was achieved by incubation with a CampyGen sachet (Oxoid, Basingstoke, United Kingdom) or by evacuation of the jar to -25 in Hg followed by equilibration with a gas mixture comprising 80% N2, 10% CO2, and 10% H2. Colonies were harvested, suspended in TSB, spread into lawns on fresh TSAB plates, and incubated overnight at 37°C in the microaerobic environment. The following day, lawns were harvested and suspended in TSB, aiming to achieve an optical density at 600 nm of approximately 1.0 in a 1:10 dilution of culture. Motility, spiral morphology, and purity of both inoculums were confirmed by wet mount and Gram stain preparations. Mice were inoculated with 0.1 mL of C. jejuni 260.94 or TSB via intra-gastric gavage using a 3.5 Fr red rubber catheter. Immediately before and after inoculation, the inocula were serially diluted in TSB, spread on TSAB plates, and the plates incubated at 37°C in the microaerobic environment for approximately 48 hours. Final calculated doses of C. jejuni 260.94 inoculum were 3.1  108 CFU for BALB/c WT mice, and 2.8  108 CFU for BALB/c IL-10-/- mice. Monitoring for Clinical Signs. Mice were checked at least once daily for clinical signs starting at inoculation. Evaluated clinical parameters included observation of eating and drinking and overall 51 activity level, and gauging abnormalities in respiratory rate, hair coat, posture (e.g., hunching), defecation including diarrhea, and level of movement including rearing. Any mouse reaching a pre- determined humane endpoint was immediately euthanized and necropsy was performed. Neurological Phenotyping. Mice underwent phenotyping tests weekly to evaluate clinical signs of neurological disease including gait abnormalities and loss of motor strength. Assessments included DigiGait treadmill analysis, hang test, and open field test (OFT). Prior to inoculation, mice were acclimated to these tests prior to the start of the experiment and baseline data taken prior to inoculation were included in the statistical analyses. DigiGait Imaging System. The DigiGait (Mouse Specifics, Inc., Quincy, MA) treadmill system records numerous gait indices to allow detection of subtle gait abnormalities. The mouse was placed upon a transparent moving belt. A digital camera underneath the belt captured movement using DigiGait Video Imaging Acquisition software (Mouse Specifics, Inc., Quincy, MA). Data were recorded over a 3-second run in which the mouse ran in the center of the belt, with no or minimal leaning against the clear plastic siding or contact with the front or back bumpers. Digital images were then processed using the DigiGait Imaging Analysis software (Mouse Specifics, Inc., Quincy, MA). Numerical data reflecting multiple gait parameters are calculated for each limb and exported to an Excel spreadsheet for statistical analysis. Of the 42 available gait parameters, eight were deemed most relevant for analysis: 13; 21; 43 swing (s), propel (s), stance (s), stride (s), stride length (cm), stride frequency (steps/s), absolute paw angle (deg), and stance width (cm). In this study, the mice were recorded running at speeds of 25 cm/s, 30 cm/s, and 35 cm/s. If needed, mice were given multiple chances to achieve a 3- second run, with rest periods between attempts. The belt was cleaned with 70% ethanol between each mouse. 52 Hang Test. The hang test served as a measure of motor strength and balance. This test was performed as described11; 54 with minor modifications. Briefly, the mouse was placed on a square metal grid elevated several inches above its cage and allowed to grip the metal. A stopwatch was started when the grid was flipped upside down. Mice were allowed to hang for up to 60 seconds, and the time elapsed until they fell off the grid into the cage was recorded. This was repeated 3 times, with the mice allowed to rest for 60 seconds between each trial. The grid was cleaned with 70% ethanol between mice. The average time to fall (or a maximum of 60 seconds if the mouse did not fall) over the 3 trials was used in the analyses. Open Field Test. The OFT allows assessment of gait, posture, and activity level. The OFT was performed by allowing the mouse to move freely within a clear plastic standard 18”  8” rat cage with 4 quadrants demarcated on the bottom. Activity of the mouse was recorded for 60 seconds, beginning with placement in the center of the cage, with a stationary video camera mounted on a tripod and set at approximately the same angle each week. At the end of each video, the mouse number was shown, with genotype and treatment group de-identified. The cage was cleaned with 70% ethanol between each mouse. Parameters intended for assessment included number of quadrants crossed, number of rears, and gait or posture abnormalities such as increased stance width or splayed toes. Necropsy Procedures. Mice were humanely euthanized by CO2 overdose 5 weeks following inoculation. Immediately thereafter, weight was recorded for each mouse and blood samples were collected via cardiac puncture using a 1 mL tuberculin syringe loaded with 0.1 mL 3.8% sodium citrate. Instruments were sterilized using a hot bead sterilizer between each section of the GI tract. The cecum was excised including approximately 0.5─1 cm of adjoining proximal colon and ileum. The tip of the cecum was removed for bacteriology studies. The remaining ileocecocolic junction (ICJ) was infused with 10% neutral phosphate buffered formalin (Fisher Scientific) and placed onto a sponge in a 53 histological tissue cassette (Fisherbrand Histosette II Tissue Cassette, Fisher HealthCare, Pittsburgh, PA) and immersed in 10% neutral phosphate buffered formalin. Sections of stomach, jejunum, proximal colon, and the tip of the cecum were excised and rinsed in PBS. Pieces of each tissue were divided into three sections: two were snap frozen in microfuge tubes and cryovials, and the third was streaked onto Campylobacter selective medium36 (TSAB-CVA plates: TSAB plates containing cefoperazone (20 µg/mL), vancomycin (10 µg/mL), and amphotericin B (2 µg/mL); antimicrobials were obtained from Sigma-Aldrich, St. Louis, MO). Plates were placed into air- tight jars with a CampyGen sachet (Oxoid, Basingstoke, Hampshire, UK). Fecal pellets were collected into microfuge tubes and placed on ice. The sciatic nerve and 2 to 3 dorsal root ganglia (DRG) were collected from experimental mice for histological and morphological analysis in a two-step procedure. At the time of necropsy, the mouse was skinned and the muscles overlying the spine were removed. The roof of the vertebral canal was removed using Castroviejo microsurgical scissors to expose the spinal cord. Muscle and connective tissue were bluntly dissected to expose the sciatic nerve on each side. Thereafter, the carcass was submerged in 10% neutral buffered formalin in a specimen cup for further dissection of tissues at a later date. On return to the laboratory, snap frozen tissues were stored at -80°C. Gastrointestinal samples streaked on TSAB-CVA plates in the air-tight jars were placed at 37°C. Fecal pellets were mashed with a sterile applicator stick in TSB containing 15% glycerol, vortexed, spread onto TSAB-CVA plates, and incubated at 37°C in jars with the microaerobic environment generated by evacuation and equilibration with the gas mixture. Plasma separated from whole blood by centrifugation was harvested and stored at -80°C until analysis. The ICJ cassettes and carcasses for nerve dissection were transferred from formalin to 60% ethanol after 24 and 48 hours, respectively. 54 Confirmation of Colonization by Culture and PCR. Colonization of stomach, jejunum, cecum, colon, and feces was reported according to a semi-quantitative grading system of plate coverage by C. jejuni colonies36 following 48─72 hours of incubation: 0 = no growth; 1 = light growth (approximately 1─20 CFU); 2 = moderate growth (20─200 CFU); 3 = heavier growth (>200 CFU); 4 = confluent growth. To confirm C. jejuni 260.94 colonization of infected mice at necropsy, and exclude colonization in sham-inoculated mice, C. jejuni-specific PCR for the C. jejuni gyrA gene was conducted.36; 59 For culture-positive mice, isolates of C. jejuni from cecal or colon samples obtained at necropsy were used. For culture-negative infected mice and sham inoculated mice, DNA extracted from frozen cecal tissues obtained at necropsy using a commercial kit (DNEasy Blood and Tissue Kit, QIAGEN, Valencia, CA) was used. Pathologic Changes: Gastrointestinal Tract. Gross lesions in the gastrointestinal tract and changes in the ileocecocolic lymph node and spleen noted during necropsy were recorded as noted by a veterinarian and other experienced personnel. The ICJs were processed routinely by the Investigative Histopathology Laboratory, Division of Human Pathology, MSU. Briefly, tissue samples were embedded in paraffin, sectioned at 4─5 µm, stained with hematoxylin and eosin (H&E), and coverslipped. The ICJs were examined histologically by a board certified veterinary clinical pathologist (JMB) blinded to the genotype and treatment groups. Lesions were graded using a previously published scoring system,36 with the exception that intraepithelial lymphocytes were not scored in the current study. The scale encompasses changes in the lumen (exudates, excessive mucus), epithelium (surface integrity, goblet cell hypertrophy or depletion, crypt abnormalities), lamina propria (inflammatory cell infiltrates), and submucosa (inflammation, edema, fibrosis). Raw scores (out of 42 total points) were ranked as semi- quantitative grades 0 (0─9 points; no colitis), 1 (10─19 points; mild colitis), or 2 (≥20; moderate or severe colitis). 55 Enzyme-Linked Immunosorbent Assay (ELISA). Aliquots of plasma were made to prevent repeated freeze-thaw cycles. Anti -C. jejuni, -GM1, and -GD1a antibody isotypes including IgG1, IgG2a, IgG2b, and IgG3 were evaluated. The assays were performed as previously described.16; 36 Briefly, 96-well plates (Nunc Maxisorp, Thermo Scientific, Rochester, NY) were coated with antigen and incubated at 4°C overnight. Antigens were diluted in PBS to the following concentrations: Campylobacter jejuni antigen36 1.9 µg/mL; GM1 antigen (US Biological, Swampscott, MA) 2 µg/mL; GD1a antigen (Sigma Aldrich, St. Louis, MO) 20 µg/mL. The plates were blocked with blocking buffer (10mM PBS with 3% BSA and 0.05% Tween-20 (Sigma)) overnight at 4°C. Following three washes in wash buffer (PBS with 0.025% Tween- 20), plasma samples diluted in blocking buffer (all samples were diluted 1:25, except for anti-C. jejuni IgG2b and IgG2a, which were 1:100) were loaded in triplicate. Positive controls included plasma from mice from previous experiments with a high OD or commercially available antibodies (anti-GD1a IgG1 (EMD Millipore, Temecula, CA)). Negative controls included anti-Toxoplasma gondii antibody (ViroStat, Portland, ME), and wells including only blocking buffer. Sealed plates were incubated with samples overnight at 4°C. Plates were washed, and secondary antibodies (Biotin-SP-conjugated AffiniPure Goat Anti-Mouse IgG1, IgG2a, IgG2b, or IgG3; Jackson ImmunoResearch, West Grove, PA) diluted in blocking buffer were added. Following incubation for 1 hour on a platform shaker, plates were washed again and ExtrAvidin peroxidase (Sigma-Aldrich) (diluted 1:2,000 in 10 mM PBS with 1% BSA and 0.05% Tween-20) was added. Plates were incubated for 1 hour on a platform shaker, washed, and tetramethylbenzidine (TMB substrate; Rockland Immunochemicals Inc., Gilbertsville, PA) was added. The reaction was stopped with 2N H2SO4. Absorbance was read at 450 nm using a Bio-Tek EL-800 Universal Microplate Reader with KC Junior software (Bio-Tek Instruments, Winooski, VT). In low numbers of samples, the CV% for triplicate values was 10% or greater. In these cases, if a clear outlier was present, that value was excluded. The absorbance generated from the diluent (blocking buffer) alone was subtracted from the mean absorbance obtained for each sample run in triplicate. This adjusted value was used in statistical 56 analyses. Negative values generated by subtracting the absorbance of the blocking buffer from the mean sample absorbance were treated as zero for the purpose of statistical analysis. Measurement of Colon Cytokine Production. Rinsed samples of proximal colon collected in Eppendorf tubes and snap frozen at necropsy were stored at -80°C until analysis. Samples were thawed on ice and wet weight was recorded. Samples were homogenized on ice for one minute in 400 µL of Hank’s Balanced Salt Solution (Sigma), with 0.5% Triton X-100 (Sigma) and the cOmplete Mini EDTA-free Protease Inhibitor cocktail (Roche/Sigma) using an autoclaved microtube pellet pestle rod powered by a handheld Kontes pellet pestle motor. Homogenates were centrifuged at 12,000g for 30 minutes at 4°C, and supernatants were aliquoted in cryovials and stored at -80°C until analysis. Cytokines were measured using a flow cytometry based multiplexed bead assay panel (LEGENDPlex Mouse Th Cytokine Panel, BioLegend, San Diego, CA). Cytokines included in the panel are designed to characterize the adaptive immune response by delineating Th polarization. Analytes included IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-6, IL-9, IL-10, IL-13, IL-17A, IL-17F, IL-21, and IL-22. Prior to analysis, aliquoted supernatants of the colon homogenates were thawed on ice and centrifuged at 300g for 10 minutes at 4°C. The assay was performed using undiluted supernatant and a V-bottom microplate according to the manufacturer’s instructions. Data were acquired on a BD FACSCanto II flow cytometer (BD Biosciences, San Jose, CA) and analyzed using the LEGENDplex Data Analysis software (BioLegend). A standard curve was generated for each analyte. Each cytokine had a maximum standard concentration of 10,000 pg/mL. Data are presented as pg cytokine/mg tissue weight. Assessment of Nerve Histopathology. Dissection of fixed tissue was performed using a dissecting microscope and nerves and dorsal root ganglia were embedded en bloc in a single cassette. Two to 3 lumbar dorsal root ganglia (DRG) were harvested from the left side of the mouse. Where possible and in 57 many cases, connections from the sciatic nerve to one or all DRG collected, including those in L3, L4, or L5 regions, were visualized prior to removal of DRG. The left sciatic nerve, brachial plexus, and lumbar DRG were placed in a cassette and stored in 60% ethanol until submission for histopathology. The sections were labeled immunohistochemically for the mouse macrophage marker F4/80 by the Investigative Histopathology Laboratory, Division of Human Pathology, MSU. Specimens were embedded in paraffin and sectioned on a rotary microtome at 4 m. Sections were placed on charged slides and dried at 56C overnight, deparaffinized in xylene, and hydrated through descending grades of ethyl alcohol to distilled water. Slides were then placed in Tris Buffered Saline pH 7.4 (TBS; Scytek Labs – Logan, UT) for 5 minutes for pH adjustment. Following TBS, epitope retrieval was performed using Citrate Plus Retrieval Solution pH 6.0 (Scytek) in a vegetable steamer for 30 minutes followed by a 10 minute countertop incubation and several changes of distilled water. Following pretreatment standard avidin-biotin complex staining steps were performed at room temperature on the DAKO Autostainer. All staining steps are followed by two minute rinses in Tris Buffered Saline + Tween 20 (Scytek). After blocking for non-specific protein with Normal Rabbit Serum (Vector Labs – Burlingame, CA) for 30 minutes, sections were incubated with Avidin / Biotin blocking system for 15 minutes each (Avidin D – Vector Labs / d-Biotin – Sigma). Primary antibody slides were incubated for 60 minutes with the Monoclonal Rat anti- Mouse F4/80 diluted @ 1:100 (AbD Serotec – Raleigh, NC) in Normal Antibody Diluent (NAD) (Scytek). Biotinylated Rabbit anti-Rat IgG (H + L) Mouse Absorbed prepared at 10.0g/mL in NAD incubated for 30 minutes, followed by R.T.U. Vector Elite Peroxidase Reagent (Vector) incubation for 30 minutes. Reaction development utilized Vector Nova Red Kit peroxidase chromogen incubation of 15 minutes followed by counterstain in Gill 2 Hematoxylin (Cancer Diagnostics – Durham, NC) for 30 seconds, differentiation, and dehydration, clearing, and mounting with Permount mounting media. The number of F4/80 positive cells was quantified morphometrically. Images were analyzed using ImageJ software (version 2.0.0rc-49/1.51d), distributed by Fiji (Fiji Is Just ImageJ) for Windows 58 (http://imagej.net/Fiji/Downloads).51; 52 The investigator (MM Cluett) was blinded to mouse genotype and treatment group. Contiguous images of each DRG section obtained at 100 magnification (10 objective) were opened in the ImageJ program. Positive cells were marked on the image using the “Cell Counter” plugin. After all positive cells were marked, the area was outlined using the “Freehand Selections” Trace Tool. When necessary, multiple areas were traced individually and the sum of the areas was recorded. Results are given as number of F4/80 positive cells/area, with area representing 100,000 pixels. The slides were finally unblinded for statistical analysis. Statistical Analyses. Analyses were performed using commercially available statistical software packages (GraphPad Prism version 6.00 for Windows, GraphPad Software, La Jolla, CA; SigmaStat 3.5 for Windows, Systat Software, Inc., Point Richmond, CA) and online statistical applications (VassarStats Website for Statistical Computation; vassarstats.net). DigiGait data generated for individual limbs were analyzed using 2-way repeated measures ANOVA, followed by Holm-Sidak’s multiple comparisons test. These analyses were conducted using SigmaStat due to missing values for those weeks when a mouse was unwilling to run on the DigiGait; SigmaStat uses a general linear model to generate a best estimate of the missing values. The mice were first grouped by combined genotype and treatment group (e.g., [WT TSB], [IL-10-/- C. jejuni 260.94]) to compare all experimental mice together. Subsequent analyses evaluated differences in all parameters between control and infected groups for WT mice alone, followed by analysis of IL-10-/- mice alone. Data were tested for normality and equal variance. Analyses of data not meeting the assumption of equal variance are excluded from results. Data from the hang test were analyzed using 2-way repeated measures ANOVA, followed by Holm-Sidak’s multiple comparisons test. The Freeman-Halton extension of the Fisher Exact Probability Test was used to assess differences in colonization identified by culture between infected BALB/c WT and infected BALB/c IL-10- 59 /- mice. For this analysis, semi-quantitative colonization grades included 0 (0 CFU), 1 (1─20 CFU), 2 (20─200 CFU) and 3 (>200 CFU; combining grades of >200 CFU and confluent growth). Gross pathology in sham-inoculated and C. jejuni 260.94-infected BALB/c IL-10-/- mice was similarly assessed using the Freeman-Halton extension of the Fisher Exact Probability Test. For this analysis, gross pathology changes recorded at necropsy including enlarged and/or thickened proximal colon, enlarged cecum, enlarged spleen, and enlarged ileocecocolic lymph node, were graded as 0 (no changes), 1 (1 change), or 2 (2 changes). Due to the lack of independence of observations comprising the composite score, ranks of colitis scores were analyzed using the Freeman-Halton extension of Fisher Exact Probability Test performed on overall group data, followed by 6 pair-wise comparisons. Kruskal-Wallis one-way ANOVA on ranks, followed by Dunn’s post-test, was also performed on raw histopathology scores. Data obtained from plasma ELISAs, colon cytokine measurement, and F4/80 scoring of DRG were analyzed using the non-parametric Kruskal-Wallis one-way ANOVA, followed by Dunn’s multiple comparisons test when overall significance was found. RESULTS Confirmation of Mouse Genotype and Absence of Enteropathogens. Genotype (WT, or IL-10-/-) was confirmed by PCR on DNA obtained from cecal tissue. At the start of the experiment, all mice were negative by PCR for all enteropathogens tested (Citrobacter rodentium, Enterococcus faecium and E. faecalis, Helicobacter spp., and Campylobacter spp.). Clinical Signs. One infected BALB/c WT mouse required early humane euthanasia at 14 days post infection (p.i.) due to a progressively worsening hunched posture, mildly decreased activity, rougher coat, thinner body condition, and mildly increased respiratory effort. Necropsy revealed a large thoracic 60 mass. This was considered unrelated to C. jejuni infection, and was not further investigated. Due to morbidity unrelated to C. jejuni infection and the span of time between euthanasia of this mouse and remaining experimental mice (approximately 3 weeks), data for this mouse are included in results of colonization but no other experimental parameters. Questionable or mildly hunched posture, decreased activity, and hair coat quality were monitored carefully and were noted in 3/10 sham-inoculated WT mice, 7/9 infected WT mice, 5/10 sham-inoculated IL-10-/- mice, and 6/10 infected IL-10-/- mice at least once during the study. Many of these signs were attributed to characteristic BALB/c posture or hair coat appearance. No other mice required early humane euthanasia, and the remaining 39 experimental mice were humanely euthanized at 5 weeks p.i. Neurological Phenotyping. Of all neurological phenotyping tests the results of the Digigait treadmill analysis were most informative. Data from the OFT were not analyzed, because the mice did not display enough spontaneous activity to provide sufficient information regarding rearing, quadrant crossing, or gait abnormalities. DigiGait. Images recorded at 25 cm/s were processed as the most mice consistently ran at this speed each week. An infected male BALB/c IL-10-/- mouse with repeated weeks of missing data was excluded from analysis. A small number of analyses failing equal variance were excluded from results: right hind limb stride and stride length in the combined analysis; left hind limb swing in the IL-10-/- only analysis; left forelimb absolute paw angle in the WT only analysis. Only parameters for which P <0.05 was found for either treatment group or time in the overall analysis are given (Table 2.1). Significant differences related to week of testing were found for multiple parameters, most frequently involving Baseline and Week 4. Significant treatment group differences were identified in the front limbs, including stance width and paw angle. When WT and IL-10-/- genotypes were analyzed together (combined analysis), infected IL-10-/- mice had wider forelimb stance width than sham-inoculated WT 61 mice. When the genotypes were analyzed separately, forelimb stance width was wider in infected than in sham-inoculated WT mice, but the difference was not significant within IL-10-/- mice. A significant effect of treatment group was also seen in the combined analysis in absolute paw angle of the right forelimb, but no pairwise comparisons were significant. Collectively, these data suggest that a relatively small proportion of gait parameters analyzed were impacted by C. jejuni infection. Hang test. No significant differences between treatment groups were found (P = 0.199). However, a significant time effect (P <0.001) was found, with significant differences seen in pairwise comparisons between baseline and weeks 2, 4, and 5 (data not shown). Colonization. At sacrifice, 9/10 infected mice from each group were culture positive in at least one area of the GI tract. Although samples from the GI tract included stomach, jejunum, cecum, colon, and feces, the cecum has been reported to be the most consistently and heavily colonized area of the GI tract36; 37 and thus results from the cecum are shown (Figure 2.1). Positive C. jejuni 260.94 culture results in all samples from infected WT BALB/c mice were as follows: 1/10 in the stomach; 7/10 in the jejunum; 8/10 in the cecum; 9/10 in the proximal colon; 9/10 in the feces. Positive C. jejuni 260.94 culture results in all samples from infected BALB/c IL-10-/- mice were as follows: 1/10 in the stomach; 0/10 in the jejunum; 9/10 in the cecum; 9/10 in the proximal colon; 8/10 in the feces. PCR for the C. jejuni gyrA gene was positive on isolates cultured from necropsy samples. All 20 TSB-inoculated mice were negative for C. jejuni by culture in all 5 areas sampled from the GI tract. The two infected but culture-negative mice and all sham inoculated animals were negative for C. jejuni by gyrA PCR on DNA isolated from frozen cecal tissue. Results of Fisher Exact Probability Test showed that there was no significant difference in colonization of the cecum in infected BALB/c WT versus infected BALB/c IL-10-/- mice (PB = 0.775). 62 Gross Pathology and Histological Assessment of Colitis. Gross pathological changes noted at necropsy included enlarged and/or thickened proximal colon or cecum and enlarged spleen or ileocecocolic lymph node. Gross pathology in WT mice was minimal, regardless of infection status. Nearly all mice exhibiting one or more gross pathological changes were BALB/c IL-10-/- mice, regardless of C. jejuni 260.94 infection status (Figure 2.2A). Nine sham-inoculated IL-10-/- mice had an enlarged or thickened proximal colon, and one of these mice also had an enlarged spleen. Similarly, 8/10 C. jejuni-infected IL-10-/- mice had an enlarged or thickened proximal colon; of these 8 mice, 3 also had an enlarged spleen, one also had an enlarged lymph node, and one also had a thickened cecal wall. As a single mouse from each of the WT treatment groups had only one gross pathological change and the remaining mice from both groups had no reported changes, gross pathology was not further analyzed in WT mice. Within BALB/c IL-10-/- mice, comparison of gross pathology between sham-inoculated and C. jejuni 260.94-infected mice was assessed by Fisher Exact Probability Test. No significant difference was found in gross pathology between infected and control IL-10-/- mice (PB = 0.087). Histological scoring of the ICJs placed all mice from all four treatment groups into grade 0 (0─9 points; no colitis) or grade 1 (10─19 points; mild colitis). As with changes noted grossly, more BALB/c IL- 10-/- mice had histologic lesions in the ICJ relative to BALB/c WT mice, regardless of C. jejuni infection status (Figure 2.2B). All 10 sham-inoculated and all 9 C. jejuni-infected WT mice had scores ≤ 9, indicating no colitis. The IL-10-/- mice inoculated with TSB showed the greatest range of raw scores (3─17) of any of the four treatment groups. Four of 10 sham-inoculated IL-10-/- mice had mild colitis (scores ranging from 10─17) while the other 6 in this group had no colitis (scores ≤ 9). The most frequent changes seen in TSB-inoculated IL-10-/- mice were subjectively mild overall, reflecting the numerical score in the mild colitis category, and included luminal exudates comprising mucus and few neutrophils, damage to single cells in the surface epithelium, goblet cell depletion, crypt hyperplasia, increased mononuclear cells in the lamina propria, and extension of inflammation into the submucosa in 3 mice. 63 Similarly, 4/10 C. jejuni-infected IL-10-/- mice had mild colitis (scores ranging from 11─16), while the other 6 in this group had no colitis (scores ≤ 9). Pathological changes seen in C. jejuni-infected IL-10-/- mice with mild colitis were similar to those seen in sham-inoculated mice, including mucus and few neutrophils in the lumen, damage to single cells in the surface epithelium, goblet cell depletion, crypt hyperplasia, increased mononuclear cells in the lamina propria, and extension of inflammation into the submucosa in one mouse. While significance was seen overall using both Fisher Exact Probability Test on ranked data (PB = 0.012) and Kruskal-Wallis on raw scores (P = 0.0265), no pairwise comparisons were significant in either analysis. Overall, these data suggest that IL-10-/- mice had mildly higher baseline levels of gross GI pathology and histologic lesions in the ICJ than WT mice, attributed to spontaneous colitis in IL-10-/- mice unrelated to C. jejuni infection. Immune Response to C. jejuni Infection. Systemic (plasma) and local (colonic) immune responses to infection were measured. Plasma Antibodies. Multiple antibody isotypes reacting with C. jejuni and gangliosides GM1 and GD1a antigens were analyzed in an indirect ELISA format. Isotypes were chosen to reflect systemic Th1 (IgG3, IgG2a, IgG2b), Th17 (IgG2b), and Th2 (IgG1) mediated responses.1; 3; 38; 57 Plasma antibody responses to C. jejuni are shown in Figure 2.3A. Infected mice mounted primarily a Th1/Th17 response to C. jejuni, as shown by statistically significant increases in plasma anti-C. jejuni IgG2a and IgG2b levels compared to control mice, within both WT and IL-10-/- genotypes. Th1-mediated IgG3 was significantly elevated in C. jejuni-infected IL-10-/- mice when compared to sham-inoculated mice of either WT or IL- 10-/- genotype. However, infection with C. jejuni did not produce significantly elevated IgG3 in WT mice compared to sham-inoculated WT mice, indicating that C. jejuni infection only stimulated a significant IgG3 response in IL-10-/- mice. Th2-mediated IgG1 responses were significantly increased in both sham- inoculated and infected IL-10-/- mice compared to sham-inoculated WT mice. Collectively, these data 64 indicate that BALB/c mice infected with C. jejuni 260.94 mounted a primary Th1/Th17 systemic response that was exacerbated by IL-10 deficiency. Because anti-ganglioside antibodies are a hallmark of GBS and immunopathogenesis of GBS is thought to involve molecular mimicry between C. jejuni LOS and peripheral nerve gangliosides, anti - GM1 and -GD1a antibodies were also measured by ELISA. In general, magnitude of response and patterns of elevation in different groups were similar for anti-GM1 antibodies (Figure 2.3B) and anti- GD1a antibodies (Figure 2.3C). Significant increases in anti-GM1 and anti-GD1a IgG2b were seen in IL-10- /- mice, with infected IL-10-/- mice having significantly higher levels compared to infected WT mice. However, anti-ganglioside IgG2b was not significantly increased in C. jejuni-infected mice compared to control mice within either genotype. Anti-GM1 IgG3 was significantly increased in sham-inoculated WT mice compared to infected and control IL-10-/- mice, and infected WT mice had significantly increased IgG3 compared to sham-inoculated IL-10-/- mice. No other significant differences were seen between groups, including in Th2-mediated IgG1 antibodies. Collectively, these data suggest that elevations in anti-ganglioside antibodies were more closely related to presence or absence of IL-10, rather than C. jejuni infection status. Colon Cytokine Production. Cytokines reflecting differentiation of Th1, Th2, Th17, Th9, Th22, and T follicular helper (Tfh) cells were measured in a panel including IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-6, IL-9, IL- 10, IL-13, IL-17A, IL-17F, IL-21, and IL-22. Cytokines not statistically analyzed included IL-5, IL-17F, and IL- 21, of which only 1/39 mice produced a detectable amount; IL-10, in which only one WT mouse and no IL-10-/- mice produced a detectable amount; and IL-4 and IL-13, of which no mouse of either genotype produced a detectable amount. No significant increases related to infection status within the WT or IL-10-/- genotype were seen in any cytokine (Figure 2.4). However, sham-inoculated IL-10-/- mice produced significantly more TNF-α and IFN-γ than either sham-inoculated or C. jejuni 260.94-infected WT mice. Interestingly, although the 65 differences were not statistically significant, C. jejuni 260.94-infected IL-10-/- mice produced less TNF-α, IFN-γ, IL-6, IL-17A, and IL-22 than did sham-inoculated IL-10-/- mice. Significance overall was seen in analysis of IL-17A (Kruskal-Wallis, P = 0.0289), although no pairwise comparisons were significant. With the exception of IL-2 and IL-9, WT mice produced less of all analyzed cytokines compared to IL-10-/- mice, regardless of C. jejuni infection status, although these differences were not significant. Collectively, these data suggest that IL-10-/- mice trended toward increased production of cytokines reflecting Th1/Th17/Th22 differentiation than did WT mice, with a more pronounced effect seen in sham- inoculated IL-10-/- mice. Nerve Histopathology. Cells positively labeled with the F4/80 macrophage marker in DRG were quantified by morphometry (Figure 2.5). C. jejuni-infected BALB/c IL-10-/- mice showed the most variation in number of F4/80+ cells, with scores ranging from 2─33, while F4/80+ scores in sham- inoculated WT mice were the most tightly clustered (range 5─18). No statistically significant differences between treatment groups were found (P = 0.290). These data suggest that, in this model, neither absence of IL-10 nor infection with C. jejuni 260.94 significantly influenced macrophage numbers in DRG after five weeks of infection. DISCUSSION The aim of the current study was to develop a mouse model of GBS subsequent to C. jejuni infection to further our understanding of the complex immunopathogenesis of this important human disease. The model aimed to exploit the reported Th2 bias of BALB/c mice that, in the absence of anti- inflammatory IL-10, were expected to mount strong Th2-mediated immunity following infection with GBS patient-derived C. jejuni 260.94 harboring a GM1a ganglioside mimic. Production of anti-ganglioside antibodies and associated nerve lesions were expected, as had been reported for C57BL/6 IL-10-/- and 66 WT, B7-2-/-, and IL-10-/- NOD mice infected with C. jejuni 260.94.35; 54 Instead, WT and IL-10-/- BALB/c mice orally inoculated with C. jejuni 260.94 mounted primarily Th1/Th17 immune responses and did not develop colitis, anti-ganglioside antibody production, an overt neurological phenotype, or nerve lesions. When considered in the context of previously published models from our group utilizing various mouse/C. jejuni strain combinations,4; 35; 36; 54 the current study further highlights the importance of both infecting C. jejuni strain characteristics (presence of ganglioside mimics, invasiveness, colitogenic potential) and host factors (genetic background, immunological biases) in determining disease outcome. We first hypothesized that Th2-mediated immune responses, including production of anti- ganglioside antibodies but without induction of colitis, would predominate following oral inoculation of WT and IL-10-/- BALB/c mice with C. jejuni 260.94. This reasoning was based upon reports in the literature of a relative Th2 bias exhibited by BALB/c mice in comparison to other mouse strains, including C57BL/6 mice, in in vivo infection models and in vitro studies,33; 39; 48 and the Th2-mediated response shown in C57BL/6 IL-10-/- mice infected with C. jejuni 260.94.35 Instead, infected WT and IL-10-/- BALB/c mice responded with significantly increased C. jejuni-specific plasma antibodies including IgG2a, IgG2b, and IgG3 antibody isotypes, but without significantly increased IgG1 production related to infection (Figure 2.3), indicating Th1/Th17-mediated class switching.1; 3; 57 This unexpected result offers deeper insight into the diverse regulatory actions of IL-10, and the interplay of infecting pathogen and immunological bias of the host based upon genetic background. While C57BL/6 and BALB/c mice reportedly exhibit Th1- and Th2-mediated immune biases, respectively, the current study and others demonstrate that characteristics of the infecting pathogen or stimulus contribute substantially to immunological outcomes. C57BL/6 and C.B-17 (BALB/c strain congenic for C57BL/6 Ig heavy chain gene segment) mice infected with Cryptococcus neoformans, a fungal lung mucosal pathogen, showed differing immune responses and ability to clear the pathogen: susceptible C57BL/6 mice exhibited increasing lung burden over time, while resistant C.B-17 mice mounted an early local Th1-mediated 67 response correlating with superior pathogen clearance.22 BALB/c mice are resistant to developing chemically induced (dextran sulfate sodium, DSS) colitis, while C57BL/6 mice are susceptible. Assessment of cytokine production following exposure to DSS indicated that BALB/c mice exhibited Th2/Th17/Treg (T regulatory cell) mediated immunity, characterized by IL-4, IL-6, IL-10, IL-17 production and higher numbers of Treg cells in the regional lymph node, while C57BL/6 mice responded with strong IFN-γ production.61 C. jejuni is a mucosal enteric pathogen, and a Th1/Th17-mediated response has been identified in most in vivo and in vitro mouse and ex vivo human C. jejuni infection models.4; 15; 16; 35; 36; 46; 54 Thus, WT BALB/c mice may be predisposed to Th2 responses in some models, but results of the current study suggest that Th1/Th17 responses are primarily induced in BALB/c mice following C. jejuni infection, even with a C. jejuni strain that previously induced Th2 responses in C57BL/6 IL-10-/- mice.35 Th1/Th17 responses were exacerbated in IL-10-/- mice, but because a similar pattern was observed in WT mice, absence of IL-10 and subsequent enhancement of Th1-mediated immunity cannot be the sole reason for this outcome. These results further highlight the complexity of GBS immunopathogenesis and the interplay of both host and pathogen characteristics in induction of immunity. Induction of anti-ganglioside antibodies is a hallmark of GBS and a partial reflection of the immune response to C. jejuni infection. Wild-type and IL-10-/- BALB/c mice infected with C. jejuni 260.94 were expected to produce antibodies to GM1 and GD1a gangliosides, as previously reported in other mouse genotypes.35; 54 In the current study, Th1/Th17-mediated IgG3 and IgG2b isotypes reacting with GM1 and GD1a antigens showed significant elevations related only to presence or absence of IL-10 (Figure 2.3), and not to C. jejuni infection status with a strain known to elicit anti-ganglioside antibodies. The primary Th1/Th17 response to C. jejuni 260.94 in this study, instead of Th2-mediated immunity reported previously with C57BL/6 IL-10-/- mice,35 apparently does not preclude generation of anti- ganglioside antibodies. In a separate study, BALB/c IL-10-/- mice infected with the enteritis-associated C. jejuni 11168 developed marked local and systemic Th1/Th17-mediated immunity with severe colitis, and 68 produced Th1/Th17-associated anti -GM1 and -GD1a antibodies related to C. jejuni infection (Brudvig et al., unpublished (Chapter 3)). Thus, the possibility that severe colitis or strong local colonic adaptive immune responses are required for generation of anti-ganglioside antibodies in some models cannot be excluded. It is also possible that in the current study, IL-10-mediated host-microbiota interactions influenced anti-ganglioside antibody production in response to ganglioside mimics expressed by commensal flora, but further studies would be required to establish such a connection. An interesting observation in the current study was the similar patterns of anti -GM1 and -GD1a antibody production. C. jejuni 260.94 reportedly harbors a GM1a but apparently no other ganglioside mimics.34 Closely similar production of anti -GM1 and -GD1a antibodies was noted in another study involving C. jejuni 11168 and 260.94 strains (Brudvig et al., unpublished (Chapter 3)), neither of which reportedly possess GD1a mimics.34 GM1 and GD1a gangliosides are extremely similar in structure,2 and the possibility that the plasma antibodies directed at GM1 also bound GD1a antigen in the ELISA format should be considered. Additionally, the C. jejuni strain 81-176 LOS possesses structures mimicking several gangliosides, and phase variation can result in changing expression of ganglioside mimics in both C. jejuni 81-176 and 11168 strains.18; 31 Therefore, the possibility that commensal flora or C. jejuni 260.94 can express different or multiple ganglioside mimics in vivo, resulting in generation of more than one type of anti-ganglioside antibody, warrants further investigation. Despite persistent colonization of 90% of infected mice at the end of the 5-week study, neither gross pathology nor histologic changes in the ICJ were significantly increased in either WT or IL-10-/- mice compared to sham-inoculated mice (Figure 2.2). A higher baseline of overall mild colitis in IL-10-/- mice regardless of C. jejuni infection status likely reflects the development of spontaneous colitis in IL-10-/- mice due to unchecked responses to the intestinal microbiota. Uncontrolled immune responses to enteric antigens may also in part explain the presence of C. jejuni-specific plasma IgG1 and IgG2a antibodies in some sham-inoculated IL-10-/- mice (Figure 2.3). All sham-inoculated mice were negative 69 for C. jejuni by both culture and PCR, and the C. jejuni-specific plasma response may reflect generation of antibodies against a structurally similar antigen in the microbiota. Colitis can occur in IL-10-/- mice as early as 3─4 weeks of age,6; 28 and mice in the current study were inoculated at 14─15 weeks and humanely euthanized 5 weeks p.i. Therefore, the mild pathology observed in IL-10-/- mice in the current study is attributable to spontaneous disease, rather than C. jejuni infection. The relatively later age at inoculation compared with other studies was to allow the mice to increase in body size prior to intra- gastric gavage. Results of the current study are consistent with other mouse models demonstrating lack of colitis despite high colonization rates following infection with C. jejuni 260.94.4; 9; 35; 54 C. jejuni strains vary widely in ability to colonize and cause colitis, with pathogenicity related to genomic content of certain open reading frames,4 and even in IL-10-/- mice C. jejuni 260.94 has not been colitogenic in our models. C. jejuni-specific plasma antibody isotypes reflected a systemic Th1/Th17 response in infected mice, but a significant local adaptive immune response was not identified in the proximal colon (Figure 2.4). Consistent with a shift toward Th1-mediated immunity in the absence of IL-10 and infection with a mucosal enteric pathogen, production of Th2 cytokines including IL-4, IL-5, and IL-13 was virtually undetectable. Significant increases were only identified in TNF-α and IFN-γ production in sham- inoculated IL-10-/- mice compared to WT mice, reflecting the Th1-mediated spontaneous colitis occurring in IL-10-/- mice. No significant difference in production of any cytokine by either WT or IL-10-/- mice related to C. jejuni infection was found. Interestingly, production of TNF-α, IFN-γ, IL-6, IL-17A, and IL-22 all tended to decrease in C. jejuni-infected compared to sham-inoculated IL-10-/- mice, suggesting that C. jejuni infection led to a relative dampening of local immune responses. The reason for this is not clear, but an alteration in the local microbiota or cytokine milieu induced by persistent C. jejuni colonization may have altered cytokine production by Th cells. Reduced intracellular survival and/or invasion efficiency of C. jejuni 260.94 compared to other C. jejuni strains has been shown by 70 immunohistochemistry in the ileocecocolic junction of infected BALB/c IL-10-/- mice (Brudvig et al., unpublished (Chapter 3)) and in vitro using murine bone marrow-derived dendritic cells (Brudvig et al., unpublished (Chapter 4)) and young adult mouse colon cells.35 Persistent but relatively superficial colonization of C. jejuni 260.94, with heavier burdens in the mucus layer and epithelium as opposed to deeper in the lamina propria and submucosa, could be especially conducive to changes in the local microbiota and lead to altered Th responses. Further studies would be needed to confirm and determine the significance of decreased cytokine production in infected compared to sham-inoculated IL-10-/- mice. The second aim of this study was to determine if C. jejuni 260.94 infection leads to neuropathology in BALB/c mice, manifested by gait abnormalities, neuromuscular weakness, and macrophage infiltration into DRG. Three rigorous phenotyping tests were performed prior to inoculation and weekly thereafter until sacrifice at 5 weeks p.i. to determine neurological deficits. Of these, the DigiGait treadmill analyses were most informative (Table 2.1). Two-way repeated measures ANOVA revealed numerous significant differences in variables with time, most of which including baseline compared to 4 weeks p.i. The only variable exhibiting significant differences related to infection status with all treatment groups analyzed together was front limb stance width, with C. jejuni-infected IL-10-/- mice having a wider stance width than sham-inoculated WT mice. The difference in stance width between infected and sham-inoculated mice remained significant when WT mice were analyzed separately, but was non-significant within IL-10-/- mice. The wider stance width could indirectly reflect hind limb weakness with more body weight being front-loaded to compensate, leading to wider stance width in the front. The DigiGait system is designed to detect subtle gait changes, but this single change is more difficult to interpret without concurrent gait disturbances to clarify its significance. Gait abnormalities also may have been missed when a run of sufficient quality could not be obtained for that week: at least 50% of the mice in each of the four treatment groups was missing at least one week of 71 DigiGait data. Similar difficulty in obtaining quality data also occurred with some NOD IL-10-/- mice in a separate study by our group.54 We can therefore conclude that, with the wider stance width related to C. jejuni infection, mild neurological abnormalities may have been present and other gait abnormalities might have been missed due to inability to acquire a quality 3-second run. The hang test and OFT test were not informative in the current study. The hang test was employed as a measure of motor strength and balance. Significant differences determined by two-way repeated measures ANOVA were related to time but not infection status. Subjective observations suggest that in the current study, this test was likely not adequate for detecting neurological deficits. More active mice tended to move around more on the grid, leading to a shorter hang time that was not related to weakness. In contrast, mice less interested in exploring the grid tended to hook their feet through the bars and hang on until the end of the test periods. These observations suggest that for BALB/c mice, the hang test performed with this technique did not reflect motor deficits but instead activity level, and was thus not a useful neurological indicator. Similarly, recordings of the OFT in which mice were placed in a standard rat cage in order to observe the gait, movement, and posture, were not analyzed because the mice did not consistently exhibit enough spontaneous movement to allow meaningful observation. Reduced movement and rearing by BALB/c mice compared to other mouse strains in the OFT has been reported previously,11; 29 and may be related to photophobia in albinos such as BALB/c mice.12 Despite the disadvantages of the OFT and hang test, the three neurological phenotyping tests combined with careful daily observation would have detected an overt neurological phenotype if present. Taken together, we conclude that infection with GBS patient-derived C. jejuni 260.94 did not lead to obvious or severe neurological deficits in the current study, though mild manifestations may have been missed. Neuropathology was assessed by evaluation of macrophage numbers in lumbar DRG. Macrophages were chosen as indicators of neuropathology due to their postulated role in GBS 72 pathogenesis, particularly in acute motor axonal form of GBS considered by some to be the subtype most closely associated with preceding C. jejuni infection.14; 41; 55 DRG were chosen for assessment as pain and sensory defects are frequently described by GBS patients.55 Cell bodies of sensory fibers reside in DRG, which are also reported to have a particularly leaky blood-nerve barrier predisposing this area to immune-mediated damage.41 Pathology in dorsal roots has been described following autopsy of patients succumbing to the motor-sensory axonal form of GBS,17 and increased F4/80 positive cells were seen in the DRG, but not sciatic nerve or brachial plexus, of NOD IL-10-/- mice infected with C. jejuni 260.94.54 In the current study, no significant difference in F4/80 positive cells in lumbar DRG was seen between any treatment groups (Figure 2.5), despite significant differences in anti-ganglioside GM1 and GD1a IgG2b and IgG3 isotypes in the plasma (Figure 2.3). A recent study also demonstrated a lack of correlation between significantly increased anti-GM1 IgG1 in C. jejuni 260.94 and 11168-infected mice with a human-derived microbiota and F4/80 positive cells in sciatic nerves and DRG.9 The immunopathogenesis of nerve damage in GBS is incompletely understood. The lack of increased macrophages in DRG despite significantly increased plasma anti-ganglioside antibodies in some groups suggests that anti-ganglioside antibodies alone are insufficient to cause nerve lesions, the timing of higher anti-GM1 or anti-GD1a antibody levels may not coincide with cellular infiltration in the DRG, or the damage may be occurring in a different location in the peripheral nervous system. In addition to macrophages, labeling of other immune components implicated in nerve damage in GBS, such as T cells, complement, and IgG41; 55 in peripheral nerve tissues and conducting a time-course study to determine if and when neuropathology is correlated with anti-ganglioside antibody production would provide a more complete assessment of GBS immunopathogenesis. Taken together, absence of neurological deficits identified by rigorous phenotyping tests, including DigiGait treadmill analysis, OFT, and hang testing, and the lack of increased macrophages in DRG indicate that neuropathology was not induced by C. jejuni 260.94 in BALB/c WT or IL-10-/- mice in the current model. 73 The final aim of this study was to determine if the absence of the regulatory actions of IL-10 resulted in exacerbation of immune responses and neuropathology. The systemic Th1/Th17 immune response to C. jejuni 260.94, manifested by significant increases in C. jejuni-specific IgG2a, IgG2b, and IgG3 antibody isotypes, was increased in magnitude in IL-10-/- compared to WT mice (Figure 2.3). Furthermore, C. jejuni induced significant IgG3 production only in the IL-10-/- group. Thus, absence of IL- 10 enhanced systemic Th1/Th17 immunity in C. jejuni infected mice. When DigiGait parameters for WT and IL-10-/- mice were analyzed separately, a significantly wider forelimb stance width was associated with C. jejuni infection in WT but not IL-10-/- mice, suggesting that IL-10 deficiency was protective rather than conducive to development of C. jejuni-induced gait defects. Presence or absence of IL-10 led to differences in anti-ganglioside antibody production, and absence of IL-10 induced mild colitis and increased cytokine production in the colon. However, aside from the increased C. jejuni-specific plasma antibodies in IL-10-/- relative to WT mice, IL-10 deficiency did not lead to significant exacerbation of C. jejuni-induced changes in immunity or neuropathology. There was no significant difference in cecal C. jejuni colonization between WT and IL-10-/- mice. An equal number of sham-inoculated and infected IL-10-/- mice developed mild colitis with similar histologic lesions, indicating that IL-10-/- mice had underlying spontaneous colitis that was not exacerbated by C. jejuni infection. Similarly, production of TNF-α and IFN-γ in the proximal colon was increased overall in IL-10-/- mice but no significant differences related to C. jejuni infection were seen. No significant difference in macrophage numbers in lumbar DRG was found between treatment groups. Thus, in this model, synergism between C. jejuni infection and IL-10 deficiency resulting in exacerbated immunity or pathology was not definitively observed. Our group has previously shown that C57BL/6 WT mice stably colonize with C. jejuni 11168 while IL-10-/- mice develop severe colitis, emphasizing the importance of IL- 10 in regulating intestinal inflammatory responses.36; 37 Further, even with IL-10 deficiency, conducive to impaired Th2- and enhanced Th1-mediated responses, C57BL/6 IL-10-/- mice infected with C. jejuni 74 260.94 still produced Th2-mediated anti -GM1 and -GD1a IgG1 antibodies,35 indicating that a lack of IL- 10 does not always preclude a Th2-mediated response. The lack of susceptibility of C. jejuni 260.94- infected BALB/c WT or IL-10-/- mice to developing Th2-driven autoimmunity and the lack of exacerbation of pathology in IL-10-/- mice in the current study further emphasize the importance of both C. jejuni strain characteristics and host genetic background in development of pathology and autoimmunity following C. jejuni infection. Results of this study indicate that WT and IL-10-/- BALB/c mice orally inoculated with GBS patient-derived C. jejuni 260.94 were stably colonized and mounted systemic Th1/Th17 responses, but did not develop colitis, produce anti-ganglioside antibodies, or develop neuropathology, including overt neurological deficits and inflammatory lesions in DRG. Systemic Th1/Th17-mediated immunity was unexpected, considering the reported Th2 immunological bias in BALB/c mice and the previous induction of Th2-mediated immunity in C57BL/6 IL-10-/- mice infected with C. jejuni 260.94. These results indicate BALB/c mice are a useful model for studying C. jejuni infection and could provide critical insights into the impact of host genetic background in polarization of the immune response and resulting pathology. 75 APPENDIX 76 Table 2.1. Results of DigiGait analysis. Gait parameters including swing, propel, stance, stride, stride length, stride frequency, absolute paw angle, and stance width were analyzed in BALB/c wild-type (WT) and IL-10-/- mice given either C. jejuni 260.94 or tryptic soy broth (sham; TSB). Gait parameters were analyzed within individual feet. Analyses included baseline (prior to inoculation) and Weeks 1-5 post infection. Two-way repeated measures ANOVA was performed on all groups together and also within individual genotypes. Where overall significance was found, Holm-Sidak post-testing was performed. P values <0.05 are reported for the factor (Week or Treatment Group) in the overall analysis, with significant pairwise comparison(s) listed. Significance was not found in any other analyses of gait parameters. Results are not shown in 4/90 total analyses, in which the assumption of equal variance was not met. LH = left hind; RH = right hind; LF = left front; RF = right front. IL-10-/- = BALB/c IL-10-/-; WT = BALB/c wild-type. 260.94 = C. jejuni 260.94; TSB = tryptic soy broth (sham). ns = nonsignificant. Gait Parameter LH Propel LH Stance RH Propel RH Stance RH Absolute Paw Angle LF Absolute Paw Angle LF Stance Width RF Absolute Paw Angle RF Absolute Paw Angle Gait Parameter RH Absolute Paw Angle LF Stance Width Gait Parameter LH Propel LH Stance RH Propel LF Absolute Paw Angle RF Absolute Paw Angle ANALYSIS INCLUDING BOTH GENOTYPES P value 0.011 0.049 <0.001 0.017 0.042 0.017 0.003 0.031 0.010 Significant Comparison Week: Baseline v. Week 4 Week: pairwise comparisons ns Week: Baseline v. Weeks 2, 3, 4, and 5 and Weeks 1 v. 3 Week: Baseline v. Week 4 Week: pairwise comparisons ns Week: Baseline v. Week 4 Treatment Group: IL-10-/- 260.94 v. WT TSB Treatment Group: pairwise comparisons ns Week: Baseline v. Weeks 4 and 5 ANALYSIS WITHIN BALB/C WT MICE Significant Comparison Week: Weeks 1 v. 3 Treatment Group: 260.94 v. TSB P value 0.031 0.034 ANALYSIS WITHIN BALB/C IL-10-/- MICE P value 0.017 0.035 <0.001 0.021 0.004 Significant Comparison Week: Baseline v. Weeks 2 and 4 Week: Baseline v. Week 4 Week: Baseline v. Weeks 3, 4, and 5 Week: Baseline v. Week 4 Week: Baseline v. Weeks 3, 4 and 5 77 Figure 2.1. Culture results of Campylobacter jejuni 260.94 colonization of the cecum at the time of necropsy. Wild-type (WT) or interleukin (IL)-10 knockout (IL-10-/-) BALB/c mice were inoculated with either C. jejuni 260.94 or vehicle (tryptic soy broth; TSB). Colonization rates were semi-quantitatively graded by approximate number of colony forming units (CFU) on the plate. With the exception of one infected WT mouse requiring early sacrifice unrelated to C. jejuni infection, mice were sacrificed 5 weeks post infection. Colonization in the cecum was not significantly different between infected WT and IL-10-/- mice (Fisher Exact Probability Test, PB = 0.775). 78 Figure 2.2. Gross pathology and ileocecocolic junction histopathology. Gross pathological changes were noted at necropsy (A) and histopathologic assessment of the ileocecocolic junction (B) was performed for 19 BALB/c wild-type (WT) and 20 BALB/c interleukin-10-deficient (BALB/c IL-10-/-) mice receiving either Campylobacter jejuni 260.94 or vehicle control (tryptic soy broth; TSB). Mice were humanely sacrificed 5 weeks post infection. One infected WT mouse required early sacrifice unrelated to C. jejuni, and is excluded from these data. (A) Gross lesions noted at necropsy included enlarged or thickened wall of proximal colon (present in all 17 IL-10-/- mice with at least one change); other more sporadic changes included enlarged or thickened cecum, enlarged spleen, and enlarged ileocecocolic lymph nodes. Gross pathology scores between sham-inoculated and C. jejuni 260.94-infected IL-10-/- mice were not significantly different (Fisher Exact Probability Test, PB = 0.087). (B) Histologic assessment of colitis in the ileocecocolic junction involved assigning a raw score out of 42 possible points, and subsequent ranking of raw scores for statistical purposes. Whiskers span minimum to maximum values, the box extends from 25th – 75th percentiles, and the line in the middle of the box denotes the median. Mice with scores ≤ 9 points are considered to have no colitis, while scores of 10—19 indicate mild colitis. Significance was seen overall using both Fisher Exact Probability Test on ranked data (PB = 0.012) and Kruskal-Wallis on raw scores (P = 0.0265), but no pairwise comparisons were significant in either analysis. 79 Figure 2.3. Assessment of plasma anti-C. jejuni and anti-ganglioside IgG isotypes. Quantification of plasma IgG1, IgG2a, IgG2b, and IgG3 specific for C. jejuni (panel A) and gangliosides GM1 (panel B) and GD1a (panel C) was performed by indirect ELISA. BALB/c wild-type (WT) and interleukin (IL)-10-/- mice were inoculated with either C. jejuni 260.94 or vehicle (tryptic soy broth, TSB), and sacrificed 5 weeks post infection. One infected WT mouse required early sacrifice unrelated to C. jejuni, and is excluded from these data. P values <0.05 were considered significant. Lines above indicate significant differences between groups (Kruskal-Wallis, followed by Dunn’s post-test). Mean ± SEM. 80 Figure 2.4. Th cytokine production in the proximal colon. Cytokine production, presented as picogram of cytokine per milligram of colon tissue, was measured in supernatants of homogenized colon samples using a multiplexed, flow cytometry-based bead array. Treatment groups included wild-type (WT) and interleukin (IL)-10 deficient (IL-10-/-) BALB/c mice, either sham inoculated (tryptic soy broth, TSB) or orally infected with C. jejuni 260.94. Mice were sacrificed 5 weeks post infection. One infected WT mouse requiring early sacrifice unrelated to C. jejuni was excluded. P values <0.05 were considered significant. Lines above groups indicate significant differences (Kruskal-Wallis ANOVA on ranks, followed by Dunn’s post-test). Overall significance was found in analysis of IL-17A (P = 0.0289), although no pairwise comparisons were significant. Mean ± SEM. 81 Figure 2.5. Quantification of macrophages in dorsal root ganglia. Lumbar dorsal root ganglia (DRG) were dissected from 19 BALB/c wild-type (WT) and 20 BALB/c interleukin (IL)-10-/- mice given either C. jejuni 260.94 or TSB (sham; tryptic soy broth). Mice were sacrificed at 5 weeks post infection. One infected WT mouse required early sacrifice unrelated to C. jejuni, and was excluded from these data. Macrophages identified by immunohistochemical labeling with F4/80 were quantitatively scored using morphometry in Image J. Results are given as number of F4/80 positive cells per unit area (100,000 pixels). No significant differences between groups were seen (Kruskal-Wallis, P = 0.290). Mean ± SEM. 82 REFERENCES 83 REFERENCES Abbas AK, Murphy KM, Sher A. 1996. Functional diversity of helper T lymphocytes. Nature 383:787-793. Ang CW, Laman JD, Willison HJ, Wagner ER, Endtz HP, De Klerk MA, Tio-Gillen AP, Van den Braak N, Jacobs BC, Van Doorn PA. 2002. Structure of Campylobacter jejuni lipopolysaccharides determines antiganglioside specificity and clinical features of Guillain-Barre and Miller Fisher patients. Infection and immunity 70:1202-1208. Bai Y, Liu R, Huang D, La Cava A, Tang YY, Iwakura Y, Campagnolo DI, Vollmer TL, Ransohoff RM, Shi FD. 2008. CCL2 recruitment of IL-6-producing CD11b+ monocytes to the draining lymph nodes during the initiation of Th17-dependent B cell-mediated autoimmunity. European journal of immunology 38:1877-1888. Bell JA, Jerome JP, Plovanich-Jones AE, Smith EJ, Gettings JR, Kim HY, Landgraf JR, Lefebure T, Kopper JJ, Rathinam VA, St Charles JL, Buffa BA, Brooks AP, Poe SA, Eaton KA, Stanhope MJ, Mansfield LS. 2013. Outcome of infection of C57BL/6 IL-10(-/-) mice with Campylobacter jejuni strains is correlated with genome content of open reading frames up- and down-regulated in vivo. Microbial pathogenesis 54:1-19. Bereswill S, Fischer A, Plickert R, Haag LM, Otto B, Kuhl AA, Dasti JI, Zautner AE, Munoz M, Loddenkemper C, Gross U, Gobel UB, Heimesaat MM. 2011. Novel murine infection models provide deep insights into the "menage a trois" of Campylobacter jejuni, microbiota and host innate immunity. PloS one 6:e20953. Berg DJ, Davidson N, Kuhn R, Muller W, Menon S, Holland G, Thompson-Snipes L, Leach MW, Rennick D. 1996. Enterocolitis and colon cancer in interleukin-10-deficient mice are associated with aberrant cytokine production and CD4(+) TH1-like responses. The Journal of clinical investigation 98:1010-1020. Berndtson E, Danielsson-Tham ML, Engvall A. 1994. Experimental colonization of mice with Campylobacter jejuni. Veterinary microbiology 41:183-188. Blaser MJ, Duncan DJ, Warren GH, Wang WL. 1983. Experimental Campylobacter jejuni infection of adult mice. Infection and immunity 39:908-916. Brooks PT, Brakel KA, Bell JA, Bejcek CE, Gilpin T, Brudvig JM, Mansfield LS. 2017. Transplanted human fecal microbiota enhanced Guillain Barre syndrome autoantibody responses after Campylobacter jejuni infection in C57BL/6 mice. Microbiome 5:92. Chang C, Miller JF. 2006. Campylobacter jejuni colonization of mice with limited enteric flora. Infection and immunity 74:5261-5271. Crawley JN. 2007. What's Wrong With My Mouse? Behavioral Phenotyping of Transgenic and Knockout Mice, 2nd ed. Hoboken, N.J., U.S.A.: John Wiley & Sons, Inc. 84 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. DeFries JC, Hegmann JP, Weir MW. 1966. Open-field behavior in mice: evidence for a major gene effect mediated by the visual system. Science 154:1577-1579. Dorman CW, Krug HE, Frizelle SP, Funkenbusch S, Mahowald ML. 2014. A comparison of DigiGait and TreadScan imaging systems: assessment of pain using gait analysis in murine monoarthritis. J Pain Res 7:25-35. Drenthen J, Yuki N, Meulstee J, Maathuis EM, van Doorn PA, Visser GH, Blok JH, Jacobs BC. 2011. Guillain-Barre syndrome subtypes related to Campylobacter infection. Journal of neurology, neurosurgery, and psychiatry 82:300-305. Edwards LA, Nistala K, Mills DC, Stephenson HN, Zilbauer M, Wren BW, Dorrell N, Lindley KJ, Wedderburn LR, Bajaj-Elliott M. 2010. Delineation of the innate and adaptive T-cell immune outcome in the human host in response to Campylobacter jejuni infection. PloS one 5:e15398. Fox JG, Rogers AB, Whary MT, Ge Z, Taylor NS, Xu S, Horwitz BH, Erdman SE. 2004. Gastroenteritis in NF-kappaB-deficient mice is produced with wild-type Camplyobacter jejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent colonization with both strains. Infection and immunity 72:1116-1125. Griffin JW, Li CY, Ho TW, Tian M, Gao CY, Xue P, Mishu B, Cornblath DR, Macko C, McKhann GM, Asbury AK. 1996. Pathology of the motor-sensory axonal Guillain-Barre syndrome. Annals of neurology 39:17-28. Guerry P, Szymanski CM, Prendergast MM, Hickey TE, Ewing CP, Pattarini DL, Moran AP. 2002. Phase variation of Campylobacter jejuni 81-176 lipooligosaccharide affects ganglioside mimicry and invasiveness in vitro. Infection and immunity 70:787-793. Hadden RD, Karch H, Hartung HP, Zielasek J, Weissbrich B, Schubert J, Weishaupt A, Cornblath DR, Swan AV, Hughes RA, Toyka KV, Plasma Exchange/Sandoglobulin Guillain-Barre Syndrome Trial G. 2001. Preceding infections, immune factors, and outcome in Guillain-Barre syndrome. Neurology 56:758-765. Halstead SK, Zitman FM, Humphreys PD, Greenshields K, Verschuuren JJ, Jacobs BC, Rother RP, Plomp JJ, Willison HJ. 2008. Eculizumab prevents anti-ganglioside antibody-mediated neuropathy in a murine model. Brain : a journal of neurology 131:1197-1208. Hansen ST, Pulst SM. 2013. Response to ethanol induced ataxia between C57BL/6J and 129X1/SvJ mouse strains using a treadmill based assay. Pharmacology, biochemistry, and behavior 103:582-588. Hoag KA, Street NE, Huffnagle GB, Lipscomb MF. 1995. Early cytokine production in pulmonary Cryptococcus neoformans infections distinguishes susceptible and resistant mice. Am J Respir Cell Mol Biol 13:487-495. 85 Jacobs BC, Hazenberg MP, van Doorn PA, Endtz HP, van der Meche FG. 1997. Cross-reactive antibodies against gangliosides and Campylobacter jejuni lipopolysaccharides in patients with Guillain-Barre or Miller Fisher syndrome. The Journal of infectious diseases 175:729-733. Jacobs BC, Koga M, van Rijs W, Geleijns K, van Doorn PA, Willison HJ, Yuki N. 2008. Subclass IgG to motor gangliosides related to infection and clinical course in Guillain-Barre syndrome. Journal of neuroimmunology 194:181-190. Jacobs BC, Rothbarth PH, van der Meche FG, Herbrink P, Schmitz PI, de Klerk MA, van Doorn PA. 1998. The spectrum of antecedent infections in Guillain-Barre syndrome: a case-control study. Neurology 51:1110-1115. Keithlin J, Sargeant J, Thomas MK, Fazil A. 2014. Systematic review and meta-analysis of the proportion of Campylobacter cases that develop chronic sequelae. BMC public health 14:1203. Koga M, Yuki N, Hirata K, Morimatsu M, Mori M, Kuwabara S. 2003. Anti-GM1 antibody IgG subclass: a clinical recovery predictor in Guillain-Barre syndrome. Neurology 60:1514-1518. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. 1993. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75:263-274. Lalonde R, Strazielle C. 2008. Relations between open-field, elevated plus-maze, and emergence tests as displayed by C57/BL6J and BALB/c mice. J Neurosci Methods 171:48-52. Li CY, Xue P, Tian WQ, Liu RC, Yang C. 1996. Experimental Campylobacter jejuni infection in the chicken: an animal model of axonal Guillain-Barre syndrome. Journal of neurology, neurosurgery, and psychiatry 61:279-284. Linton D, Gilbert M, Hitchen PG, Dell A, Morris HR, Wakarchuk WW, Gregson NA, Wren BW. 2000. Phase variation of a beta-1,3 galactosyltransferase involved in generation of the ganglioside GM1-like lipo-oligosaccharide of Campylobacter jejuni. Molecular microbiology 37:501-514. Liu T, Matsuguchi T, Tsuboi N, Yajima T, Yoshikai Y. 2002. Differences in expression of toll-like receptors and their reactivities in dendritic cells in BALB/c and C57BL/6 mice. Infection and immunity 70:6638-6645. Liu T, Nishimura H, Matsuguchi T, Yoshikai Y. 2000. Differences in interleukin-12 and -15 production by dendritic cells at the early stage of Listeria monocytogenes infection between BALB/c and C57 BL/6 mice. Cellular immunology 202:31-40. Louwen R, Horst-Kreft D, de Boer AG, van der Graaf L, de Knegt G, Hamersma M, Heikema AP, Timms AR, Jacobs BC, Wagenaar JA, Endtz HP, van der Oost J, Wells JM, Nieuwenhuis EE, van Vliet AH, Willemsen PT, van Baarlen P, van Belkum A. 2013. A novel link between Campylobacter jejuni bacteriophage defence, virulence and Guillain-Barre syndrome. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 32:207-226. 86 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Malik A, Sharma D, St Charles J, Dybas LA, Mansfield LS. 2014. Contrasting immune responses mediate Campylobacter jejuni-induced colitis and autoimmunity. Mucosal immunology 7:802- 817. Mansfield LS, Bell JA, Wilson DL, Murphy AJ, Elsheikha HM, Rathinam VA, Fierro BR, Linz JE, Young VB. 2007. C57BL/6 and congenic interleukin-10-deficient mice can serve as models of Campylobacter jejuni colonization and enteritis. Infection and immunity 75:1099-1115. Mansfield LS, Patterson JS, Fierro BR, Murphy AJ, Rathinam VA, Kopper JJ, Barbu NI, Onifade TJ, Bell JA. 2008. Genetic background of IL-10(-/-) mice alters host-pathogen interactions with Campylobacter jejuni and influences disease phenotype. Microbial pathogenesis 45:241-257. Martin RM, Brady JL, Lew AM. 1998. The need for IgG2c specific antiserum when isotyping antibodies from C57BL/6 and NOD mice. Journal of immunological methods 212:187-192. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. 2000. M-1/M-2 macrophages and the Th1/Th2 paradigm. Journal of immunology 164:6166-6173. Moran AP, Annuk H, Prendergast MM. 2005. Antibodies induced by ganglioside-mimicking Campylobacter jejuni lipooligosaccharides recognise epitopes at the nodes of Ranvier. Journal of neuroimmunology 165:179-185. Nachamkin I, Allos BM, Ho T. 1998. Campylobacter species and Guillain-Barre syndrome. Clinical microbiology reviews 11:555-567. Nyati KK, Prasad KN, Kharwar NK, Soni P, Husain N, Agrawal V, Jain AK. 2012. Immunopathology and Th1/Th2 immune response of Campylobacter jejuni-induced paralysis resembling Guillain-Barre syndrome in chicken. Medical microbiology and immunology 201:177- 187. Piesla MJ, Leventhal L, Strassle BW, Harrison JE, Cummons TA, Lu P, Whiteside GT. 2009. Abnormal gait, due to inflammation but not nerve injury, reflects enhanced nociception in preclinical pain models. Brain research 1295:89-98. Poropatich KO, Walker CL, Black RE. 2010. Quantifying the association between Campylobacter infection and Guillain-Barre syndrome: a systematic review. Journal of health, population, and nutrition 28:545-552. Prendergast MM, Moran AP. 2000. Lipopolysaccharides in the development of the Guillain- Barre syndrome and Miller Fisher syndrome forms of acute inflammatory peripheral neuropathies. Journal of endotoxin research 6:341-359. Rathinam VA, Hoag KA, Mansfield LS. 2008. Dendritic cells from C57BL/6 mice undergo activation and induce Th1-effector cell responses against Campylobacter jejuni. Microbes and infection 10:1316-1324. 87 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. Rees JH, Gregson NA, Hughes RA. 1995. Anti-ganglioside GM1 antibodies in Guillain-Barre syndrome and their relationship to Campylobacter jejuni infection. Annals of neurology 38:809- 816. Reiner SL, Locksley RM. 1995. The regulation of immunity to Leishmania major. Annual review of immunology 13:151-177. Sabat R, Grutz G, Warszawska K, Kirsch S, Witte E, Wolk K, Geginat J. 2010. Biology of interleukin-10. Cytokine & growth factor reviews 21:331-344. Salomon B, Rhee L, Bour-Jordan H, Hsin H, Montag A, Soliven B, Arcella J, Girvin AM, Padilla J, Miller SD, Bluestone JA. 2001. Development of spontaneous autoimmune peripheral polyneuropathy in B7-2-deficient NOD mice. The Journal of experimental medicine 194:677-684. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. 2012. Fiji: an open-source platform for biological-image analysis. Nature methods 9:676-682. Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years of image analysis. Nature methods 9:671-675. Sheikh KA, Nachamkin I, Ho TW, Willison HJ, Veitch J, Ung H, Nicholson M, Li CY, Wu HS, Shen BQ, Cornblath DR, Asbury AK, McKhann GM, Griffin JW. 1998. Campylobacter jejuni lipopolysaccharides in Guillain-Barre syndrome: molecular mimicry and host susceptibility. Neurology 51:371-378. St Charles JL, Bell JA, Gadsden BJ, Malik A, Cooke H, Van de Grift LK, Kim HY, Smith EJ, Mansfield LS. 2017. Guillain Barre Syndrome is induced in Non-Obese Diabetic (NOD) mice following Campylobacter jejuni infection and is exacerbated by antibiotics. Journal of autoimmunity 77:11-38. van den Berg B, Walgaard C, Drenthen J, Fokke C, Jacobs BC, van Doorn PA. 2014. Guillain- Barre syndrome: pathogenesis, diagnosis, treatment and prognosis. Nature reviews Neurology 10:469-482. 56. Watanabe H, Numata K, Ito T, Takagi K, Matsukawa A. 2004. Innate immune response in Th1- and Th2-dominant mouse strains. Shock 22:460-466. 57. Weaver CT, Harrington LE, Mangan PR, Gavrieli M, Murphy KM. 2006. Th17: an effector CD4 T cell lineage with regulatory T cell ties. Immunity 24:677-688. 58. WHO. [Internet]. 2016. Campylobacter Fact Sheet. Available at: http://www.who.int/mediacentre/factsheets/fs255/en/. 59. Wilson DL, Abner SR, Newman TC, Mansfield LS, Linz JE. 2000. Identification of ciprofloxacin- resistant Campylobacter jejuni by use of a fluorogenic PCR assay. Journal of clinical microbiology 38:3971-3978. 88 Xia RH, Yosef N, Ubogu EE. 2010. Clinical, electrophysiological and pathologic correlations in a severe murine experimental autoimmune neuritis model of Guillain-Barre syndrome. Journal of neuroimmunology 219:54-63. Yang F, Wang D, Li Y, Sang L, Zhu J, Wang J, Wei B, Lu C, Sun X. 2017. Th1/Th2 Balance and Th17/Treg-Mediated Immunity in relation to Murine Resistance to Dextran Sulfate-Induced Colitis. Journal of immunology research 2017:7047201. Young KT, Davis LM, Dirita VJ. 2007. Campylobacter jejuni: molecular biology and pathogenesis. Nature reviews Microbiology 5:665-679. Yuki N, Susuki K, Koga M, Nishimoto Y, Odaka M, Hirata K, Taguchi K, Miyatake T, Furukawa K, Kobata T, Yamada M. 2004. Carbohydrate mimicry between human ganglioside GM1 and Campylobacter jejuni lipooligosaccharide causes Guillain-Barre syndrome. Proceedings of the National Academy of Sciences of the United States of America 101:11404-11409. Yuki N, Taki T, Inagaki F, Kasama T, Takahashi M, Saito K, Handa S, Miyatake T. 1993. A bacterium lipopolysaccharide that elicits Guillain-Barre syndrome has a GM1 ganglioside-like structure. The Journal of experimental medicine 178:1771-1775. 89 60. 61. 62. 63. 64. CHAPTER 3: INFECTING CAMPYLOBACTER JEJUNI STRAIN DETERMINES TH1/TH17-MEDIATED IMMUNITY AND COLITIS IN INTERLEUKIN-10-DEFICIENT BALB/C MICE To be submitted to: Infection and Immunity ABSTRACT Campylobacter jejuni is an important cause of bacterial enteritis worldwide and is associated with the post-infectious neuropathy Guillain-Barré syndrome (GBS). The immunopathogenesis of GBS is incompletely understood, but both infecting C. jejuni strain characteristics and host genetic background are thought to contribute. The purpose of this study was to develop an in vivo mouse model characterizing the differences in immune response to colitogenic C. jejuni strain 11168 and GBS patient- derived C. jejuni strain 260.94 in orally inoculated C57BL/6 IL-10-/- and BALB/c IL-10-/- mice. We hypothesized that 1) infection with C. jejuni 11168 would induce significant systemic and local Th1/Th17 immune responses in both mouse genotypes, culminating in severe colitis, and 2) infection with C. jejuni 260.94 would lead to Th2-mediated anti-ganglioside antibody production and nerve lesions mimicking GBS in C57BL/6 IL-10-/- mice, while BALB/c IL-10-/- mice would mount Th1/Th17 responses with protection from autoimmunity. Following a 4-week infection period, C. jejuni 11168-infected BALB/c IL- 10-/- mice exhibited strong Th1/Th17-mediated immunity, severe colitis, and anti -GM1 and -GD1a ganglioside antibody production. C. jejuni 260.94-infected BALB/c IL-10-/- mice also exhibited Th1/Th17 responses, but of lesser magnitude than following C. jejuni 11168 infection and without colitis or anti- ganglioside antibody production. C57BL/6 IL-10-/- mice infected with either C. jejuni strain exhibited less colonization, no colitis, and overall milder immune responses than in previously published models from our group. The unexpected finding that C57BL/6 IL-10-/- but not BALB/c IL-10-/- mice carried Lactobacillus murinus, a potential probiotic organism, combined with the atypical results for C57BL/6 IL-10-/- mice presents the possibility of a protective probiotic effect exerted by the resident strain of L. murinus. Results of the current study indicate that BALB/c IL-10-/- mice are a useful model for studying mucosal 90 enteric pathogens such as C. jejuni, which stimulates a marked Th1/Th17 response in mouse strain that is reportedly Th2-biased. Furthermore, colitogenic and GBS patient-derived C. jejuni strains stimulate Th1/Th17 responses of different magnitude and result in distinct disease outcomes in BALB/c IL-10-/- mice. 91 INTRODUCTION Campylobacter spp., spiral, microaerophilic, Gram-negative rods, are considered to be the most frequent cause of bacterial gastroenteritis worldwide.65 While C. jejuni colonizes chickens in high numbers without inducing disease,36 infection in humans leads to inflammatory enteritis including symptoms of abdominal cramping, fever, diarrhea, and myalgia.12 C. jejuni is also associated with numerous post-infectious sequelae, including reactive arthritis, irritable bowel syndrome, and Guillain- Barré syndrome (GBS).30 GBS is a peripheral neuropathy associated with an antecedent infection with viral or bacterial agents, including Epstein-Barr virus, cytomegalovirus, Mycoplasma, or C. jejuni.25; 29 C. jejuni is the most commonly reported antecedent infection,25; 29; 61 associated with approximately 30% of GBS cases.50 The pathogenesis of GBS subsequent to C. jejuni infection is incompletely understood, but is thought to be mediated by generation of cross-reactive antibodies to lipooligosaccharide (LOS) in the C. jejuni outer membrane and structurally similar gangliosides on peripheral nerves.4; 46; 61; 69 Antibodies to gangliosides, especially GM1 and often of the IgG subtype, are frequently found in GBS patients51 and correlate with clinical severity in GBS patients following C. jejuni infection.27; 52 Subtype of anti-ganglioside IgG also impacts patient prognosis: IgG1 subtypes are associated with preceding C. jejuni infection and worse clinical outcome, while IgG3 alone or in combination with IgG1 is associated with preceding respiratory infection and improved outcome.28; 32 Both the infecting C. jejuni strain and the patient genetic background and immune response are thought to contribute to susceptibility to GBS following C. jejuni infection. Structural similarity between the LOS of some C. jejuni strains and GM1 and GD1a gangliosides,22; 69 in addition to the overrepresentation of certain C. jejuni serotypes preceding GBS,2; 34 support a role for C. jejuni strain characteristics in GBS pathogenesis. However, while expression of ganglioside mimics in LOS of C. jejuni isolates from GBS and Miller Fisher syndrome (MFS; a subtype of GBS) patients was more common than 92 in isolates from uncomplicated enteritis patients,4 ganglioside mimics can also be found in C. jejuni isolates causing only enteritis.4; 58 The low proportion (0.07%) of C. jejuni infections associated with subsequent GBS, the rarity of GBS outbreaks, and the geographic clustering of GBS subtypes30; 47 further suggest differences in host genetic background also influence susceptibility to GBS. Thus, bacterial factors such as ganglioside mimicry in the LOS, in addition to patient immune response, likely both contribute to a disease outcome of uncomplicated enteritis versus development of GBS. Recent establishment of murine models exploiting alterations in host immune system, microbiota, or both have been critical in furthering our understanding of pathogenesis and immune response in C. jejuni-induced colitis.9; 15; 21; 42 One landmark model developed by Mansfield et al. utilizes infection of C57BL/6 mice lacking interleukin (IL)-10 with C. jejuni 11168.42 IL-10 is an essential anti- inflammatory mediator and a critical regulator of immune responses to intestinal flora.33 Enterocolitis developing in IL-10-/- mice is thought to be mediated by unchecked Th1 responses, can manifest as early as 3-4 weeks of age, and was reportedly more severe in 3-month-old BALB/c IL-10-/- than C57BL/6 IL-10-/- mutants.10; 33 C. jejuni 11168 was originally isolated from an enteritis patient in the United Kingdom. This strain harbors both GM1 and GM2 mimics,37 but based on the literature has not been associated with GBS in humans. C. jejuni 11168 has an LOS class C and encodes cst-III sialyl transferase, leading to monosialylation rather than the disialylated form most often associated with high anti-ganglioside activity.22 C. jejuni 11168 has colonized C57BL/6 IL-10-/- mice well, producing enterocolitis and primarily Th1/Th17 systemic responses identified by anti-C. jejuni plasma IgG2c, IgG3, and IgG2b isotypes6; 7; 41-43; 54 and increased colonic expression of IFN-γ and IL-17.41 The few exceptions in which marked colitis did not result in C57BL/6 IL-10-/- mice following infection with C. jejuni 11168 include: development of colitis in only 1/5 mice in an initial passage, but followed by severe colitis with subsequent passages;7 more mice in a dose-response experiment displaying mild than severe colitis, although the majority of mice exhibited some colitis;42 and lower than typical colonization rates and gross pathological changes in a 93 pilot experiment.13 Aside from these examples, infection of C57BL/6 IL-10-/- mice with C. jejuni 11168 has provided a repeatable colitis model. C. jejuni strains can vary widely in ability to colonize and cause disease in mice6; 7 and can also elicit contrasting immune responses in mice of the same genetic background.41 C. jejuni 260.94 was originally isolated from a GBS patient in South Africa. It possesses a GM1a ganglioside mimic, has an outer core classified as an LOS class A and encodes a cst-II sialyl transferase.39 C. jejuni 260.94 has colonized different mouse strains at high prevalence with high enteric colony-forming units (CFU), is not colitogenic and can induce anti-ganglioside antibodies in mice.6; 13; 41; 59 Interestingly, C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 did not exhibit colitis, but produced Th2-mediated responses, including plasma anti -C. jejuni, -GM1, and -GD1a ganglioside IgG1 antibodies and upregulated colonic Gata-3 and IL-4 expression.41 C. jejuni 260.94 also induced production of anti-ganglioside antibodies in non-obese diabetic (NOD) mice with and without knockout of the IL-10 gene.59 Murine models of campylobacteriosis assessing both pathogen and host factor interactions would further our understanding of the complex GBS immunopathogenesis. Several reports suggest that mice of C57BL/6 and BALB/c genetic backgrounds develop polarized immune responses to some well- studied pathogens, including in aspects of innate and adaptive immunity. The intracellular protozoan pathogen Leishmania major provides an excellent example: upon infection, resistant C57BL/6 mice down-regulate Th2-associated IL-4 production and produce Th1-associated IFN-γ; in contrast, susceptible BALB/c mice produce IL-4 upon infection leading to non-healing lesions.53 Similarly, when stimulated with various ligands in vitro, peritoneal macrophages, spleen cells, and dendritic cells derived from C57BL/6 mice produced higher levels of pro-inflammatory cytokines including TNF-α, IL-12, IFN-γ, while cells from BALB/c mice produced more IL-4 and MCP-1.38; 45; 63 These reports suggest that BALB/c mice predominantly produce Th2-polarized responses that may enhance the molecular mimicry driving GBS. 94 Thus, our rationale for this study design was based on the contrasting immune responses and disease outcomes elicited in C57BL/6 IL-10-/- mice infected with C. jejuni 11168 or C. jejuni 260.94 strains,41 and contrasting immune responses of C57BL/6 and BALB/c mice to the same stimulus or pathogen.38; 45; 53; 63; 68 We expected that infection of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice with colitogenic and GBS patient-derived C. jejuni strains would elicit polarized immune and disease responses and could further elucidate the underlying adaptive immune mechanisms. We hypothesized that 1) both C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with colitogenic C. jejuni 11168 would develop colitis and primarily Th1/Th17 immune responses, with Th2 responses also contributing in BALB/c IL-10-/- mice; 2) C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 would be protected from colitis and would develop both Th2-mediated responses, including anti-ganglioside antibodies, and nerve lesions manifested by increased macrophage infiltration into dorsal root ganglia; 3) BALB/c IL-10-/- mice infected with C. jejuni 260.94 would similarly be protected from colitis but would mount primarily Th1/Th17-mediated responses with lesser Th2 responses, without developing anti-ganglioside antibodies or nerve lesions. To test these hypotheses, C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were orally infected with C. jejuni strains 11168 and 260.94 and sacrificed after four weeks. Enzyme-linked immunosorbent assays (ELISAs) were used to measure anti -C. jejuni, -GM1, and -GD1a specific IgG subtypes in plasma and C. jejuni-specific IgA in fecal supernatants. Histological lesions in the gastrointestinal tract were graded according to a previously published scale.42 Production of 13 cytokines reflecting adaptive immune responses in the proximal colon was assessed by a flow-cytometry based multiplexed bead assay. Morphometry was used to quantitate number of macrophages labeled immunohistochemically (IHC) with F4/80 in dorsal root ganglia (DRG). Results from this study show that BALB/c IL-10-/- mice mounted Th1/Th17-mediated immune responses to C. jejuni infection that varied in magnitude and disease manifestation depending upon the 95 infecting C. jejuni strain. Infection of C57BL/6 IL-10-/- mice with either C. jejuni strain did not produce adaptive immune responses or disease consistent with previous experiments in our laboratory. All C57BL/6 IL-10-/-, but no BALB/c IL-10-/-, mice were unexpectedly found to harbor Lactobacillus murinus in the gastrointestinal tract, and we speculate that L. murinus may have conferred protection against C. jejuni-induced pathology in C57BL/6 IL-10-/- mice. MATERIALS AND METHODS Experimental Animals. All mouse experiments were performed according to recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. All procedures were approved by the Michigan State University (MSU) Institutional Animal Care and Use Committee (approval numbers 06/12-107-00 and 06/15-101-00). B6.129P2-IL-10tm1Cgn/J (referred to as C57BL/6 IL- 10-/-) and C.129P2(B6)-IL-10tm1Cgn/J (referred to as BALB/c IL-10-/-) mice were originally obtained from the Jackson Laboratories (Bar Harbor, ME). Experimental animals were bred and maintained in-house in a specific pathogen-free colony. DNA was extracted from pooled fecal samples from experimental mice prior to the start of the experiment, and PCR was used to screen for enteropathogens including Helicobacter spp., Campylobacter spp., Enterococcus faecalis, E. faecium, and Citrobacter rodentium as previously described.42 Absence of IL-10 was assessed by a PCR assay based on the method of the Jackson Laboratories (https://www2.jax.org/protocolsdb/f?p=116:5:0::NO:5:P5_MASTER_PROTOCOL_ID,P5_JRS_CODE:23475 ,004333) using DNA extracted from cecal tissue from experimental mice taken at necropsy. Prior to inoculation, mice were transported to MSU’s University Research Containment Facility (URCF) and allowed to acclimate before placement in individual cages. At the humane endpoint or at the end of the experiment, all mice were humanely euthanized using an overdose of CO2 according to the guidelines of 96 the American Veterinary Medical Association (https://www.avma.org/KB/Policies/Pages/Euthanasia- Guidelines.aspx). Experimental Design. Prior to this study, a 15-mouse pilot experiment was performed to verify ability of experimental C. jejuni strains to colonize mice following oral inoculation with sterile pipette tip, in contrast to the gastric gavage method used previously. Five wild-type (WT) BALB/c mice were each orally inoculated with C. jejuni 11168 (3.3 × 109 CFU), C. jejuni 260.94 (4.3 × 109 CFU), or tryptic soy broth (TSB; sham/vehicle). Stable colonization of mice throughout and at the end of the 3-week pilot experiment was verified by culture and C. jejuni gyrA gene-specific PCR as described below. Sixty mice were included in the present study: 30 C57BL/6 IL-10-/- mice and 30 BALB/c IL-10-/- mice. Of the 30 mice of each genotype, 10 were inoculated with C. jejuni 11168, C. jejuni 260.94, or TSB (Table 3.1). Age matched male and female mice were randomized by sex, litter, rack and slot position, and treatment group as previously described.42 Mice were inoculated at 6 weeks of age and sacrificed 4 weeks post-infection. Campylobacter jejuni Inocula Preparation and Inoculation. C. jejuni inocula were prepared as previously described.42 C. jejuni strains were originally obtained from the American Type Culture Collection (ATCC; Manassas, VA) and stored as glycerol stock cultures at -80°C. C. jejuni ATCC 700819 (referred to as C. jejuni 11168) and C. jejuni ATCC BAA-1234 (referred to as C. jejuni 260.94) were streaked onto tryptic soy agar supplemented with 5% defibrinated sheep’s blood (Cleveland Scientific, Bath, OH) (TSAB plates). The plates were incubated for approximately 42 hours at 37°C in sealed jars containing a single CampyGen sachet (Oxoid, Basingstoke, United Kingdom). Growth was harvested using a sterile swab, resuspended in TSB to an optical density (OD) 600 nm of 0.262 for C. jejuni 11168 and 0.255 for C. jejuni 260.94, and 100 µL aliquots were spread on TSAB plates. Following incubation for 97 approximately 18 hours at 37°C with a CampyGen sachet, growth was harvested and resuspended in TSB so that a 1:10 dilution of the culture exhibited an OD600 1.031 for C. jejuni 11168 and 0.956 for C. jejuni 260.94. Gram stain and wet mount preparations of each strain were used to verify purity, spiral morphology, and motility. The cultures were placed on ice and transported to the URCF for immediate inoculation. Each mouse received 100 µL of its assigned treatment (C. jejuni 11168, C. jejuni 260.94, or TSB) by oral inoculation using a sterile 200 µL pipette tip. Serial dilutions of the inocula were made pre- and post-inoculation. CFU in the inocula were determined following approximately 72 hours’ incubation at 37°C with a CampyGen sachet. Each infected mouse received 5.2 × 109 CFU of C. jejuni 11168, or 5.9 × 109 CFU of C. jejuni 260.94. Monitoring for Clinical Signs. Following infection mice were checked at least once daily by a veterinarian, in addition to checks by MSU University Laboratory Animal Resources personnel for clinical signs of illness including rough hair coat, lethargy, hunched posture, dehydration, and soft feces or diarrhea. Mice reaching a pre-determined humane endpoint based on a previously developed standardized scoring sheet42; 59 were promptly euthanized and necropsied. Necropsy Procedures. Mice were humanely euthanized by CO2 overdose when an early humane endpoint was reached, or at the scheduled end of the 4-week experiment. Weight was recorded for each mouse. Blood was collected via cardiac puncture with a 25g needle on a 1 mL tuberculin syringe pre-loaded with 0.1 mL 3.8% sodium citrate. Mice were necropsied in a hood using instruments sterilized with a hot bead sterilizer between isolations from each section of the gastrointestinal (GI) tract. One fecal pellet was collected into an Eppendorf tube and placed on ice for culture. The remaining fecal pellets were placed into a second Eppendorf tube containing 1 mL of 1% Protease 98 Inhibitor Cocktail (Sigma-Aldrich, St. Louis, MO) in phosphate buffered saline (PBS) per 100 mg feces and placed on ice for fecal IgA determination. The cecum was excised including approximately 0.5—1 cm of adjoining proximal colon and ileum. The tip of the cecum was removed. The remaining ileocecocolic junction (ICJ) was injected with 10% neutral buffered formalin before placing onto a sponge in a histological tissue cassette (Fisherbrand Histosette II Tissue Cassette, Fisher HealthCare, Hampton, NH) and immersing in 10% phosphate buffered formalin (Fisher Scientific). Sections of stomach, jejunum, proximal colon, and the tip of the cecum were excised and rinsed in sterile PBS in separate Petri dishes. Pieces of each tissue were divided into three sections: two were snap frozen in microcentrifuge tubes and cryovials, and the third was streaked onto TSAB plates containing cefoperazone (20 µg/mL), vancomycin (10 µg/mL), and amphotericin B (2 µg/mL) (TSAB-CVA plates; all antimicrobials obtained from Sigma-Aldrich, St. Louis, MO). Plates were placed into air-tight jars with a CampyGen sachet. Finally, the mice were skinned and muscle overlying the spine was removed. The roof of the vertebral canal was removed using Castroviejo scissors to expose the spinal cord. Muscle and connective tissue were bluntly dissected to expose the sciatic nerve on each side. Thereafter, the carcass was immersed in a specimen cup containing 10% phosphate buffered formalin for further dissection and retrieval of tissue at a later date. On return to the laboratory, snap frozen tissues were stored at -80°C. Jars with the CampyGen sachet containing TSAB-CVA plates were placed at 37°C. Fecal pellets were crushed with a sterile applicator stick in 300 µL TSB containing 15% glycerol, vortexed, and a loopful was spread onto TSAB- CVA plates. Plates were incubated at 37°C in a microaerobic environment generated by evacuation of anaerobic jars to -25 in Hg and equilibration with a gas mixture comprising 80% N2, 10% CO2, and 10% H2. The remainder of the fecal suspension was frozen at -80°C. Plasma separated from whole blood by centrifugation was harvested and stored at -80°C until analysis. Fecal pellets collected in PBS with 99 protease inhibitor were vortexed for 30 seconds and centrifuged at 16,000 rcf for 10 minutes. Supernatants were collected and stored at -80°C. The ICJ cassettes and carcasses for nerve dissection were transferred from phosphate buffered formalin to 60% ethanol after 24 and 48 hours, respectively. Confirmation of Colonization. Culture. Colonization of the stomach, jejunum, cecal tip, proximal colon, and presence of C. jejuni in the feces sampled at necropsy were reported following a semi-quantitative grading system of plate coverage by C. jejuni colonies.42 After 72 hours of incubation, C. jejuni growth was enumerated as follows: 0 = no growth; 1 = 1-20 CFU; 2 = 20-200 CFU; 3 = over 200 CFU; 4 = confluent growth. C. jejuni colonies were harvested and stored in 15% glycerol in TSB at -80°C. Colonies inconsistent with C. jejuni morphology were swabbed and observed microscopically. In cases where another organism grew on the same plate as C. jejuni, only numbers of C. jejuni were used in semi-quantitative grading. Colonies of different morphology were harvested and stored separately. Following conclusion of the experiment, an isolate of the second organism was streaked onto TSAB and TSAB-CVA plates and incubated in the microaerophilic environment at 37°C for 72 hours. Growth was observed on plates with and without antibiotics. Colony morphology was recorded, growth was swabbed from the two types of plates, and Gram staining was performed. Colonies subsequently grown on TSAB-CVA plates were suspended and serially diluted in TSB and spread upon both TSAB and TSAB- CVA plates. Colony morphology and Gram staining were again assessed, and the two types of plates with growth were submitted to the Diagnostic Center for Population and Animal Health (currently the Veterinary Diagnostic Laboratory), MSU, for analysis by matrix-assisted laser desorption/ionization time of flight (MALDI-TOF) mass spectrometry. PCR. C. jejuni gyrA gene-specific PCR66 was performed to confirm colonization of infected mice at sacrifice and exclude colonization in control mice. Isolates of C. jejuni cultured from samples obtained 100 at necropsy were used for culture-positive mice; DNA extracted from cecal tissue obtained during necropsy was used for TSB-inoculated and infected but culture-negative mice. Immunohistochemical Labeling of C. jejuni in Ileocecocolic Junctions. IHC was performed by the Investigative Histopathology Laboratory, MSU. Specimens were processed, embedded in paraffin and sectioned on a rotary microtome at 4 m. Sections were placed on slides coated with 2% 3- aminopropyltriethoxysilane and dried at 56C overnight. The slides were subsequently deparaffinized in xylene, hydrated through descending grades of ethyl alcohol to distilled water, and placed in Tris Buffered Saline pH 7.4 (TBS; Scytek Labs, Logan, UT) for 5 minutes for pH adjustment. Following TBS, slides underwent heat induced epitope retrieval utilizing Scytek Citrate Plus Retrieval pH 6.0 in a Vegetable Steamer for 30 minutes, were allowed to cool on the counter at 25° for 10 minutes, and were rinsed in several changes of distilled water. Endogenous Peroxidase was blocked utilizing a 3% Hydrogen Peroxide/Methanol bath for 30 minutes followed by running tap and distilled water rinses. Following pretreatment, standard micro-polymer complex staining steps were performed at room temperature on the IntelliPath™ Flex Autostainer. All staining steps were followed by rinses in TBS Autowash buffer (Biocare Medical, Concord, CA). After blocking for nonspecific protein with Background Punisher (Biocare) for 5 minutes, sections were incubated with Rabbit Polyclonal anti – Campylobacter jejuni (US Biological Canada) (Cat# C1037-08) @ 1:500 diluted in Normal Antibody diluent (NAD-Scytek) for 60 minutes. Rabbit anti – Rodent™ Micro-Polymer (Biocare) was applied for 60 minutes followed by reaction development with Romulin AEC™ (Biocare) for 10 minutes and counterstaining with Cat Hematoxylin (Biocare) for 1 minute. Presence, abundance, and location of positive labeling within the ileum, cecum, and colon were graded using a semi-quantitative scale by a single investigator (LS Mansfield) who was blinded to mouse identity and experimental group. The grading system included semi-quantitative scoring for positive labeling in the lumen (associated with mucus, contents, crypts, and crypt abscesses); epithelium (apical 101 or basolateral surfaces, within paracellular junctions or effacing lesions); lamina propria (between cells, or intracellularly within polymorphonuclear cells or macrophages/dendritic cells); and submucosa (within connective tissue, intracellularly within polymorphonuclear cells or macrophages/dendritic cells, associated with vasculature, lymphoid tissues, or histiocytes). Pathologic Changes: Gastrointestinal Tract. Gross lesions in the gastrointestinal tract, such as thickening, enlargement, or watery or soft contents in the cecum or colon and changes in the ileocecocolic lymph node and spleen, noted by a veterinarian and other experienced personnel during necropsy were recorded. The fixed ICJ sections were embedded in paraffin, finely sectioned at 4-5 µm, routinely stained with Hematoxylin and Eosin (H&E), and coverslipped by the Investigative Histopathology Laboratory at MSU. The sections were evaluated by a board-certified veterinary clinical pathologist (JM Brudvig) who was blinded to mouse identity and experimental group. With the exception that intraepithelial lymphocytes were not scored in the current study, a previously published grading scale42 was used to evaluate changes in the lumen (exudates, excessive mucus), epithelium (surface integrity, goblet cell hypertrophy or depletion, crypt abnormalities), lamina propria (inflammatory cell infiltrates), and submucosa (inflammation, edema, fibrosis). Raw scores (out of 42 total points) were subsequently ranked into semi-quantitative grades as 0 (0-9 points; no colitis), 1 (10- 19 points; mild colitis), or 2 (≥20; moderate or severe colitis). ELISAs. Plasma ELISAs. Plasma stored at -80°C was thawed on ice and aliquoted to prevent repeated freeze-thaw cycles. ELISAs were used to quantify anti -C. jejuni, -GM1, and -GD1a ganglioside IgG1, IgG2a (present in BALB/c mice),44 IgG2c (present in C57BL/6 mice),44 IgG2b, and IgG3 subtypes. Assays were performed as previously described.42 Nunc Maxisorp (Thermo Scientific, Waltham, MA) 96-well plates 102 were coated with antigen and incubated at 4°C overnight. Antigens were diluted in PBS to the following concentrations: Campylobacter jejuni antigen 1.9 µg/mL;42 GM1 antigen (US Biological, Swampscott, MA) 2 µg/mL; GD1a antigen (Sigma Aldrich, St. Louis, MO) 20 µg/mL. The plates were blocked with blocking buffer (10mM PBS with 3% BSA and 0.05% Tween-20 (Sigma)) overnight at 4°C. Following three washes in wash buffer (PBS with 0.025% Tween-20), plasma samples diluted in blocking buffer (all samples were diluted 1:25, except for anti-C. jejuni IgG2b and IgG2a, which were 1:100) were loaded in triplicate. Positive and negative controls and wells containing only blocking buffer were run on each plate. Positive controls included plasma from mice from previous experiments with a high OD or commercially available antibodies (anti-Campylobacter IgG1 (Virostat, Portland, ME)). Anti-Toxoplasma gondii (ViroStat, Portland, ME) antibody was used as a negative control. Sealed anti -GM1 and -GD1a antibody plates were incubated with samples overnight at 4°C, while anti-C. jejuni antibody plates were incubated for 1 hour at room temperature on a platform shaker. Plates were washed, and secondary antibodies (Biotin-SP-conjugated AffiniPure Goat Anti-Mouse IgG1, IgG2a, IgG2b, IgG2c, or IgG3; Jackson ImmunoResearch, West Grove, PA) diluted in blocking buffer were added. Following incubation for 1 hour on a platform shaker, plates were washed again and ExtrAvidin peroxidase (Sigma-Aldrich, St. Louis, MO) diluted 1:2,000 in 10 mM PBS with 1% BSA and 0.05% Tween-20 was added. Plates were incubated for 1 hour on a platform shaker, washed, and tetramethylbenzidine (TMB substrate; Rockland Immunochemicals Inc., Limerick, PA) was added. The reaction was stopped with 2N H2SO4. Absorbance was read at 450 nm using a Bio-Tek EL-800 Universal Microplate Reader with KC Junior software (Bio-Tek Instruments, Winooski, VT). The absorbance generated from the diluent (blocking buffer) alone was subtracted from the mean absorbance obtained for each sample run in triplicate. This adjusted value was used in statistical analyses. Negative values generated by subtracting the absorbance of the blocking buffer from the mean sample absorbance were treated as zero for the purpose of statistical analysis. 103 Fecal Supernatant (anti-C. jejuni IgA) ELISA. Measurement of C. jejuni-specific IgA from fecal supernatants was performed in similar fashion to plasma antibodies. Nunc Maxisorp (Thermo Scientific, Waltham, MA) 96-well plates were coated with C. jejuni antigen diluted in PBS to 1.9 µg/mL42 and incubated at 4°C overnight. The plates were then blocked with blocking buffer overnight. Following washing, 100 µL of undiluted supernatants were loaded onto the plate in duplicate. Fecal supernatants from the pilot experiment with high ODs were used as positive controls on each plate. Wells containing only blocking buffer (used to dilute the secondary antibody) and PBS (for background of fecal supernatant) were included on each plate. Anti-Toxoplasma gondii antibody (ViroStat, Portland, ME) was used as a negative control on each plate. Following 1 hour of incubation at room temperature, the plates were washed and biotin-conjugated goat anti-mouse IgA (Sigma) was added. After 1 hour incubation, plates were washed and ExtrAvidin peroxidase was added. Plates were incubated for 1 hour, washed, and TMB substrate was added. The reaction was stopped with 2N H2SO4 and absorbance was read at 450 nm. The absorbance generated from the PBS alone was subtracted from the mean absorbance obtained for each sample run in duplicate. This adjusted value was used in statistical analyses. For purposes of statistics, negative values generated by subtracting the absorbance of the PBS from the mean sample absorbance were treated as zero, and the two samples with absorbances returning a value of “OUT” (out-of-range; set to ≥3.000 on plate reader) were converted to 3.000 before subtracting the blank. Measurement of Colon Cytokine Production. At necropsy, rinsed proximal colon samples were collected in Eppendorf tubes and snap frozen. Tissues were stored at -80°C until analysis. Subsequently, samples were thawed on ice and wet tissue weight was recorded. Tissue pieces were homogenized using autoclaved microtube pellet pestle rods attached to a handheld Kontes pellet pestle motor for one minute, on ice, in 400 µL of Hank’s Balanced Salt Solution (Sigma), with 0.5% Triton X-100 (Sigma) and 104 the cOmplete Mini EDTA-free Protease Inhibitor cocktail (Roche/Sigma). Homogenates were centrifuged at 12,000  g for 30 minutes at 4°C. Supernatants were aliquoted in cryovials and stored at -80°C until further analysis. Cytokines were measured using a commercially available flow cytometry based, multiplexed, bead assay panel (LEGENDPlex Mouse Th Cytokine Panel, BioLegend, San Diego, CA). Cytokines included in the panel are designed to characterize the Th adaptive immune response by delineating specific Th polarization. Analytes included IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-6, IL-9, IL-10, IL-13, IL-17A, IL-17F, IL-21, and IL-22. Just before analysis, aliquoted supernatants of the colon homogenates were thawed on ice and centrifuged at 300  g for 10 minutes at 4°C to further pellet any debris and ensure clarity. The assay was performed according to the manufacturer’s instructions using undiluted supernatant and a V- bottom microplate. Data were acquired on a BD FACSCanto II flow cytometer and analyzed using the LEGENDplex Data Analysis software. Standard curves were generated for each analyte. Each cytokine had a maximum standard concentration of 10,000 pg/mL. Data are presented as pg cytokine/mg tissue weight. Assessment of Nerve Histopathology. Dissection of fixed tissue was performed using a dissecting microscope. Two to 3 lumbar dorsal root ganglia (DRG) were harvested from the left side of the mouse. Where possible and in many cases, connections from the sciatic nerve to one or all DRG collected, including those in L3, L4, or L5 regions, were visualized prior to removal of DRG. The left sciatic nerve, brachial plexus, and lumbar DRG were placed en bloc in a cassette and stored in 60% ethanol until submission for histopathology. The sections were labeled by IHC for the mouse macrophage marker F4/80 by the Investigative Histopathology Laboratory, Division of Human Pathology, MSU. Specimens were embedded in paraffin and sectioned on a rotary microtome at 4 m. Sections were placed on charged slides and dried at 56C 105 overnight, deparaffinized in xylene, and hydrated through descending grades of ethyl alcohol to distilled water. Slides were then placed in Tris Buffered Saline pH 7.4 (TBS; Scytek Labs – Logan, UT) for 5 minutes for pH adjustment. Following TBS, epitope retrieval was performed using Citrate Plus Retrieval Solution pH 6.0 (Scytek) in a vegetable steamer for 30 minutes followed by a 10 minute countertop incubation and several changes of distilled water. Following pretreatment standard avidin-biotin complex staining steps were performed at room temperature on the DAKO Autostainer. All staining steps are followed by two minute rinses in Tris Buffered Saline + Tween 20 (Scytek). After blocking for non-specific protein with Normal Rabbit Serum (Vector Labs – Burlingame, CA) for 30 minutes, sections were incubated with Avidin / Biotin blocking system for 15 minutes each (Avidin D – Vector Labs / d- Biotin – Sigma). Primary antibody slides were incubated for 60 minutes with the Monoclonal Rat anti- Mouse F4/80 diluted @ 1:100 (AbD Serotec – Raleigh, NC) in Normal Antibody Diluent (NAD) (Scytek). Biotinylated Rabbit anti-Rat IgG (H + L) Mouse Absorbed prepared at 10.0 g/mL in NAD incubated for 30 minutes, followed by R.T.U. Vector Elite Peroxidase Reagent (Vector) incubation for 30 minutes. Reaction development utilized Vector Nova Red Kit peroxidase chromogen incubation of 15 minutes followed by counterstain in Gill 2 Hematoxylin (Cancer Diagnostics – Durham, NC) for 30 seconds, differentiation, and dehydration, clearing, and mounting with Permount mounting media. Number of F4/80 positive cells was quantified by morphometry. Images were analyzed using ImageJ software (version 2.0.0rc-49/1.51d), distributed by Fiji (Fiji Is Just ImageJ)56; 57 for Windows (http://imagej.net/Fiji/Downloads). The investigator (MM Cluett) was blinded to mouse genotype and treatment group. Contiguous images of each DRG section obtained at 100 magnification (10 objective and ocular) were opened in the ImageJ program. Positive cells were marked on the image using the “Cell Counter” plugin. After all positive cells were marked, the area was outlined using the “Freehand Selections” Trace Tool. When necessary, multiple areas were traced individually and the sum of the 106 areas was recorded. Results are given as number of F4/80 positive cells/area, with area representing 100,000 pixels. The slides were finally unblinded for statistical analysis. Statistical Analyses. Analyses were performed and figures generated using commercial statistical software packages (SigmaStat 3.5, Systat Software, San Jose, CA; and GraphPad Prism 6, GraphPad Software, San Diego, CA) or through online applications (VassarStats: Website for Statistical Computation; vassarstats.net). P-values ≤0.05 were considered statistically significant. Data were tested for normality and equal variance. Survival curves were compared by log-rank (Mantel-Cox) testing. The Freeman-Halton extension of the Fisher Exact Probability Test was used to assess differences in C. jejuni colonization depending upon mouse genotype and infecting C. jejuni strain. Colonization was assessed between all infected C57BL/6 IL-10-/- and BALB/c IL-10-/- mice (regardless of infecting C. jejuni strain), along with testing for (1) differences in colonization of each of the two C. jejuni strains within each mouse genotype, and (2) differences in colonization of the two mouse genotypes with each of the two C. jejuni strains considered individually. Semi-quantitative colonization grades were entered as 0, no CFU; 1, 1-20 CFU; 2, 21-200 CFU; and 3, >200 CFU (a single grade, combining grades 3 and 4 used in assigning values of plate coverage by C. jejuni colonies). Differences in gross pathology in infected C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were similarly assessed by the Freeman-Halton extension of the Fisher Exact Probability Test. Gross pathology was assessed between all infected C57BL/6 IL-10-/- and BALB/c IL-10-/- mice (regardless of infecting C. jejuni strain), along with testing for (1) differences in pathology resulting from of each of the two C. jejuni strains within each mouse genotype, and (2) differences in pathology in the two mouse genotypes with each of the two C. jejuni strains considered individually. For these analyses, gross pathology recorded at necropsy was graded as 0 changes; 1 change; 2 changes; and 3-4 changes as a single grade. 107 Kruskal-Wallis analysis followed by Dunn’s post-testing was performed on semi-quantitative data following ranking of raw scores for assessment of colon histopathology. The non-parametric ANOVA and analysis of semi-quantitative grades, rather than raw scores, were chosen to account for non-independence of some parameters of the grading system. One- or two-way ANOVA and Kruskal-Wallis ANOVA on ranks with appropriate post-hoc testing (Holm-Sidak, Tukey, Dunn’s) were used to evaluate impact of mouse genotype and infecting C. jejuni strain on production of anti -C. jejuni and -ganglioside plasma antibodies, C. jejuni-specific IgA in fecal supernatant, cytokines reflecting Th polarization in the proximal colon, and number of F4/80 positive cells in DRG. Data not meeting the assumption of equal variance were analyzed non-parametrically by Kruskal-Wallis. RESULTS Confirmation of Mouse Genotype and Absence of Enteropathogens. Absence of IL-10 was confirmed by PCR in all 60 experimental mice (data not shown). Additionally, pooled fecal samples (at least one fecal pellet per group-housed cage) obtained pre-inoculation were negative by PCR for Helicobacter spp., Campylobacter spp., Enterococcus faecalis, E. faecium, and Citrobacter rodentium (data not shown). Clinical Signs and Survivorship. Clinical signs of illness, including a slight hunch, decreased activity, roughened hair coat, and soft or sticky stool were first noted on day 16 post-infection in BALB/c IL-10-/- mice infected with C. jejuni 11168. In total, 5/10 C. jejuni 11168-infected BALB/c IL-10-/- mice required early humane euthanasia. In the C57BL/6 IL-10-/- groups, 4 mice infected with C. jejuni 11168 showed mild intermittent clinical signs including slight hunching, slightly decreased activity, and a questionably slight rough hair coat starting on day 20; one of these mice required early humane euthanasia because the clinical signs score was greater than 10. In all other experimental mice, no clinical signs aside from 108 questionable and intermittent findings including slight hunching, decreased activity levels, or roughened hair coat were noted. In all mice, signs were monitored carefully using our scoring sheet by at least one observer and least once per day, and mice were promptly euthanized if humane endpoint was determined. In total, 5 BALB/c IL-10-/- mice and one C57BL/6 IL-10-/- mouse, all infected with C. jejuni 11168, required early euthanasia between days 17 and 25 post infection because their clinical signs scores exceeded the limit. No mouse inoculated with TSB or C. jejuni 260.94 required early humane euthanasia. Survival curves for all treatment groups are shown in Figure 3.1A; curves were significantly different from each other when assessed by log-rank (Mantel-Cox) testing (P = 0.0001). C57BL/6 IL-10-/- Mice: Clinical Signs and Survivorship. Clinical signs in C57BL/6 IL-10-/- mice included mild intermittent hunching, lethargy, or roughened hair coat. One mouse infected with C. jejuni 11168 was humanely sacrificed on day 25 due to a hunched posture. No mouse in the TSB or C. jejuni 260.94 groups required early sacrifice. Survival curves for C57BL/6 IL-10-/- mice are shown in Figure 3.1B; curves were not significantly different from each other (Mantel-Cox, P = 0.3679). BALB/c IL-10-/- Mice: Clinical Signs and Survivorship. Clinical signs in BALB/c IL-10-/- mice were first observed in C. jejuni 11168-infected mice on day 16, and included slight hunching, soft or sticky stools with wet or matted fur around the anus, decreased activity, and slightly roughened hair coat. Two of these mice were euthanized and necropsied on day 17, one on day 21, and two on day 25. No mouse in the TSB or C. jejuni 260.94 groups required early sacrifice. Survival curves for BALB/c IL-10-/- mice are shown in Figure 3.1C; when compared by Mantel-Cox analysis, curves were significantly different from each other (P = 0.002). 109 Colonization. Culture. Colonization data, represented by culturable C. jejuni isolated from the cecum, are shown in Figure 3.2. Although feces and sections of stomach, jejunum, and proximal colon taken at necropsy were also assessed by culture for presence of C. jejuni, the cecum was previously shown to be the most consistently and heavily colonized location in the GI tract.42; 43 Colonization for all experimental mice together is shown in Figure 3.2A. When all 20 infected BALB/c IL-10-/- mice were compared with all 20 infected C57BL/6 IL-10-/- mice, a significantly higher proportion of BALB/c IL-10-/- mice were more heavily colonized than C57BL/6 IL-10-/- mice in the cecum at the time of necropsy (PB < 0.0001). Differences in colonization between C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with C. jejuni 11168 were also compared, as were differences between mouse genotypes infected with C. jejuni 260.94. A significantly higher proportion of BALB/c IL-10-/- mice infected with C. jejuni 11168 had higher cecal colonization scores than C. jejuni 11168-infected C57BL/6 IL-10-/- mice (PB = 0.001). Similarly, BALB/c IL-10-/- mice infected with C. jejuni 260.94 had significantly higher cecal colonization rates than C57BL/6 IL-10-/- mice also infected with C. jejuni 260.94 (PB = 0.015). When all 5 sampled areas of the gastrointestinal tract are included, a total of 9/10 BALB/c IL-10- /- mice inoculated with C. jejuni 260.94 were culture positive, 5/10 C57BL/6 IL-10-/- mice inoculated with C. jejuni 260.94 were positive, 10/10 BALB/c IL-10-/- mice inoculated with C. jejuni 11168 were positive, and 9/10 C57BL/6 IL-10-/- mice inoculated with C. jejuni 11168 were positive. C. jejuni gyrA gene-specific PCR confirmed positive culture results, and PCR performed on extracted cecal DNA from infected but culture negative mice also confirmed negative culture results. All TSB inoculated mice were culture negative for C. jejuni in all areas sampled; C. jejuni gyrA gene specific PCR performed on extracted cecal DNA from TSB mice was also negative in all 20 control mice. Immunohistochemical Labeling of C. jejuni in Ileocecocolic Junctions. Labeling of C. jejuni by IHC in ileocecocolic junctions of control and infected mice showed that 9/10 mice in the BALB/c IL-10-/-, C. 110 jejuni 11168-infected group exhibited positive intracellular labeling within macrophage/dendritic cell types within the submucosa, versus between 0-3/10 mice in the other five treatment groups. Similarly, 4/10 C. jejuni 11168-infected BALB/c IL-10-/- mice exhibited intracellular labeling within macrophage/dendritic cell types within the lamina propria; this was not seen in any other mice in any group. Non-specific labeling does occur with this method. Positive staining was noted in at least one area of the epithelium or submucosa in 6/10 sham-inoculated BALB/c IL-10-/- mice and 3/10 sham- inoculated C57BL/6 IL-10-/- mice, along with staining of luminal contents in virtually all experimental mice, regardless of treatment group. C57BL/6 IL-10-/- Mice: Colonization. Culture. Cecal colonization by C. jejuni is shown for C57BL/6 IL-10-/- groups in Figure 3.2B. C57BL/6 IL-10-/- mice infected with C. jejuni 11168 did not have significantly different colonization scores in the cecum than those infected with C. jejuni 260.94 (PB = 0.184). Positive C. jejuni 260.94 culture results in all samples were as follows: 0/10 in the stomach; 4/10 in the jejunum; 3/10 in the cecum; 4/10 in the proximal colon; 3/10 in the feces. Positive C. jejuni 11168 culture results were as follows: 1/10 in the stomach; 5/10 in the jejunum; 7/10 in the cecum; 5/10 in the proximal colon; 3/10 in the feces. Immunohistochemical Labeling of C. jejuni in Ileocecocolic Junctions of C57BL/6 IL-10-/- Mice. In the epithelium, 3/10 TSB mice exhibited labeling associated with the apical surface. In the C. jejuni 260.94 and C. jejuni 11168 infection groups, 2/10 mice in each group were positive for labeling in the epithelium, one in each group in association with the apical surface and another in each group associated with an effacing lesion. No C57BL/6 IL-10-/- mouse exhibited labeling associated with the lamina propria. The single C57BL/6 IL-10-/- mouse with positive labeling intracellularly in the submucosa was infected with C. jejuni 260.94. 111 BALB/c IL-10-/- Mice: Colonization. Culture. Cecal colonization by C. jejuni in BALB/c IL-10-/- mice is shown in Figure 3.2C. BALB/c IL- 10-/- mice infected with C. jejuni 11168 did not have significantly different colonization scores in the cecum than those infected with C. jejuni 260.94 (PB = 0.195). Positive C. jejuni 260.94 culture results in all samples were as follows: 2/10 in the stomach; 2/10 in the jejunum; 9/10 in the cecum; 9/10 in the proximal colon; 9/10 in the feces. Positive C. jejuni 11168 culture results were as follows: 8/10 in the stomach; 8/10 in the jejunum; 10/10 in the cecum; 10/10 in the proximal colon; 10/10 in the feces. Immunohistochemical Labeling of C. jejuni in Ileocecocolic Junctions of BALB/c IL-10-/- Mice. In the epithelium, 5/10 sham-inoculated BALB/c IL-10-/- mice exhibited positive labeling at the apical surface, with one of these also staining positively in an effacing lesion. In BALB/c IL-10-/- mice infected with C. jejuni 260.94, 6/10 mice exhibited positive labeling on the apical surface. Two of these 6 also stained positively in an effacing lesion and one of these also was positive on the basolateral surface. Within the C. jejuni 11168-infected mice, 8/10 showed positive labeling on the apical surface. Of these 8, staining in an effacing lesion was noted in 3, and additional staining in paracellular junctions was seen in one. Additionally, only the C. jejuni 11168 group had mice exhibiting positivity within macrophages/dendritic cells in the lamina propria (4/10 in this group). Positive labeling in the submucosa was seen in all three groups of BALB/c IL-10-/- mice. 3/10 TSB mice had positive staining intracellularly in the macrophage/dendritic cell types, and one of these also had staining identified in regional lymphoid tissue. BALB/c IL-10-/- mice infected with C. jejuni 260.94 also were positive in the macrophage/dendritic cell types in the submucosa (3/10; one of these also had vascular tissue involvement). Finally, in the 9/10 C. jejuni 11168-infected mice showing positive submucosal staining, C. jejuni was found intracellularly in macrophages/dendritic cells in all 9, in the lymphoid tissue in 2, and associated with vasculature in 2. 112 Presence of Lactobacillus murinus in C57BL/6 IL-10-/-, but not BALB/c IL-10-/-, Mice. When assessing colonization of C. jejuni by culture, TSAB-CVA plates from all 30 C57BL/6 IL-10-/- mice (irrespective of treatment group), but no BALB/c IL-10-/- mice, were noted to exhibit growth of a second organism in all five samples of the gastrointestinal tract (stomach, jejunum, cecum, colon, and feces). These colonies had a small, dry, flat, puckered appearance, were hemolytic, and exhibited robust growth on all plates, with or without concurrent growth of C. jejuni colonies. A greenish-orange hue was observed on the swab while harvesting these colonies. Wet mount preparation revealed non-motile rods larger than C. jejuni. When a stored isolate from the cecum of a TSB-inoculated mouse was later streaked onto TSAB and TSAB-CVA plates, differences in colony morphology and microscopic appearance were observed. On plates without antibiotics, colonies appeared more mucoid and on Gram staining, the organisms appeared as mostly fat, Gram-positive rods. In contrast, colonies grown on plates with antibiotics had a more dry and puckered appearance and microscopically the organisms appeared as mostly Gram- positive filamentous rods with occasional fat rods. Colonies from the stored isolate grown on TSAB-CVA plates were diluted and spread onto plates with and without antibiotics. Gram staining revealed a near-pure growth of filamentous rods from plates containing antibiotics, and near-pure growth of fatter rods from plates without antibiotics. The fatter rods were identified by MALDI-TOF mass spectrometry as Lactobacillus murinus with a score of 2.3, and the filamentous rods also were identified as L. murinus with a score of 2.2. The unexpected presence of L. murinus only in C57BL/6 IL-10-/- mice, combined with the atypical results for infected C57BL/6 IL-10-/- mice compared to previous studies from our group, complicated the comparison of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice as originally intended. Data are thus described first with C57BL/6 IL-10-/- and BALB/c IL-10-/- mice together, followed by analyses of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice separately. 113 Gross Pathology and Histological Assessment of Colitis. Gross pathological data are shown in Figure 3.3. Pathology in the cecum, colon, mesenteric lymph nodes (MLN), or spleen at the time of necropsy was recorded. Possible changes included thickening, enlargement, or watery or soft contents in the cecum or colon and enlarged MLN or spleen. Pathology for all experimental mice together is shown in Figure 3.3A. When all 20 infected BALB/c IL-10-/- mice were compared with all 20 infected C57BL/6 IL-10- /- mice, a significantly higher proportion of BALB/c IL-10-/- mice had greater pathology than C57BL/6 IL- 10-/- mice at the time of necropsy (PB < 0.0001). Differences in gross pathology between C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with C. jejuni 11168 were also compared, as were differences between mouse genotypes infected with C. jejuni 260.94. A significantly higher proportion of BALB/c IL-10-/- mice infected with C. jejuni 11168 had greater pathology scores than C. jejuni 11168-infected C57BL/6 IL-10-/- mice (PB = 0.001). Similarly, BALB/c IL-10-/- mice infected with C. jejuni 260.94 had greater pathology scores than C. jejuni 260.94-infected C57BL/6 IL-10-/- mice (PB < 0.01). Assessment of histopathologic lesions in the ileocecocolic junctions (ICJ) is shown in Figure 3.4. Raw scores are displayed graphically for all mice together in Figure 3.4A. Raw scores were subsequently placed into grades, and statistical analysis was performed on these data using the non-parametric Kruskal Wallis one-way ANOVA on ranks to account for non-independence within some parameters of the previously published42 grading system. A raw score of 0-9 indicates no colitis is present (semi- quantitative grade 0), raw scores of 10-19 indicate mild colitis (grade 1), and scores ≥20 reflect moderate to severe colitis (grade 2). BALB/c IL-10-/- mice infected with C. jejuni 11168 had significantly higher colitis scores than all other treatment groups, excepting BALB/c IL-10-/- mice infected with C. jejuni 260.94. The most pronounced changes in BALB/c IL-10-/- mice infected with C. jejuni 11168 compared to other treatment groups included higher frequency of mucus and cellular exudate within the lumen, higher numbers of mice with groups of damaged surface epithelial cells, more severe goblet cell depletion, more severity in crypt changes including irregular architecture, dilation, and 114 cryptitis/abscesses, increased cellular infiltrates in the lamina propria, and more inflammation in the submucosa. C57BL/6 IL-10-/- Mice: Gross Pathology and Histological Assessment of Colitis. Gross pathology in C57BL/6 IL-10-/- mice is shown in Figure 3.3B. Within the C. jejuni 11168-infected mice, with the exception of one mouse with a thickened proximal colon, the changes were limited to mildly enlarged MLN and/or spleen. Similarly, within the C57BL/6 IL-10-/- mice infected with C. jejuni 260.94, one mouse had a slightly thickened proximal colon and the remaining changes were attributable to mildly to moderately enlarged MLN. C57BL/6 IL-10-/- mice infected with C. jejuni 11168 did not have significantly different gross pathology scores than those infected with C. jejuni 260.94 (PB = 0.262). No gross pathology was found in any mouse receiving TSB. Assessment of histopathology in the ICJ of C57BL/6 IL-10-/- mice is shown in Figure 3.4B. One sham-inoculated mouse had a score of 11, indicating mild colitis. Of the C. jejuni 260.94-infected mice, one had a score of 10 (mild colitis), and one other had a score of 32, reflecting severe colitis; no other mice in this group had raw scores above 8. In the C. jejuni 11168-infected group, a single mouse had a score of 13, indicating mild colitis, and all other mice had no colitis (scores ranging from 3─8). There was no statistical difference between groups (P = 0.715). BALB/c IL-10-/- Mice: Gross Pathology and Histological Assessment of Colitis. Gross pathology in BALB/c IL-10-/- mice is shown in Figure 3.3C. Pathology was noted in 9/10 C. jejuni 11168-infected mice, with 7 of these 9 mice exhibiting 3 or 4 changes. Pathology was noted in 6/10 C. jejuni 260.94-infected mice; 4 of these mice exhibited changes in the cecum and/or colon, and all 6 had enlarged MLN, spleen, or both. BALB/c IL-10-/- mice infected with C. jejuni 11168 did not have significantly different gross pathology scores than those infected with C. jejuni 260.94 (PB = 0.084). No gross pathology was found in any mouse receiving TSB. 115 Assessment of histopathology in the ICJ of BALB/c IL-10-/- mice is shown in Figure 3.4C. In the TSB group, 3 mice exhibited mild colitis (raw scores 10─19), while all other mice in this group had no colitis (raw score range 6─9). In the C. jejuni 260.94 group, 7/10 mice had scores between 10─17 indicating mild colitis, and the other 3 mice had no colitis. Pathology was most severe in the C. jejuni 11168-infected group. The majority (8/10) of mice had moderate or severe colitis, with scores ranging from 23─33; one mouse had mild colitis and the last had a raw score of 9, indicating no colitis. Statistically significant differences were found between the C. jejuni 11168 and 260.94 groups (P = 0.024) and between C. jejuni 11168 and sham-inoculated groups (P = 0.0004). The colon of a sham- inoculated BALB/c IL-10-/- mouse with a grade of 0 (no colitis) is shown in Figure 3.4D, compared to the severe colitis in a C. jejuni 11168-infected BALB/c IL-10-/- mouse (Figure 3.4E). Systemic and Local Immune Response to C. jejuni Infection. Plasma ELISAs. Systemic immune responses to C. jejuni infection were evaluated by measurement of C. jejuni-specific IgG1, IgG2b, IgG3, and IgG2c (in C57BL/6 IL-10-/- mice) or IgG2a (in BALB/c IL-10-/- mice)44 antibodies in plasma by indirect ELISA (Figure 3.5). Additionally, as antibodies cross-reacting to LOS on the C. jejuni outer membrane and gangliosides on peripheral nerves are a hallmark of GBS and thought to be involved in GBS immunopathogenesis, anti -GM1 antibodies (Figure 3.6) and -GD1a antibodies (Figure 3.7) also were assessed by indirect ELISA. When anti-C. jejuni specific IgG1, IgG2b, and IgG3 antibody levels were compared between all 6 groups (Figure 3.5A), the most pronounced responses were seen in C. jejuni 11168-infected BALB/c IL- 10-/- mice. This group had significantly higher production of Th2-mediated IgG1 than BALB/c IL-10-/- mice receiving TSB or C. jejuni 11168-infected C57BL/6 IL-10-/- mice. A robust Th1/Th17 response, evidenced by production of IgG2b and IgG3, also was most pronounced in BALB/c IL-10-/- mice infected with C. jejuni 11168. Mice in this group had significantly higher plasma IgG2b than TSB-inoculated mice of either 116 mouse genotype. BALB/c IL-10-/- mice infected with C. jejuni 11168 also produced the most robust IgG3 response, which was significantly higher than all other treatment groups excepting the BALB/c IL-10-/- C. jejuni 260.94 group. Collectively, these results suggest that of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with C. jejuni 11168 or 260.94 strains, BALB/c IL-10-/- mice infected with C. jejuni 11168 mount the most pronounced immune response and that this response is mixed Th1/Th2/Th17 in character. Production of anti-ganglioside GM1 (Figure 3.6A) and GD1a (Figure 3.7A) antibodies was compared between all 6 treatment groups. IgG1, IgG2b, and IgG3 isotypes reacting with GM1 were nearly identical in pattern and magnitude of response to those isotypes reacting with GD1a. Significant differences between groups were found only in IgG2b production. C57BL/6 IL-10-/- mice infected with C. jejuni 11168 produced the strongest IgG2b response, although a low-level background production in all C57BL/6 IL-10-/- groups may be present, as suggested by the background in C57BL/6 IL-10-/- TSB- inoculated mice. Local Response: Fecal IgA and Colon Cytokine Production. Local immune responses in the gastrointestinal tract were assessed by measurement of C. jejuni-specific IgA in fecal supernatants (Figure 3.8) and evaluation of the production of cytokines reflecting character of Th adaptive immunity in the proximal colon (Figure 3.9). When IgA production was compared in all 6 treatment groups (Figure 3.8A), results were similar to the systemic response assessed by plasma anti-C. jejuni antibodies. The strongest response was seen in BALB/c IL-10-/- mice infected with C. jejuni 11168. These mice produced significantly higher amounts of C. jejuni-specific IgA than all other treatment groups, excepting the C. jejuni 260.94-infected BALB/c IL-10-/- group. Measurement of colon cytokine production was performed with a panel including IFN-γ, TNF-α, IL-2, IL-4, IL-5, IL-6, IL-9, IL-10, IL-13, IL-17A, IL-17F, IL-21, and IL-22. Detectable amounts of IL-4, IL-5, IL- 17F, and IL-21 were found in only 1/60 mice, and IL-13 production was not detected in any mouse. As expected, no detectable IL-10 was produced by any mouse. Therefore, only production of IFN-γ, TNF-α, 117 IL-2, IL-6, IL-9, IL-17A, and IL-22 was analyzed. Comparison of cytokine production in the proximal colon between all 6 groups (Figure 3.9) similarly showed that BALB/c IL-10-/- mice infected with C. jejuni 11168 produced significantly higher amounts of IFN-γ, TNF-α, IL-17A, and IL-22, and although non-significant, BALB/c IL-10-/- mice infected with C. jejuni 11168 had the highest IL-6 levels. Collectively, these results show that of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with C. jejuni 260.94 or C. jejuni 11168, the most potent mucosal immunity and robust Th1/Th17/Th22-mediated gastrointestinal responses are elicited in BALB/c IL-10-/- mice infected with C. jejuni 11168. C57BL/6 IL-10-/- Mice: Immune Response to C. jejuni Infection. Plasma ELISAs. Production of anti-C. jejuni specific IgG1, IgG2b, IgG3, and IgG2c is compared between C57BL/6 IL-10-/- infection groups in Figure 3.5B. Infected mice produced slightly higher levels of IgG2b, IgG3, and IgG2c anti-C. jejuni antibodies than TSB control mice, but these responses were mild and no significant difference was found between any C57BL/6 IL-10-/- groups in any of the four anti-C. jejuni antibody isotypes tested. Production of anti-GM1 antibodies (Figure 3.6B) and anti-GD1a antibodies (Figure 3.7B) by mice in different C57BL/6 IL-10-/- infection groups was compared. Again, production of IgG1, IgG2b, IgG3, and IgG2c antibodies reacting with GM1 closely mirrored those reacting with GD1a. IgG2b production by mice infected with C. jejuni 11168 represented the strongest response, although no significant differences were seen between any treatment groups, in any of the four isotypes tested reacting with GM1 or GD1a. Local Response: Fecal IgA and Colon Cytokine Production. Comparison of C. jejuni-specific IgA production between mice in C57BL/6 IL-10-/- treatment groups is shown in Figure 3.8B. Although both C. jejuni 260.94- and C. jejuni 11168-infected C57BL/6 IL-10-/- mice produced slightly higher levels of IgA than control mice, these responses were mild overall and no significant difference was found between any treatment groups. Local cytokine production in the proximal colon of C57BL/6 IL-10-/- mice was 118 similarly mild (Figure 3.10), and no statistically significant difference was found between any treatment groups. IL-6 and IL-17A were not produced in detectable amounts in any group. Although the differences were not significant, there appeared to be mild relative up- or down-regulation of three cytokines in the panel depending upon infection status. C. jejuni 11168-infected mice produced the greatest amount of IFN-γ, while this group produced no IL-22 or IL-9. Mice infected with C. jejuni 260.94 also produced no IL- 22. Collectively, the data support a mild overall local response to C. jejuni infection in C57BL/6 IL-10-/- mice, reflected by C. jejuni-specific IgA and colonic cytokine production, that was not significantly different between infected and control mice. BALB/c IL-10-/- Mice: Immune Response to C. jejuni Infection. Plasma ELISAs. Production of anti-C. jejuni specific IgG1, IgG2b, IgG3, and IgG2a is compared between BALB/c IL-10-/- infection groups in Figure 3.5C. The most pronounced responses were elicited in mice infected with C. jejuni 11168. The most robust response was Th1/Th17-mediated IgG2b: mice infected with either C. jejuni 11168 or C. jejuni 260.94 had significantly increased production compared to sham-inoculated mice. Additionally, C. jejuni 11168-infected mice had significantly higher IgG2b production than C. jejuni 260.94-infected mice. Th1-mediated IgG3 and IgG2a isotypes mirrored each other, with significantly higher production of both isotypes in C. jejuni 11168-infected mice than in C. jejuni 260.94- or sham-inoculated groups. Overall significance was identified in IgG1 production by one- way ANOVA (P = 0.044), although no significant pairwise comparisons were identified following Holm- Sidak post-testing. Together, these data suggest a primary Th1/Th17-mediated immune response to C. jejuni 11168 and a similar response of lesser magnitude to C. jejuni 260.94 by infected BALB/c IL-10-/- mice. Production of anti-GM1 antibodies (Figure 3.6C), and anti-GD1a antibodies (Figure 3.7C) was compared between BALB/c IL-10-/- infection groups. The pattern of production of IgG1, IgG2b, IgG3, and IgG2a antibody isotypes reacting with GM1 again closely resembled those reacting with GD1a. Mice 119 infected with C. jejuni 11168 produced the most robust responses, manifested mainly by significantly increased levels of IgG2b and IgG2a isotypes to both GM1 and GD1a gangliosides in C. jejuni 11168- infected mice, compared to the other two BALB/c IL-10-/- treatment groups. No significant differences between treatment groups were found in production of anti -GM1 or -GD1a IgG1 or IgG3. Collectively, these data suggest that BALB/c IL-10-/- mice infected with C. jejuni 11168 produce Th1/Th17-mediated anti-ganglioside antibodies. Local Response: Fecal IgA and Colon Cytokine Production. Production of C. jejuni-specific IgA measured in fecal supernatants in BALB/c IL-10-/- mice was compared (Figure 3.8C). Similar to C. jejuni- specific IgG3 and IgG2a antibodies measured in the plasma, the most pronounced IgA response was elicited in C. jejuni 11168-infected mice. These mice showed significantly increased IgA production compared to both sham-inoculated and C. jejuni 260.94-infected mice. Colon cytokines reflecting adaptive immune responses also were compared between BALB/c IL-10-/- treatment groups (Figure 3.11). Mice infected with C. jejuni 11168 produced significantly higher amounts of IFN-γ and TNF-α than the other two treatment groups, and significantly more IL-22 than C. jejuni 260.94-infected mice. C. jejuni 11168-infected mice also trended toward increased production of IL-6 and IL-17A compared to the other groups. Similar to comparisons within the C57BL/6 IL-10-/- group, production of IL-9 was decreased, although non-significantly, in C. jejuni 11168-infected BALB/c IL-10-/- mice compared to the sham- or C. jejuni 260.94-inoculated groups. Collectively, these data support the conclusion that C. jejuni 11168 elicits the strongest mucosal immunity and primarily a Th1/Th17/Th22-mediated local adaptive response in BALB/c IL-10-/- mice. Nerve Histopathology. Macrophage densities in the lumbar dorsal root ganglia were identified by labeling of cells by IHC with the F4/80 macrophage marker. Number of macrophages in DRG were quantified by morphometry and presented as F4/80 positive cells per unit area of DRG. Results of 120 comparisons between all 6 treatment groups are shown in Figure 3.12A. No statistically significant difference was found between any treatment groups (two-way ANOVA; P-values for mouse genotype, treatment group, and interaction were all >0.05). The widest range of scores was seen in sham- inoculated BALB/c IL-10-/- mice (7─39). Overall these data suggest that, in this model, production of anti - GM1 and -GD1a IgG2b antibodies was alone not sufficient to mediate increased macrophage infiltration or proliferation in the DRG. C57BL/6 IL-10-/- Mice: Nerve Histopathology. Numbers of macrophages in DRG were compared within C57BL/6 IL-10-/- treatment groups (Figure 3.12B). C. jejuni 260.94-infected mice showed the greatest range (7─28) of F4/80 positive cells per unit area. Although mice infected with C. jejuni 11168 had higher scores than the other two groups, the difference between the three groups was not statistically significant (Kruskal-Wallis, P = 0.157). BALB/c IL-10-/- Mice: Nerve Histopathology. Comparison of F4/80-positive cells in DRG of BALB/c IL-10-/- mice is shown in Figure 3.12C. TSB-inoculated mice showed the greatest variability in scores, while C. jejuni 11168-infected mice had the most tightly clustered scores. No significant differences were seen between groups (one-way ANOVA, P = 0.641). As in the all-group comparisons, these data suggest that production of anti -GM1 and -GD1a IgG2a and IgG2b antibodies in C. jejuni 11168-infected BALB/c IL-10-/- mice was alone not sufficient to mediate increased macrophage numbers in DRG. DISCUSSION The purpose of this study was to develop a mouse model to evaluate the contributions of both infecting C. jejuni strain characteristics and host genetic background in determining disease outcome, specifically colitis and susceptibility to GBS. Results of this study indicate that BALB/c IL-10-/- mice infected with the colitogenic C. jejuni 11168 strain provide an additional model of colitis mimicking 121 human disease, presenting opportunities for studying the pathogenesis of campylobacteriosis in mice of an additional genetic background; to our knowledge, BALB/c IL-10-/- mouse models of C. jejuni-induced colitis have not been previously reported. Infection of BALB/c IL-10-/- mice with GBS-associated C. jejuni 260.94 produced a less robust Th17 systemic response than that of C. jejuni 11168, without anti- ganglioside antibody production, significantly increased IgA production, or development of colitis. These results indicate that in BALB/c IL-10-/- mice, magnitude of mucosal immunity and Th1/Th17-mediated immune responses vary with infecting C. jejuni strain, providing an additional model to further study mechanisms by which C. jejuni induces a variable immune response. In contrast, C57BL/6 IL-10-/- mice infected with either C. jejuni 11168 or the GBS patient-derived C. jejuni strain 260.94 remained clinically healthy, with the exception of one C. jejuni 11168-infected mouse deemed to require early humane sacrifice. These results contrast with previously published reports indicating C57BL/6 IL-10-/- mice infected with C. jejuni 11168 develop severe colitis, while infection with C. jejuni 260.94 stimulates anti-ganglioside antibody production.6; 7; 41-43; 54 C57BL/6 IL-10-/- mice mounted only weak immune responses and did not develop the expected colitis or anti-ganglioside antibodies with the respective C. jejuni strains reported in previous studies. The original intent of the study was to compare C. jejuni infection in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice to evaluate the impact of immune bias due to host genetic background in determining immunity. However, the atypical lack of immune response or colitis despite C. jejuni colonization of infected C57BL/6 IL-10-/- mice, with the concurrent unexpected finding of L. murinus carriage, necessitated the additional interpretation of results within C57BL/6 IL-10-/- and BALB/c IL-10-/- genotypes separately. The first aim of this study involved determining polarization of the adaptive immune response and susceptibility to colitis depending upon mouse genotype and infecting C. jejuni strain. We hypothesized that C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with colitogenic C. jejuni 11168 would develop colitis and primarily Th1/Th17 immune responses, with a mixed Th1/Th17/Th2 response 122 in BALB/c IL-10-/- mice due to the reported Th2 bias in this mouse strain. This hypothesis was proposed based upon previous studies demonstrating colitis, systemic (plasma antibodies) or local colonic Th1/Th17 responses in C. jejuni 11168-infected mice, including C57BL/6 IL-10-/- 6; 7; 41-43; 54 and NOD IL-10- /- genotypes.59 Indeed, when all treatment groups were compared together in the current study, BALB/c IL-10-/- mice infected with C. jejuni 11168 mounted the most robust systemic C. jejuni-specific IgG2b and IgG3 responses, along with significantly increased IgG2a (Figure 3.5), reflecting Th1/Th17-mediated class switching.1; 5; 64 A robust Th1/Th17 cytokine response also was identified locally in the proximal colon: C. jejuni 11168-infected BALB/c IL-10-/- mice again showed the most pronounced IFN-γ, TNF-α, IL-6, IL-17A, and IL-22 production of any treatment group (Figure 3.9). Interestingly, when C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were compared together, BALB/c IL-10-/- mice infected with C. jejuni 11168 also had a significant Th2-mediated IgG1 plasma response, but when BALB/c IL-10-/- mice were analyzed individually, no pairwise comparisons were significant (Figure 3.5). In contrast to the results presented in Malik et al., in which C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 exhibited upregulated colonic expression of Th2-associated Gata-3, IL-4, and IL-13 measured by real-time PCR,41 mice in the current study were not producing detectable amounts of colonic Th2-associated IL-4, IL-5, or IL-13 at the time of euthanasia. These results are consistent with a shift to Th1-mediated immunity in the absence of IL-1010 and with primary systemic and local Th1/Th17 responses to C. jejuni infection. The robust Th1/Th17 and essentially negligible Th2 response in BALB/c IL-10-/- mice, especially those infected with C. jejuni 11168, reflects a consistent Th1/Th17 response to C. jejuni as seen in previous in vivo mouse6; 21; 41; 42; 59 and ex vivo human models.18 Interestingly, production of Th17-related cytokines IL-17A and IL-22 by C. jejuni 11168-infected BALB/c IL-10-/- mice was significantly higher than all three C57BL/6 IL-10-/- treatment groups (Figure 3.9), but when BALB/c IL-10-/- groups were compared alone production was not significantly different than in sham inoculated BALB/c IL-10-/- mice (Figure 3.11). This suggests a higher 123 baseline production of Th17-related cytokines in BALB/c IL-10-/- than C57BL/6 IL-10-/- mice in the current study, regardless of infection status, reflecting a stronger default to Th17- as well as Th1-mediated immunity in the absence of IL-10. While WT BALB/c mice may be predisposed to Th2 responses in various models, in the absence of IL-10, a shift to a heightened combined Th1/Th17 response was induced with C. jejuni infection in the current and previous studies (Brudvig et al., unpublished, Chapter 2)). This primary Th1/Th17 response with apparent suppression of a Th2 response is exacerbated in IL- 10-/- mice, but an absent or muted Th2 response is likely not only due to absence of regulatory IL-10 but also due to C. jejuni itself: in the previous study, infected WT BALB/c mice also mounted Th1/Th17, but not Th2, responses (Brudvig et al., unpublished (Chapter 2)). Th17 cells are a relatively recently described subset of Th cells distinct from Th1 and Th2 subsets. Induction of Th17 cells is closely related to that of Treg cells but differentiation of these two subsets is mutually exclusive: TGF-β in steady-state induces Foxp3+ Treg cells, but with concurrent production of IL-6 by innate cells during infection or inflammation, TGF-β and IL-6 together induce Th17 cell differentiation.11 Th17 cells play a primary role in control of extracellular bacteria at mucosal surfaces, including through recruitment of inflammatory cells such as neutrophils, and in maintaining mucosal homeostasis in the gut.31; 64 Interestingly, the chemically induced (dextran sulfate sodium) colitis model is characterized by resistance to colitis in BALB/c mice, which mount a primarily Th2/Th17/Treg response characterized by higher levels of IL-4, IL-6, IL-10, and IL-17, compared to susceptibility in C57BL/6 mice producing more IFN-γ and TNF-α.68 These data suggest that, in the current study, the absence of the regulatory actions of IL-10 compounded with infection with an enteric mucosal pathogen such as C. jejuni 11168 led to an unchecked Th1/Th17-mediated response in BALB/c IL-10-/- mice. This response is apparently dependent upon characteristics of the infecting C. jejuni strain, as infection with C. jejuni 260.94 induced significant plasma IgG2b production but did not result in significant colitis or increased colon cytokine production (Figures 3.4, 3.5, and 3.11). Differences in 124 colitogenic potential and immune responses induced by infection with C. jejuni 11168 and 260.94 strains may be explained in part by the enhanced invasion efficiency and intracellular survival of C. jejuni 11168 compared to C. jejuni 260.94, demonstrated by IHC in the current study and in vitro using dendritic cells (Brudvig et al., unpublished (Chapter 4)). BALB/c IL-10-/- mice infected with C. jejuni 11168 exhibited the most severe colitis of any treatment group, and compared to sham-inoculated and C. jejuni 260.94-infected BALB/c IL-10-/- groups alone (Figure 3.4). The severe colitis seen in this group was characterized by inflammatory exudates in the lumen consisting primarily of mononuclear cells and neutrophils; damage to groups of surface epithelial cells; marked goblet cell depletion; crypt abnormalities including irregular architecture, dilation, and cryptitis/abscesses; diffuse and marked inflammatory infiltrates consisting of mononuclear cells, neutrophils, and plasma cells in the lamina propria; and frequent extension of inflammatory infiltrates into the submucosa and mesenteric fat. These lesions are similar to those described previously in C57BL/6 IL-10-/- mice,42 although severe colitis was not observed in C. jejuni 11168-infected C57BL/6 IL-10-/- mice in the current study. BALB/c IL-10-/- mice infected with C. jejuni 11168 also exhibited the most marked production of C. jejuni-specific IgA in fecal supernatants (Figure 3.8), reflecting a strong mucosal immune response. Although direct comparison between BALB/c IL-10-/- and C57BL/6 IL-10-/- mice in the current study is complicated by atypical results and carriage of L. murinus in C57BL/6 IL-10-/- mice, it is worth noting that BALB/c mice were reported to have enhanced vitamin A metabolism in the intestine and increased ability to induce IgA class switching compared to C57BL/6 mice.23 Evaluation of systemic immunity reflected by C. jejuni-specific plasma IgG antibody isotypes, local immunity assessed by colonic cytokines reflecting Th polarization, and mucosal immunity gauged by IgA production all suggest that BALB/c IL-10-/- mice infected with C. jejuni 11168 mounted strong Th1/Th17 responses resulting in unchecked inflammatory pathology in the colon. Thus, the hypothesis 125 that Th1/Th17 responses mediated pathology in BALB/c IL-10-/- mice, although without a significant Th2 component, can be accepted. In the current study, the virtual lack of immune response or pathology as typically reported in C. jejuni 11168-infected C57BL/6 IL-10-/- mice precludes accepting or rejecting this hypothesis for C57BL/6 IL-10-/- mice. Though additional experiments will be necessary to rigorously validate BALB/c IL-10-/- mice as a model of C. jejuni-induced colitis, these results suggest that a bias toward Th1/Th17-mediated immunity in the absence of IL-10 makes BALB/c IL-10-/- mice a useful model for studying enteric mucosal pathogens such as C. jejuni. The second aim of this study involved assessing the importance of host genetic background in determining susceptibility to C. jejuni-induced GBS, using production of anti-ganglioside antibodies, a hallmark of GBS, and increased numbers of macrophages in DRG as indicators. We hypothesized that mice of both genetic backgrounds infected with C. jejuni 260.94 would not develop colitis, but Th- mediation of immunity would vary between mouse strains. We reasoned that based upon previous models by Malik et al41 and Brudvig et al (unpublished (Chapter 2)), the Th2 response in C57BL/6 IL-10-/- mice would lead to anti -GM1 and -GD1a antibodies and nerve lesions, while a mixed Th1/Th17/Th2 response would protect BALB/c IL-10-/- mice from neurological manifestations. Histopathological scoring of the ICJ is shown in Figure 3.4. Seven of 10 C. jejuni 260.94-infected BALB/c IL-10-/- mice had mild colitis (raw scores 10─17; the remaining 3 in this group were <10, indicating no colitis). Despite higher numbers of C. jejuni 260.94-infected BALB/c IL-10-/- mice with raw colitis scores ≥10, this group was not statistically different from sham-inoculated BALB/c IL-10-/- mice, likely owing to 3 uninfected mice also having mild colitis. Similarly, within the C57BL/6 IL-10-/- group, one mouse in the C. jejuni 260.94-infected group had severe colitis (score = 32), but overall C. jejuni 260.94 did not induce significant colitis in C57BL/6 IL-10-/- mice. Lack of colitis induced with C. jejuni 260.94 in these mouse strains is consistent with previous studies6; 41 and with Brudvig et al. (unpublished, (Chapter 126 2)), indicating that despite colonization and stimulation of immunity, this strain is less colitogenic than C. jejuni 11168. Following assessment of colitis, susceptibility to C. jejuni 260.94-induced GBS was evaluated. Measurement of anti-ganglioside antibodies by ELISA (Figures 3.6 and 3.7) demonstrated significant differences in anti-GM1 and anti-GD1a IgG2b between many groups when all treatment groups were compared. C57BL/6 IL-10-/- mice showed the most robust responses, especially those infected with C. jejuni 11168; however, when analyzed separately, C57BL/6 IL-10-/- mice showed no significant increases in any anti-GM1 or anti-GD1a IgG isotype. Clearly, C57BL/6 IL-10-/- mice infected with C. jejuni 11168 produced high levels of anti-GM1 and anti-GD1a IgG2b antibodies, although some TSB-inoculated mice also had detectable levels of these antibodies indicating that C57BL/6 IL-10-/- mice had a higher baseline production. C. jejuni 11168 also induced the most pronounced production of Th1/Th17-mediated anti- ganglioside IgG2a and IgG2b isotypes in BALB/c IL-10-/- mice. When only BALB/c IL-10-/- groups were compared, C. jejuni 11168-induced increases were statistically significant when compared to both sham- and C. jejuni 260.94-inoculated groups. This Th1/Th17 mediated production of anti-ganglioside IgG2a and IgG2b in BALB/c IL-10-/- mice infected with C. jejuni 11168, compared with Th2-mediated production of anti -GM1 and -GD1a IgG1 in C. jejuni 260.94-infected C57BL/6 IL-10-/- mice previously reported,41 further highlights the potential importance of both host and C. jejuni strain characteristics in disease outcomes. Though associated with enteritis and not GBS, C. jejuni 11168 produces GM1 and GM2 ganglioside mimics and thus production of anti-ganglioside antibodies could occur, possibly associated with the strong Th1/Th17 response to a mucosal enteric pathogen. Persistent colonization of C. jejuni 11168 and resulting severe inflammation may have led to alterations in the microbiota; expression of ganglioside mimics by commensal flora leading to anti-ganglioside antibody production cannot be excluded. The lack of anti -GM1 or -GD1a antibody production in C. jejuni 260.94-infected BALB/c IL-10-/- mice is consistent with a previous study, in which increased anti-ganglioside antibodies related to 127 presence or absence of IL-10 but not to infection status (Brudvig et al., unpublished (Chapter2)). Lack of anti-ganglioside antibody production in C. jejuni 260.94-infected C57BL/6 IL-10-/- mice in the current study is consistent with relatively low colonization rates identified by culture, lack of invasion into deeper layers in the proximal colon reflected by IHC, and overall mild and non-significant responses to C. jejuni infection in these mice. A second indicator of GBS susceptibility chosen for this study was number of macrophages in lumbar DRG. Macrophages were chosen due to their postulated role in pathogenesis of GBS, particularly in the acute motor axonal neuropathy pattern of GBS reported by some to be the subtype most closely associated with C. jejuni infection.17; 47; 61 Dorsal root ganglia were chosen because some GBS patients describe pain and sensory defects, and the cell bodies of sensory fibers are located in DRG.47 The blood- nerve barrier is also reportedly particularly leaky within DRG, increasing vulnerability to immune attack.47 Furthermore, pathology was described in dorsal roots following autopsies of patients with motor-sensory axonal GBS24 and a significant increase in F4/80 positive cells was seen in the DRG, but not sciatic nerve or brachial plexus, of C. jejuni 260.94-infected NOD IL-10-/- mice.59 There was no significant increase in F4/80 positive cells in DRG in any treatment group in the current study (Figure 3.12) despite significantly increased production of anti-ganglioside antibodies by C. jejuni 11168-infected BALB/c IL-10-/- mice. Similarly, significantly increased anti-GM1 IgG1 in C. jejuni 260.94 and 11168-infected wild type C57BL/6 mice with a humanized microbiota did not result in increased F4/80 positive cells in sciatic nerves and DRG.13 The lack of correlation between macrophage infiltration or proliferation and increased plasma anti -GM1 and -GD1a antibodies may indicate that anti- ganglioside antibodies alone are insufficient to cause nerve lesions. The exact immunopathogenesis of nerve damage in the AMAN form of GBS is incompletely understood, but macrophages, complement, and IgG have all been implicated.47; 61 Labeling of these other components may reveal that nerve pathology in GBS is mediated by multiple factors. Nerve lesions in this study were assessed in tissues 128 obtained at the scheduled necropsy 4 weeks post-infection or when earlier sacrifice was necessary; a time-course study would more clearly delineate if and when inflammatory nerve lesions occur in relation to elevations in anti-ganglioside antibodies. In considering susceptibility to GBS, we hypothesized that infection with GBS patient strain C. jejuni 260.94 would not lead to colitis in either C57BL/6 IL-10-/- or BALB/c IL-10-/- mice, but would drive Th2-mediated immunity with susceptibility to GBS in C57BL/6 IL-10-/- mice, while predominant Th1/Th17 responses in BALB/c IL-10-/- mice would result in protection from neurological manifestations. Considering colon histopathology, anti-ganglioside antibody production, and macrophage numbers in DRG, we conclude that C. jejuni 260.94 was not colitogenic in either mouse strain, but also did not induce production of anti -GM1 and -GD1a antibodies. Surprisingly, significant anti-ganglioside antibody production was induced only by infection with the colitogenic C. jejuni 11168 strain in BALB/c IL-10-/- mice, albeit without corresponding nerve lesions. The lack of nerve lesions in mice with anti-ganglioside antibodies resulting from infection may also indicate IgG2a and IgG2b isotypes do not mediate nerve damage. The carriage of L. murinus and atypically mild immune responses in C. jejuni infected C57BL/6 IL-10-/- mice precludes the rejection of C57BL/6 IL-10-/- mice as a suitable GBS model. Here L. murinus colonization of C57BL/6 IL-10-/- mice was associated with a lack of significant disease manifestations and immune responses, including IgG1 anti-ganglioside antibodies and nerve lesions following C. jejuni infection. It will be necessary to validate the possibility that L. murinus acts as a protective probiotic preventing GBS. The severe colitis, marked IgA production, and upregulation of Th1/Th17 related cytokines in the colon of C. jejuni 11168-infected BALB/c IL-10-/- mice compared to all other treatment groups correlated with enhanced intracellular labeling and subepithelial presence of C. jejuni 11168 in this group as evaluated by IHC in the ICJ. C. jejuni 11168-infected BALB/c IL-10-/- mice had the highest proportion of mice exhibiting intracellular labeling within macrophage/dendritic cell types in both the lamina propria 129 and the submucosa. These results are consistent with in vitro studies demonstrating enhanced invasion efficiency and intracellular survival of C. jejuni 11168 compared to C. jejuni 260.94 in cultured dendritic cells assessed by gentamicin killing assay (Brudvig et al., unpublished (Chapter 4)). Studies attempting to correlate C. jejuni invasiveness or association with cultured cells and clinical symptoms or pathology have shown conflicting results. Association of C. jejuni and C. coli strains with cultured HeLa cells was related to symptoms of fever and diarrhea, but not blood in the feces.20 A subsequent study determined that association of C. jejuni and C. coli strains with HeLa cells was not adequate in discriminating between isolates from patients with colitis versus non-inflammatory diarrhea, although a higher proportion of isolates causing colitis displayed transcytosis through polarized Caco-2 monolayers compared to isolates from patients with non-inflammatory diarrhea.19 Invasion and survival ability of C. jejuni strains in vitro did not impact intestinal lesion development in a pig model,35 but C. jejuni invasion was positively correlated with immunopathology in mice by comparison with mutant strains capable of colonization but showing reduced invasion capacity in vitro.9 In the current studies, the enhanced invasion efficiency and intracellular survival ability of C. jejuni 11168 shown in vitro (Brudvig et al., unpublished, Chapter 4)) fits with in vivo IHC assessment of intracellular organisms, subepithelial location of C. jejuni 11168, and severity of colitis in BALB/c IL-10-/- mice. The IHC labeling method of C. jejuni in intestinal tissues exhibits non-specific staining, as evidenced by positive labeling in the epithelium or submucosa in some sham-inoculated mice. All 20 sham inoculated mice in this study were culture negative for C. jejuni in all sampled areas of the gastrointestinal tract, and negative status was confirmed by C. jejuni gyrA gene-specific PCR performed on DNA extracted from cecal tissue. This non- specific labeling precludes the use of this method for more than identification of broad pattern. Despite the disadvantage of non-specific staining, this method was chosen as a way to assess C. jejuni load, location within the mucosa, submucosa, and extraintestinal areas such as the lymph node, and presence of intracellular organisms. 130 Colonization of C57BL/6 IL-10-/- mice with C. jejuni 11168 or C. jejuni 260.94 did occur, but these mice displayed virtually none of the pathology or immune responses reported in previous studies involving these C. jejuni strains and C57BL/6 IL-10-/- mice.6; 7; 41-43; 54 Ninety percent of C. jejuni 11168- infected C57BL/6 IL-10-/- mice were colonized at the end of the study, but the gross and histopathological lesions expected in the cecum and colon were not seen. In contrast to previous models, no significant C. jejuni-specific IgG2c, IgG2b, or IgG3 plasma responses were seen, and these mice exhibited virtually no C. jejuni- specific mucosal IgA or cytokines reflecting a local adaptive immune response in the colon. Similarly, 50% of the C. jejuni 260.94-infected C57BL/6 IL-10-/- mice were colonized at the end of the 4-week study, but this group did not exhibit any significant increase in C. jejuni-specific plasma antibodies of any isotype or mucosal IgA, and no significant differences in production of cytokines reflecting adaptive immunity in the colon were seen. No GM1 or GD1a ganglioside-specific antibodies were made, in contrast to significant production of anti-GM1 and anti-GD1a IgG1 previously reported.41 Additionally, C. jejuni 260.94 has stably colonized high percentages of inoculated mice, including C57BL/6 IL-10-/- mice, for up to several weeks in previous studies; (Brudvig et al., unpublished (Chapter 2))6; 13; 41; 59 colonization with C. jejuni 260.94 in the current study is comparably lower. Despite colonization determined by culture and PCR methods in a total of 14/20 infected C57BL/6 IL-10-/- mice (70%) at the end of the 4-week study or at humane endpoint for one mouse at day 25, in vivo assessment of intracellular organisms and location beyond the surface epithelium into the lamina propria and submucosa by IHC labeling of C. jejuni in the ICJ suggested decreased migration or invasiveness of C. jejuni in C57BL/6 IL-10-/- compared to BALB/c IL-10-/- mice: no C57BL/6 IL-10-/- mouse exhibited labeling in the lamina propria, and only a single C57BL/6 IL-10-/- mouse infected with C. jejuni 260.94, and none infected with C. jejuni 11168, showed positive staining intracellularly in the submucosa. Interestingly, despite a lack of detectable colitis or systemic or local immune responses in 131 infected mice, 12/20 infected C57BL/6 IL-10-/- mice exhibited an enlarged regional lymph node and/or spleen at necropsy. This suggests that the persistent colonization, though present more superficially compared to BALB/c IL-10-/- mice, was stimulating a local, but relatively mild, immune response that did not lead to systemic morbidity or pathology in these mice. It was recently shown that a C. jejuni mutant strain lacking the formic acid receptor Tlp7 was able to colonize mice with similarly high loads to the parental WT strain, although the mutant displayed five times reduced invasion capacity in vitro and did not induce immunopathology in vivo in the mice compared to the WT strain.9 Thus, the results of the current study indicating that infected C57BL/6 IL-10-/- mice exhibited virtually no colitis and no systemic or local immunity may reflect not only decreased colonization rates compared to previous studies, but also decreased invasiveness or translocation of the colonized C. jejuni beyond the epithelium. The relative protection of C57BL/6 IL-10-/- mice infected with C. jejuni 11168 or C. jejuni 260.94 in this study, compared with previously published outcomes and those of BALB/c IL-10-/- mice in the current study receiving the same inocula, warrants further exploration to determine if carriage of the L. murinus isolated from C57BL/6 IL-10-/- mice may have contributed to this unexpected outcome. Members of the Lactobacillus genus can be found in the intestinal tract of humans, though many species are apparently not stable inhabitants but instead originate in the oral cavity or from exogenous sources such as food.62 In contrast, lactobacilli form stable populations in the gastrointestinal tracts of animals, including mice, where colonization of the stomach occurs with epithelial associations resembling biofilms.62 Presence of L. murinus in mice, including C57BL/6 substrains, can reportedly vary by vendor.26 L. murinus has also been found in healthy BALB/c mice, colonizing the upper gastrointestinal tract in an arrangement described as mimicking a biofilm.3 Lactobacillus spp. may therefore increase colonization resistance by forming a physical barrier inhibiting colonization of pathogenic bacteria. Beyond colonization resistance, commensal bacteria have also demonstrated local immunomodulatory effects.14; 60 Interestingly, through interactions with dietary components in the gut, L. reuteri contributed 132 to differentiation of CD4+ T cells into CD4+CD8αα+ intraepithelial lymphocytes, T cells with regulatory functions similar to Tregs.14 L. murinus did not have the same effect as L. reuteri in that study,14 but an unidentified immunomodulatory effect of L. murinus, in addition to possible colonization resistance, cannot be excluded in C57BL/6 IL-10-/- mice in our study. Studies evaluating the potential effect of lactobacilli on colitis in mice report differences in Lactobacillus spp. and strains present in mice with and without colitis, and suggest a potential protective role of lactobacilli in preventing development of spontaneous colitis in IL-10 deficient mice. L. murinus and other Lactobacillus spp. were isolated from Swiss Webster and inducible nitric oxide synthetase (iNOS)-deficient C57BL/6 mice without colitis, while only L. johnsonii was isolated from C57BL/6 IL-10-/- mice with colitis.49 Prior to development of colitis, 129 Sv/Ev IL-10-/- mice showed increased mucosal invasion of aerobic bacteria with a concurrent decrease in Lactobacillus spp., and repopulation with L. reuteri reduced aerobic bacterial invasion and attenuated the colitis.40 A subsequent study corroborated these findings, demonstrating that IL-10-/- mice on C57BL/6 and C57BL/10 backgrounds fed L. salivarius ssp. salivarius showed reduction in C. perfringens load, development of colitis, neoplasia, and death compared to control mice receiving placebo.48 In the current study, 3/10 sham-inoculated BALB/c IL-10-/- mice exhibited mild spontaneous colitis and 1/10 sham-inoculated C57BL/6 IL-10-/- mice had mild colitis. This may only reflect the relative severity of spontaneous enterocolitis in BALB/c IL-10-/- compared to C57BL/6 IL-10-/- previously reported,10 but an effect of the L. murinus in C57BL/6 IL-10-/- mice in this study should be considered. Standardized experiments specifically designed to examine the prevalence of spontaneous colitis in both BALB/c IL-10-/- and C57BL/6 IL-10-/- mice with and without L. murinus carriage would be necessary to further test the possibility that development of spontaneous colitis may have been ameliorated by L. murinus in the current study. Both in vitro and in vivo studies also have demonstrated antagonism of Lactobacillus spp. against C. jejuni. Several lactobacilli, including L. pentosus, 6 strains of L. plantarum, and L. 133 pseudomesenteroides exhibited in vitro inhibitory activity against 3 different C. jejuni strains; however, despite repeated administration of the chosen L. plantarum strain to chickens in attempt to estimate the effect on C. jejuni populations in vivo, colonization could not be established.55 Cell-free extracts of milk fermented by two Lactobacillus spp., L. acidophilus and L. rhamnosus, inhibited growth and down- regulated the flaA σ28 promoter activity of two C. jejuni strains.16 Invasion of C. jejuni into cultured epithelial cells was inhibited by L. helveticus and L. rhamnosus, but probiotic activity varied depending upon Lactobacillus spp., C. jejuni strain, and epithelial cell line.67 Protective effects, including immunomodulation, of lactobacilli against C. jejuni-induced colitis in mice have also been shown. While L. johnsonii did not reduce colonization of abiotic C57BL/6 mice with C. jejuni 81-176, reduction in colonic apoptosis, number of colonic B cells, and secretion of IL-6, MCP-1, TNF, and NO was observed.8 Considered together, lactobacilli may exhibit both in vitro and in vivo inhibitory activity against C. jejuni, but protective effects vary depending on C. jejuni strain and Lactobacillus spp. or strain, and in vitro inhibition may be difficult to reproduce in vivo due to difficulty in establishing stable colonization. Because of the inhibitory activity of Lactobacillus spp. against C. jejuni shown in vitro and in vivo, the carriage of L. murinus by C57BL/6 IL-10-/- mice exhibiting abnormally mild responses despite colonization with C. jejuni warrants further study. Detection of L. murinus in the current study was unexpected, and the source is unknown. Direct comparison to outcomes in BALB/c IL-10-/- mice without L. murinus cannot be made, as BALB/c IL-10-/- mice are of a different genetic background and are not yet an established and repeatedly validated model C. jejuni-induced colitis as are the C57BL/6 IL-10-/- mice. The possibility that the mild responses of C. jejuni-infected C57BL/6 IL-10-/- mice and the presence of L. murinus were coincidental cannot be excluded. TSAB-CVA plates are considered selective for C. jejuni growth, but L. murinus is vancomycin resistant49 and can be found in laboratory mice;3; 26 thus L. murinus present in the gut would be expected to grow simultaneously with C. jejuni on this medium. Furthermore, competition in culture medium resulting in reduced C. jejuni growth, if this occurred, does 134 not necessarily equate to inhibition of C. jejuni in vivo. L. murinus may have occupied a relatively small fraction of the gut microbiota but exhibited disproportionately robust growth in culture with reduced competition from other members of the gut microbiota; levels of L. murinus were significantly expanded during vancomycin treatment in a mouse study.26 Finally, in contrast to established outcomes of severe colitis in repeated experiments involving C57BL/6 IL-10-/- mice infected with C. jejuni 11168,6; 7; 41-43; 54 anomalous results of relatively milder colitis have occurred in individual experiments.7; 13; 42 Therefore, the possibility that the L. murinus harbored by C57BL/6 IL-10-/- mice conferred protection from C. jejuni in this study is intriguing but will require additional rigorous experiments to prove or disprove. An essential step preceding further in vitro or in vivo experiments testing the inhibitory effect of the cultured L. murinus against C. jejuni is the further assessment of the different colony morphologies and appearance on Gram staining observed following growth on plates with and without antibiotics. Furter testing is needed to confirm both morphologies as different phenotypes of the same organism, rather than L. murinus in close association with a second organism. MALDI-TOF mass spectrometry identified organisms with both morphologies as L. murinus with high accuracy scores, and further confirmation of the identity of the two types of rods by 16S sequencing is in progress. If the two colony types are indeed both L. murinus, exhibiting change in morphology due to antibiotic-induced pressure, the single genome will be sequenced. Analysis of the sequenced genome will include evaluation of genes involved in antibiotic resistance and known protective probiotic effects including bacteriocin production. In addition to genome sequencing, in vitro inhibition assays to definitively evaluate the L. murinus interaction with C. jejuni strains 11168 and 260.94 are planned. Assays may include assessment of direct inhibition of C. jejuni growth on agar plates or in wells, using live L. murinus cells or supernatants, respectively.55 Pretreatment of isolated dendritic cells or epithelial cells with L. murinus prior to infection with C. jejuni can be used to evaluate any reduction in C. jejuni invasion by gentamicin killing assay.67 If in vitro results support the possibility of in vivo inhibition of C. jejuni by the isolated strain of L. 135 murinus, an in vivo mouse experiment involving L. murinus-negative BALB/c IL-10-/- mice infected with C. jejuni with and without L. murinus inoculation(s) will be performed. Ability of L. murinus to stably colonize BALB/c IL-10-/- mice following inoculation would first have to be verified. The present study was designed to develop an in vivo mouse model characterizing the differences in immune response to colitogenic and GBS-associated C. jejuni strains in mice of two genetic backgrounds. Models like these are needed to further our understanding of the complex determinants of disease outcomes, such as colitis or GBS, following C. jejuni infection. In this study, infected C57BL/6 IL-10-/- mice mounted only mild immune responses and did not develop colitis or produce anti-ganglioside antibodies as previously reported. The unexpected finding of L. murinus carriage, coupled with atypically mild results in C57BL/6 IL-10-/- mice, presents a promising area of future research to determine whether the L. murinus can be an effective and safe probiotic conferring protection against C. jejuni-mediated pathology. In contrast, C. jejuni-infected BALB/c IL-10-/- mice mounted Th1/Th17-mediated immunity, without a significant Th2 component. Magnitude of this Th1/Th17 response, induction of mucosal immunity, and disease outcome including colitis and production of anti-ganglioside antibodies depended upon infecting C. jejuni strain. BALB/c IL-10-/- mice thus present an additional model mimicking human colitis, and provide an additional avenue for further study of the mechanisms by which different C. jejuni strains induce a variable immune response. 136 APPENDIX 137 Table 3.1. Experimental design. Sixty mice were included, 30 each of C57BL/6 interleukin (IL)-10-/- and BALB/c IL-10-/- genotypes. Ten of each mouse strain were orally inoculated with Campylobacter jejuni strain 260.94, C. jejuni 11168, or vehicle (tryptic soy broth, TSB). Mice were inoculated at 6 weeks of age and humanely euthanized after 4 weeks of infection, or earlier when necessary. Mouse Genotype C57BL/6 IL-10-/- C57BL/6 IL-10-/- C57BL/6 IL-10-/- BALB/c IL-10-/- BALB/c IL-10-/- BALB/c IL-10-/- Treatment Group TSB C. jejuni 260.94 C. jejuni 11168 TSB C. jejuni 260.94 C. jejuni 11168 Number of Mice 10 10 10 10 10 10 138 Figure 3.1. Survival of C57BL/6 interleukin (IL)-10-/- and BALB/c IL-10-/- mice inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB). Mice were inoculated at 6 weeks of age and humanely euthanized at approximately 4 weeks post-infection, or earlier if humane endpoint was reached. Survival curves were significantly different from each other when (A) all 6 treatment groups were compared (log-rank [Mantel-Cox]; P = 0.0001) and when (C) BALB/c IL-10-/- mice alone were compared (log-rank [Mantel-Cox]; P = 0.002), but not when (B) C57BL/6 IL-10-/- mice alone were compared (log-rank [Mantel-Cox]; P = 0.3679). 139 Figure 3.2. Culture results, cecum. Culturable C. jejuni strains 11168 and 260.94 isolated from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice at the time of necropsy for (A) all mice in the experiment, (B) C57BL/6 IL- 10-/- mice only, and (C) BALB/c IL-10-/- mice only. While samples of stomach, jejunum, cecum, proximal colon, and the feces were cultured, the cecum was chosen for graphic representation as C. jejuni typically colonizes the cecum most heavily. In total (out of all 5 sampled areas of the gastrointestinal tract), 9/10 BALB/c IL-10-/- mice infected with C. jejuni 260.94 were culture positive; 5/10 C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 were positive, 10/10 BALB/c IL-10-/- mice infected with C. jejuni 11168 were positive, and 9/10 C57BL/6 IL-10-/- mice infected with C. jejuni 11168 were positive. All sham (tryptic soy broth, TSB) inoculated mice were culture negative for C. jejuni in all areas sampled. 140 Figure 3.3. Gross pathology noted in the cecum, colon, mesenteric lymph nodes (MLN), or spleen at the time of necropsy in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice following inoculation with C. jejuni 11168, C. jejuni 260.94, or sham inoculation (tryptic soy broth; TSB). Pathology is shown for (A) all mice together, (B) C57BL/6 IL-10-/- mice only, and (C) BALB/c IL-10-/- mice only. Possible changes included thickening, enlargement, or watery or soft contents in the cecum or colon, and enlarged lymph nodes or spleen. In C57BL/6 IL-10-/- mice infected with C. jejuni 11168, with the exception of one mouse with a thickened proximal colon, the changes were due to mildly enlarged MLN and/or spleen. Similarly, in C57BL/6 IL-10-/- mice infected with C. jejuni 260.94, one mouse had a thickened proximal colon and the remaining changes were attributable to mildly to moderately enlarged MLN. 141 Figure 3.4. Colon histopathology. C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth, TSB). (A-C) Histopathologic scoring of the ileocecocolic junction. Sections were blinded and given a raw score (0-42 point scale, displayed graphically). Scores were then further graded semi-quantitatively for statistical analysis as Grade 0 (no colitis; 0-9 points), Grade 1 (mild colitis; 10-19 points), or Grade 2 (moderate or marked colitis; ≥20 points). All six treatment groups were analyzed together (A), followed by analysis of C57BL/6 IL-10-/- mice (B) and BALB/c IL-10-/- mice (C) separately. Kruskal-Wallis was performed on semi-quantitative scores, followed by Dunn’s multiple comparisons test. Brackets indicate statistically significant differences between groups. Data are represented by box and whisker, with the box extending from 25th-75th percentiles and the line plotted at the median. Whiskers represent minimum to maximum values. (D) Colon of a sham-inoculated BALB/c IL-10-/- mouse (Grade 0; 10 magnification, H&E). (E) Colon of a C. jejuni 11168-infected BALB/c IL-10-/- mouse; severe pathologic changes, including inflammatory cell infiltrate into the lamina propria and submucosa, mucus and necrotic debris in the lumen, and crypt abscesses, are evident (Grade 2; 10 magnification, H&E). 142 Figure 3.5. Plasma anti-C. jejuni IgG1, IgG2b, IgG3 for all mice, IgG2c (C57BL/6 IL-10-/- only), and IgG2a (BALB/c IL-10-/- only) antibodies measured by ELISA in samples taken at necropsy. C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB). All six treatment groups were analyzed together (panel A), followed by analysis of C57BL/6 IL-10-/- (panel B) and BALB/c IL-10-/- (panel C) mice separately. Statistical analyses included two- or one-way ANOVA followed by Holm-Sidak post-testing, and when the assumption of equal variance was not met, Kruskal-Wallis followed by Tukey’s post-test. Brackets above groups indicate significant pairwise comparisons; P ≤ 0.05 was considered statistically significant. Significance was found overall in analysis of IgG1 production in BALB/c IL- 10-/- mice (one-way ANOVA, P = 0.044); however, no significant pairwise comparisons were found by Holm-Sidak post-testing. Mean ± SEM. 143 Figure 3.6. Plasma anti-GM1 IgG1, IgG2b, IgG3 for all mice, and IgG2c (C57BL/6 IL-10-/- only), and IgG2a (BALB/c IL-10-/- only) antibodies measured by ELISA in samples taken at necropsy. C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB). One C57BL/6 IL-10-/- mouse inoculated with C. jejuni 260.94 did not have enough plasma for analysis and is excluded. All six treatment groups were analyzed together (panel A), followed by analysis of C57BL/6 IL-10-/- (panel B) and BALB/c IL-10-/- (panel C) mice separately. Statistical analyses included one- and two-way ANOVA, and when the assumption of equal variance was not met, Kruskal- Wallis followed by Dunn’s or Tukey’s post-testing. Brackets above groups indicate significant pairwise comparisons; P ≤ 0.05 was considered statistically significant. Mean ± SEM. 144 Figure 3.7. Plasma anti-GD1a IgG1, IgG2b, IgG3 in all mice, and IgG2c (C57BL/6 IL-10-/- only), and IgG2a (BALB/c IL-10-/- only) antibodies measured by ELISA in samples taken at necropsy. C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB). One C57BL/6 IL-10-/- mouse inoculated with C. jejuni 260.94 did not have enough plasma for analysis and is excluded. All six treatment groups were analyzed together (panel A), followed by analysis of C57BL/6 IL-10-/- (panel B) and BALB/c IL-10-/- (panel C) mice separately. Statistical analyses included one- or two-way ANOVA, or when the assumption of equal variance was not met, Kruskal-Wallis followed by Dunn’s or Tukey’s post-test. Brackets above groups indicate significant pairwise comparisons; P ≤ 0.05 was considered statistically significant. Mean ± SEM. 145 Figure 3.8. Measurement of anti-C. jejuni specific IgA in supernatants of feces collected at necropsy from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB). All six treatment groups were analyzed together (A), followed by analysis of C57BL/6 IL-10-/- (B) and BALB/c IL-10-/- (C) mice separately. Because the assumption of equal variance was not met, Kruskal-Wallis followed by Tukey post-hoc testing was performed. Brackets above groups indicate significant pairwise comparisons; P ≤ 0.05 was considered statistically significant. Mean ± SEM. 146 Figure 3.9. Assessment of colon cytokine production in C57BL/6 IL-10-/- and BALB/c IL-10-/- mice inoculated with C. jejuni 11168, C. jejuni 260.94, or sham (tryptic soy broth, TSB) by flow cytometry-based multiplexed bead assay. Production of IL-4, IL-5, IL-10, IL-13, IL-17F, and IL-21 was undetectable in 59-60/60 mice. Statistical analyses included two- and one-way ANOVA with Holm-Sidak post-testing, or when the assumption of equal variance was not met, Kruskal-Wallis followed by Tukey post-testing. Brackets above groups indicate statistically significant pairwise comparisons; P ≤ 0.05 was considered statistically significant. Although a significant genotype effect (BALB/c IL-10-/- > C57BL/6 IL-10-/-) was found by two-way ANOVA in IL-9 production, further analysis by one-way ANOVA did not determine any overall significance. Mean ± SEM. 147 Figure 3.10. Assessment of colon cytokine production in C57BL/6 IL-10-/- mice inoculated with C. jejuni 11168, C. jejuni 260.94, or sham (tryptic soy broth, TSB) by flow cytometry-based multiplexed bead assay. Data were analyzed by one-way ANOVA. No statistically significant differences (P ≤ 0.05) between groups were found. IL-6 and IL-17A production was undetectable in all C57BL/6 IL-10-/- mice. Mean ± SEM. 148 Figure 3.11. Assessment of colon cytokine production in BALB/c IL-10-/- mice inoculated with C. jejuni 11168, C. jejuni 260.94, or sham (tryptic soy broth, TSB) by flow cytometry-based multiplexed bead assay. Data were analyzed by one-way ANOVA, or when assumption of equal variance was not met, by Kruskal-Wallis followed by Tukey’s post-testing. Brackets above groups indicate statistically significant pairwise comparisons; P ≤ 0.05 was considered statistically significant. Mean ± SEM. 149 Figure 3.12. Assessment of cells positively labeled with the F4/80 macrophage marker in lumbar dorsal root ganglia (DRG) of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice at the time of necropsy. Mice were inoculated with C. jejuni 260.94, C. jejuni 11168, or vehicle (tryptic soy broth; TSB) and sacrificed 4 weeks post-infection, or earlier when humane endpoint was reached. The DRG were labeled immunohistochemically for F4/80. Positive cells were scored by morphometry on contiguous images using the Image J program. Results are given as number of F4/80 positive cells/area, with area representing 100,000 pixels. All six treatment groups (A) were analyzed by two-way ANOVA. C57BL/6 IL-10-/- mice (B) were analyzed by Kruskal-Wallis, as the assumption of equal variance was not met. BALB/c IL-10-/- mice (C) were analyzed by one-way ANOVA. No statistically significant differences (P ≤ 0.05) were found between any treatment groups. Data are represented by box and whisker, with the box extending from 25th-75th percentiles and the line plotted at the median. Whiskers represent minimum to maximum values. 150 REFERENCES 151 REFERENCES Abbas AK, Murphy KM, Sher A. 1996. Functional diversity of helper T lymphocytes. Nature 383:787-793. Allos BM, Lippy FT, Carlsen A, Washburn RG, Blaser MJ. 1998. Campylobacter jejuni strains from patients with Guillain-Barre syndrome. Emerging infectious diseases 4:263-268. Almiron M, Traglia G, Rubio A, Sanjuan N. 2013. Colonization of the mouse upper gastrointestinal tract by lactobacillus murinus: a histological, immunocytochemical, and ultrastructural study. Current microbiology 67:395-398. Ang CW, Laman JD, Willison HJ, Wagner ER, Endtz HP, De Klerk MA, Tio-Gillen AP, Van den Braak N, Jacobs BC, Van Doorn PA. 2002. Structure of Campylobacter jejuni lipopolysaccharides determines antiganglioside specificity and clinical features of Guillain-Barre and Miller Fisher patients. Infection and immunity 70:1202-1208. Bai Y, Liu R, Huang D, La Cava A, Tang YY, Iwakura Y, Campagnolo DI, Vollmer TL, Ransohoff RM, Shi FD. 2008. CCL2 recruitment of IL-6-producing CD11b+ monocytes to the draining lymph nodes during the initiation of Th17-dependent B cell-mediated autoimmunity. European journal of immunology 38:1877-1888. Bell JA, Jerome JP, Plovanich-Jones AE, Smith EJ, Gettings JR, Kim HY, Landgraf JR, Lefebure T, Kopper JJ, Rathinam VA, St Charles JL, Buffa BA, Brooks AP, Poe SA, Eaton KA, Stanhope MJ, Mansfield LS. 2013. Outcome of infection of C57BL/6 IL-10(-/-) mice with Campylobacter jejuni strains is correlated with genome content of open reading frames up- and down-regulated in vivo. Microbial pathogenesis 54:1-19. Bell JA, St Charles JL, Murphy AJ, Rathinam VA, Plovanich-Jones AE, Stanley EL, Wolf JE, Gettings JR, Whittam TS, Mansfield LS. 2009. Multiple factors interact to produce responses resembling spectrum of human disease in Campylobacter jejuni infected C57BL/6 IL-10-/- mice. BMC microbiology 9:57. Bereswill S, Ekmekciu I, Escher U, Fiebiger U, Stingl K, Heimesaat MM. 2017. Lactobacillus johnsonii ameliorates intestinal, extra-intestinal and systemic pro-inflammatory immune responses following murine Campylobacter jejuni infection. Scientific reports 7:2138. Bereswill S, Fischer A, Plickert R, Haag LM, Otto B, Kuhl AA, Dasti JI, Zautner AE, Munoz M, Loddenkemper C, Gross U, Gobel UB, Heimesaat MM. 2011. Novel murine infection models provide deep insights into the "menage a trois" of Campylobacter jejuni, microbiota and host innate immunity. PloS one 6:e20953. Berg DJ, Davidson N, Kuhn R, Muller W, Menon S, Holland G, Thompson-Snipes L, Leach MW, Rennick D. 1996. Enterocolitis and colon cancer in interleukin-10-deficient mice are associated with aberrant cytokine production and CD4(+) TH1-like responses. The Journal of clinical investigation 98:1010-1020. 152 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, Weiner HL, Kuchroo VK. 2006. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441:235-238. Blaser M, Engberg J. 2008. Clinical Aspects of Campylobacter jejuni and Campylobacter coli Infections. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Brooks PT, Brakel KA, Bell JA, Bejcek CE, Gilpin T, Brudvig JM, Mansfield LS. 2017. Transplanted human fecal microbiota enhanced Guillain Barre syndrome autoantibody responses after Campylobacter jejuni infection in C57BL/6 mice. Microbiome 5:92. Cervantes-Barragan L, Chai JN, Tianero MD, Di Luccia B, Ahern PP, Merriman J, Cortez VS, Caparon MG, Donia MS, Gilfillan S, Cella M, Gordon JI, Hsieh CS, Colonna M. 2017. Lactobacillus reuteri induces gut intraepithelial CD4(+)CD8alphaalpha(+) T cells. Science 357:806-810. Chang C, Miller JF. 2006. Campylobacter jejuni colonization of mice with limited enteric flora. Infection and immunity 74:5261-5271. Ding W, Wang H, Griffiths MW. 2005. Probiotics down-regulate flaA sigma28 promoter in Campylobacter jejuni. Journal of food protection 68:2295-2300. Drenthen J, Yuki N, Meulstee J, Maathuis EM, van Doorn PA, Visser GH, Blok JH, Jacobs BC. 2011. Guillain-Barre syndrome subtypes related to Campylobacter infection. Journal of neurology, neurosurgery, and psychiatry 82:300-305. Edwards LA, Nistala K, Mills DC, Stephenson HN, Zilbauer M, Wren BW, Dorrell N, Lindley KJ, Wedderburn LR, Bajaj-Elliott M. 2010. Delineation of the innate and adaptive T-cell immune outcome in the human host in response to Campylobacter jejuni infection. PloS one 5:e15398. Everest PH, Goossens H, Butzler JP, Lloyd D, Knutton S, Ketley JM, Williams PH. 1992. Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. Journal of medical microbiology 37:319-325. Fauchere JL, Rosenau A, Veron M, Moyen EN, Richard S, Pfister A. 1986. Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infection and immunity 54:283-287. Fox JG, Rogers AB, Whary MT, Ge Z, Taylor NS, Xu S, Horwitz BH, Erdman SE. 2004. Gastroenteritis in NF-kappaB-deficient mice is produced with wild-type Camplyobacter jejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent colonization with both strains. Infection and immunity 72:1116-1125. Godschalk PC, Kuijf ML, Li J, St Michael F, Ang CW, Jacobs BC, Karwaski MF, Brochu D, Moterassed A, Endtz HP, van Belkum A, Gilbert M. 2007. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barre and Miller Fisher syndromes. Infection and immunity 75:1245-1254. 153 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Goverse G, Olivier BJ, Molenaar R, Knippenberg M, Greuter M, Konijn T, Cook EC, Beijer MR, Fedor DM, den Haan JM, Napoli JL, Bouma G, Mebius RE. 2015. Vitamin A metabolism and mucosal immune function are distinct between BALB/c and C57BL/6 mice. European journal of immunology 45:89-100. Griffin JW, Li CY, Ho TW, Tian M, Gao CY, Xue P, Mishu B, Cornblath DR, Macko C, McKhann GM, Asbury AK. 1996. Pathology of the motor-sensory axonal Guillain-Barre syndrome. Annals of neurology 39:17-28. Hadden RD, Karch H, Hartung HP, Zielasek J, Weissbrich B, Schubert J, Weishaupt A, Cornblath DR, Swan AV, Hughes RA, Toyka KV, Plasma Exchange/Sandoglobulin Guillain-Barre Syndrome Trial G. 2001. Preceding infections, immune factors, and outcome in Guillain-Barre syndrome. Neurology 56:758-765. Harley IT, Giles DA, Pfluger PT, Burgess SL, Walters S, Hembree J, Raver C, Rewerts CL, Downey J, Flick LM, Stankiewicz TE, McAlees JW, Wills-Karp M, Balfour Sartor R, Divanovic S, Tschop MH, Karp CL. 2013. Differential colonization with segmented filamentous bacteria and Lactobacillus murinus do not drive divergent development of diet-induced obesity in C57BL/6 mice. Molecular metabolism 2:171-183. Jacobs BC, Hazenberg MP, van Doorn PA, Endtz HP, van der Meche FG. 1997. Cross-reactive antibodies against gangliosides and Campylobacter jejuni lipopolysaccharides in patients with Guillain-Barre or Miller Fisher syndrome. The Journal of infectious diseases 175:729-733. Jacobs BC, Koga M, van Rijs W, Geleijns K, van Doorn PA, Willison HJ, Yuki N. 2008. Subclass IgG to motor gangliosides related to infection and clinical course in Guillain-Barre syndrome. Journal of neuroimmunology 194:181-190. Jacobs BC, Rothbarth PH, van der Meche FG, Herbrink P, Schmitz PI, de Klerk MA, van Doorn PA. 1998. The spectrum of antecedent infections in Guillain-Barre syndrome: a case-control study. Neurology 51:1110-1115. Keithlin J, Sargeant J, Thomas MK, Fazil A. 2014. Systematic review and meta-analysis of the proportion of Campylobacter cases that develop chronic sequelae. BMC public health 14:1203. Khader SA, Gaffen SL, Kolls JK. 2009. Th17 cells at the crossroads of innate and adaptive immunity against infectious diseases at the mucosa. Mucosal immunology 2:403-411. Koga M, Yuki N, Hirata K, Morimatsu M, Mori M, Kuwabara S. 2003. Anti-GM1 antibody IgG subclass: a clinical recovery predictor in Guillain-Barre syndrome. Neurology 60:1514-1518. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. 1993. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75:263-274. Kuroki S, Saida T, Nukina M, Haruta T, Yoshioka M, Kobayashi Y, Nakanishi H. 1993. Campylobacter jejuni strains from patients with Guillain-Barre syndrome belong mostly to 154 Penner serogroup 19 and contain beta-N-acetylglucosamine residues. Annals of neurology 33:243-247. Law BF, Adriance SM, Joens LA. 2009. Comparison of in vitro virulence factors of Campylobacter jejuni to in vivo lesion production. Foodborne pathogens and disease 6:377-385. Lee MD, Newell DG. 2006. Campylobacter in poultry: filling an ecological niche. Avian diseases 50:1-9. Linton D, Gilbert M, Hitchen PG, Dell A, Morris HR, Wakarchuk WW, Gregson NA, Wren BW. 2000. Phase variation of a beta-1,3 galactosyltransferase involved in generation of the ganglioside GM1-like lipo-oligosaccharide of Campylobacter jejuni. Molecular microbiology 37:501-514. Liu T, Matsuguchi T, Tsuboi N, Yajima T, Yoshikai Y. 2002. Differences in expression of toll-like receptors and their reactivities in dendritic cells in BALB/c and C57BL/6 mice. Infection and immunity 70:6638-6645. Louwen R, Horst-Kreft D, de Boer AG, van der Graaf L, de Knegt G, Hamersma M, Heikema AP, Timms AR, Jacobs BC, Wagenaar JA, Endtz HP, van der Oost J, Wells JM, Nieuwenhuis EE, van Vliet AH, Willemsen PT, van Baarlen P, van Belkum A. 2013. A novel link between Campylobacter jejuni bacteriophage defence, virulence and Guillain-Barre syndrome. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 32:207-226. Madsen KL, Doyle JS, Jewell LD, Tavernini MM, Fedorak RN. 1999. Lactobacillus species prevents colitis in interleukin 10 gene-deficient mice. Gastroenterology 116:1107-1114. Malik A, Sharma D, St Charles J, Dybas LA, Mansfield LS. 2014. Contrasting immune responses mediate Campylobacter jejuni-induced colitis and autoimmunity. Mucosal immunology 7:802- 817. Mansfield LS, Bell JA, Wilson DL, Murphy AJ, Elsheikha HM, Rathinam VA, Fierro BR, Linz JE, Young VB. 2007. C57BL/6 and congenic interleukin-10-deficient mice can serve as models of Campylobacter jejuni colonization and enteritis. Infection and immunity 75:1099-1115. Mansfield LS, Patterson JS, Fierro BR, Murphy AJ, Rathinam VA, Kopper JJ, Barbu NI, Onifade TJ, Bell JA. 2008. Genetic background of IL-10(-/-) mice alters host-pathogen interactions with Campylobacter jejuni and influences disease phenotype. Microbial pathogenesis 45:241-257. Martin RM, Brady JL, Lew AM. 1998. The need for IgG2c specific antiserum when isotyping antibodies from C57BL/6 and NOD mice. Journal of immunological methods 212:187-192. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. 2000. M-1/M-2 macrophages and the Th1/Th2 paradigm. Journal of immunology 164:6166-6173. 155 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. Moran AP, Annuk H, Prendergast MM. 2005. Antibodies induced by ganglioside-mimicking Campylobacter jejuni lipooligosaccharides recognise epitopes at the nodes of Ranvier. Journal of neuroimmunology 165:179-185. Nachamkin I, Allos BM, Ho T. 1998. Campylobacter species and Guillain-Barre syndrome. Clinical microbiology reviews 11:555-567. O'Mahony L, Feeney M, O'Halloran S, Murphy L, Kiely B, Fitzgibbon J, Lee G, O'Sullivan G, Shanahan F, Collins JK. 2001. Probiotic impact on microbial flora, inflammation and tumour development in IL-10 knockout mice. Alimentary pharmacology & therapeutics 15:1219-1225. Pena JA, Li SY, Wilson PH, Thibodeau SA, Szary AJ, Versalovic J. 2004. Genotypic and phenotypic studies of murine intestinal lactobacilli: species differences in mice with and without colitis. Applied and environmental microbiology 70:558-568. Poropatich KO, Walker CL, Black RE. 2010. Quantifying the association between Campylobacter infection and Guillain-Barre syndrome: a systematic review. Journal of health, population, and nutrition 28:545-552. Prendergast MM, Moran AP. 2000. Lipopolysaccharides in the development of the Guillain- Barre syndrome and Miller Fisher syndrome forms of acute inflammatory peripheral neuropathies. Journal of endotoxin research 6:341-359. Rees JH, Gregson NA, Hughes RA. 1995. Anti-ganglioside GM1 antibodies in Guillain-Barre syndrome and their relationship to Campylobacter jejuni infection. Annals of neurology 38:809- 816. Reiner SL, Locksley RM. 1995. The regulation of immunity to Leishmania major. Annual review of immunology 13:151-177. Samuelson DR, Eucker TP, Bell JA, Dybas L, Mansfield LS, Konkel ME. 2013. The Campylobacter jejuni CiaD effector protein activates MAP kinase signaling pathways and is required for the development of disease. Cell communication and signaling : CCS 11:79. Santini C, Baffoni L, Gaggia F, Granata M, Gasbarri R, Di Gioia D, Biavati B. 2010. Characterization of probiotic strains: an application as feed additives in poultry against Campylobacter jejuni. International journal of food microbiology 141 Suppl 1:S98-108. Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, Preibisch S, Rueden C, Saalfeld S, Schmid B, Tinevez JY, White DJ, Hartenstein V, Eliceiri K, Tomancak P, Cardona A. 2012. Fiji: an open-source platform for biological-image analysis. Nature methods 9:676-682. Schneider CA, Rasband WS, Eliceiri KW. 2012. NIH Image to ImageJ: 25 years of image analysis. Nature methods 9:671-675. Sheikh KA, Nachamkin I, Ho TW, Willison HJ, Veitch J, Ung H, Nicholson M, Li CY, Wu HS, Shen BQ, Cornblath DR, Asbury AK, McKhann GM, Griffin JW. 1998. Campylobacter jejuni 156 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. lipopolysaccharides in Guillain-Barre syndrome: molecular mimicry and host susceptibility. Neurology 51:371-378. St Charles JL, Bell JA, Gadsden BJ, Malik A, Cooke H, Van de Grift LK, Kim HY, Smith EJ, Mansfield LS. 2017. Guillain Barre Syndrome is induced in Non-Obese Diabetic (NOD) mice following Campylobacter jejuni infection and is exacerbated by antibiotics. Journal of autoimmunity 77:11-38. St Leger AJ, Desai JV, Drummond RA, Kugadas A, Almaghrabi F, Silver P, Raychaudhuri K, Gadjeva M, Iwakura Y, Lionakis MS, Caspi RR. 2017. An Ocular Commensal Protects against Corneal Infection by Driving an Interleukin-17 Response from Mucosal gammadelta T Cells. Immunity 47:148-158 e145. van den Berg B, Walgaard C, Drenthen J, Fokke C, Jacobs BC, van Doorn PA. 2014. Guillain- Barre syndrome: pathogenesis, diagnosis, treatment and prognosis. Nature reviews Neurology 10:469-482. 62. Walter J. 2008. Ecological role of lactobacilli in the gastrointestinal tract: implications for fundamental and biomedical research. Applied and environmental microbiology 74:4985-4996. 63. Watanabe H, Numata K, Ito T, Takagi K, Matsukawa A. 2004. Innate immune response in Th1- and Th2-dominant mouse strains. Shock 22:460-466. 64. Weaver CT, Harrington LE, Mangan PR, Gavrieli M, Murphy KM. 2006. Th17: an effector CD4 T cell lineage with regulatory T cell ties. Immunity 24:677-688. 65. WHO. [Internet]. 2016. Campylobacter Fact Sheet. Available at: http://www.who.int/mediacentre/factsheets/fs255/en/. 66. Wilson DL, Abner SR, Newman TC, Mansfield LS, Linz JE. 2000. Identification of ciprofloxacin- resistant Campylobacter jejuni by use of a fluorogenic PCR assay. Journal of clinical microbiology 38:3971-3978. 67. Wine E, Gareau MG, Johnson-Henry K, Sherman PM. 2009. Strain-specific probiotic (Lactobacillus helveticus) inhibition of Campylobacter jejuni invasion of human intestinal epithelial cells. FEMS microbiology letters 300:146-152. 68. 69. Yang F, Wang D, Li Y, Sang L, Zhu J, Wang J, Wei B, Lu C, Sun X. 2017. Th1/Th2 Balance and Th17/Treg-Mediated Immunity in relation to Murine Resistance to Dextran Sulfate-Induced Colitis. Journal of immunology research 2017:7047201. Yuki N, Taki T, Inagaki F, Kasama T, Takahashi M, Saito K, Handa S, Miyatake T. 1993. A bacterium lipopolysaccharide that elicits Guillain-Barre syndrome has a GM1 ganglioside-like structure. The Journal of experimental medicine 178:1771-1775. 157 CHAPTER 4: INVASION EFFICIENCY, INTRACELLULAR SURVIVAL, AND ELICITATION OF CYTOKINE PRODUCTION IN MURINE DENDRITIC CELLS IS DETERMINED BY BOTH CAMPYLOBACTER JEJUNI STRAIN CHARACTERISTICS AND MOUSE GENOTYPE To be submitted to: Microbial Pathogenesis ABSTRACT Campylobacter jejuni is an important cause of bacterial diarrhea worldwide and is associated with post- infectious Guillain-Barré syndrome (GBS). Immunological mechanisms underlying different disease outcomes are incompletely understood, but both infecting C. jejuni strain characteristics and host genetic background are thought to contribute. Dendritic cells (DCs) play a critical role in pathogen recognition and initiation of adaptive immunity by the polarization of naïve T helper (Th) cells. The objective of this study was to evaluate the interaction of colitogenic C. jejuni 11168 and GBS patient- derived C. jejuni 260.94 with murine bone-marrow derived DCs (BMDCs) from mice of two inbred genotypes. C57BL/6 and BALB/c mice were chosen due to reported Th1 and Th2 immunological biases, respectively. We hypothesized that BMDCs from wild-type (WT) and IL-10-/- mice on both C57BL/6 and BALB/c backgrounds would efficiently internalize and kill both C. jejuni strains, and that cytokines produced by infected IL-10-/- BMDCs would mirror adaptive immunity seen in previous in vivo mouse models: C. jejuni 11168-infected IL-10-/- BMDCs from both mouse genotypes and BALB/c IL-10-/- BMDCs infected with C. jejuni 260.94 would produce Th1/Th17-polarizing cytokines, while C. jejuni 260.94 would stimulate Th2-polarizing cytokine production in infected C57BL/6 IL-10-/- BMDCs. BMDC populations from C57BL/6 and BALB/c mice differed significantly: flow cytometry showed a more diverse population of differentiating DCs, with lower MHC II expression and higher proportions of macrophages, in C57BL/6 compared to BALB/c BMDCs. C. jejuni invasiveness and intracellular survival within BMDCs assessed by gentamicin killing assay differed significantly based on C. jejuni strain. Infection time with C. jejuni strains 11168 or 260.94 prior to gentamicin treatment and IL-10-/- BMDC lysis was varied to assess combined invasion efficiency and intracellular survival; to test intracellular survival alone, gentamicin was applied 158 after 1 hour of infection and WT or IL-10-/- BMDCs were lysed after 24 hours. C. jejuni 11168 showed significantly higher combined invasion/survival over time and intracellular survival alone compared to C. jejuni 260.94. A consistent mouse genotype difference in C. jejuni recovery was not found across all time points, but when a difference was identified, more C. jejuni was recovered from C57BL/6 BMDCs. BMDC cytokine profiles indicate that C57BL/6 IL-10-/- cells produced more MCP-1, especially following infection with C. jejuni 260.94, while BALB/c IL-10-/- cells produced more IL-6, especially when infected with C. jejuni 11168. Enhanced invasion and intracellular survival of C. jejuni 11168 in vitro correlated with higher in vivo pathogenicity and invasion than C. jejuni 260.94 in our mouse models. Production of MCP- 1 and IL-6 similarly mirrored Th2 and Th17 immune responses seen in vivo. These results support an important role of DCs in the initial host-microbe interaction and implicate invasiveness, intracellular survival, and elicitation of cytokine production of infecting C. jejuni strains as determinants of in vivo disease outcomes. 159 INTRODUCTION Campylobacter spp. are Gram negative, motile, spiral rods, and are considered the most frequent cause of bacterial gastroenteritis worldwide.61 Estimates from active surveillance data indicate that Campylobacter spp. cause over 800,000 cases of foodborne illness, and more than 8,000 hospitalizations and 70 deaths, annually in the United States.53 In addition to causing symptoms of enteritis including diarrhea and abdominal pain, Campylobacter infection has been associated with numerous severe sequelae, including inflammatory bowel disease and Guillain-Barré syndrome (GBS).30 GBS is a debilitating post-infectious neuropathy, with approximately 25% of patients requiring artificial ventilation, and many suffering from long-lasting residual pain and neurologic deficits.56 While several viral and bacterial agents are associated with GBS, campylobacteriosis is the most frequently reported preceding illness.21; 26 The immunopathogenesis of GBS subsequent to C. jejuni infection is incompletely understood, but is thought to involve generation of cross-reactive antibodies to the lipooligosaccharide (LOS) on the C. jejuni outer membrane and structurally similar peripheral nerve gangliosides.2; 43; 47; 56; 63 Patients with GBS frequently have antibodies to gangliosides GM1a, GM1b, GD1a, GT1a, and GQ1b, among others.48; 56 Rise in anti-ganglioside antibodies coincides with onset of paralysis, with subsequent waning of antibodies occurring with clinical improvement.56 GBS patients with preceding C. jejuni infection have been shown to have higher frequencies of antibodies to ganglioside GM1 and suffer worse clinical outcomes.21; 50 Subtype of anti-ganglioside IgG may also be related to type of preceding infection and clinical outcome of GBS. Anti-GM1 IgG1 and IgG1 antibodies to motor gangliosides have been associated with preceding gastroenteritis/diarrhea, positive C. jejuni serology, and slower recovery; in contrast, anti -GM1 or -ganglioside IgG3, alone or with IgG1, was associated with antecedent respiratory infection and better clinical outcome.25; 32 160 Current analyses estimate that GBS develops subsequent to Campylobacter infection in approximately 0.07% of cases.30 C. jejuni strain characteristics are thought to play a role in development of GBS, particularly due to the confirmed structural similarity between the LOS of some C. jejuni strains and gangliosides, including GM1 and GD1a,17; 63 and the overrepresentation of certain C. jejuni serotypes in GBS patients.1; 33 Multiple lines of epidemiological evidence suggest that host factors also contribute to GBS susceptibility, including the rarity of GBS outbreaks and geographic clustering of GBS subtypes, low proportion of C. jejuni infections resulting in GBS, isolation of C. jejuni strains harboring ganglioside mimics from both GBS and uncomplicated enteritis patients, and lack of identification of ganglioside mimics in some GBS-associated C. jejuni strains.3; 17; 30; 46; 47; 54; 56 Taken together, the evidence of C. jejuni molecular mimicry and host genetic features associated with GBS onset provides a rationale for further study into these mechanisms. Studies in mice and humans suggest that T helper (Th)1/Th17-driven immune responses predominate following C. jejuni infection, although Th2 responses can also occur. Multiple studies have demonstrated increased C. jejuni-specific plasma or serum IgG2a/IgG2c and IgG3 (Th1-mediated), along with IgG2b (Th1/Th17-mediated) in mice of various genetic backgrounds infected with different C. jejuni strains.16; 39; 40; 55 Th1/Th17 immune responses were similarly found in human intestinal biopsies infected with colitogenic C. jejuni in vitro; IFN-γ, along with more modest increases in IL-23, IL-12, IL-6, IL-17 and IL-1β, characterized the mucosal cytokine response.12 However, Th2-mediated plasma anti -C. jejuni, - GM1, and -GD1a IgG1 antibodies and colonic expression of Gata-3 and IL-4 were significantly increased in C57BL/6 IL-10-/- mice infected with GBS patient-derived C. jejuni 260.94,39 indicating that characteristics of the infecting C. jejuni strain impact Th polarization. Dendritic cells (DCs) are sentinel immune cells, bridging the innate and adaptive immune systems.57 DCs are important antigen presenting cells (APCs) that surveil peripheral tissues, and upon activation by antigen, migrate to a regional lymph node. DCs undergo several changes in this process in 161 preparation for interaction with naïve T cells, including increased expression of costimulatory molecules and MHC II, along with decreased phagocytosis of antigen. DCs contribute to Th polarization by direct interaction and cytokine production.57 Thus, DCs are a target for understanding the initial host-microbe interaction in C. jejuni infection and subsequent initiation of adaptive immunity. However, relatively few studies to date have investigated the role of DCs in C. jejuni pathogenesis. In vitro studies using DCs derived from human peripheral blood monocytes and mouse bone marrow have demonstrated efficient internalization and killing of C. jejuni by DCs, along with increased costimulatory molecule expression and production of pro-inflammatory and Th1-polarizing cytokines.23; 49 Furthermore, co-culture of C. jejuni-infected DCs with CD4+ T cells, or exposure of T cells to the supernatant of infected DCs, confirmed both Th1 and Th17 T cell polarization by production of IFN-γ and IL-17A.12; 49 Collectively, these studies show that Th1/Th17 responses can be initiated by DCs following interaction with C. jejuni, consistent with the isotypes of C. jejuni-specific antibodies described in mouse infection studies.16; 40; 55 However, Th1/Th17 responses are not the sole outcome of the adaptive response to C. jejuni, as demonstrated by a predominant Th2 response produced by C57BL/6 IL-10-/- mice infected with C. jejuni 260.94.39 Noteworthy is that the type of sialic acid linkage in the C. jejuni LOS can influence Th1 or Th2 polarization by DCs through binding of specific siglecs.5 These studies showed that C. jejuni adaptive responses are not limited to Th1/Th17 polarization, but that Th2 responses may also be triggered in the initial interaction of C. jejuni with DCs. Th1/Th17 adaptive immune responses to infection with colitogenic C. jejuni strains described in mouse and human studies12; 16; 40 are thought to correlate with C. jejuni clearance.16 However, production of antibodies cross-reacting to gangliosides is a hallmark of GBS. Anti-ganglioside, especially GM1, antibodies of the Th2-mediated IgG1 isotype have been identified in mice infected with GBS- associated C. jejuni 260.9439 and in GBS patients, correlate with poor clinical outcome.25; 32 Collectively, these studies suggest that contrary to a protective Th1/Th17 immune response, “aberrant” Th2 162 responses resulting from C. jejuni infection contribute to GBS development and that both C. jejuni strain characteristics and host factors contribute to susceptibility. Our rationale is that adaptive immunity is initiated following the host-microbe interaction and innate immune response to C. jejuni. Contrasting immune responses, likely influenced by both C. jejuni strain characteristics and host factors, can produce different disease outcomes following infection.39 Because DCs are potent APCs and contributors to Th cell polarization, a model system designed to evaluate the interaction of different C. jejuni strains with DCs derived from mice of different genetic backgrounds would further our understanding of how the interplay of bacterial and host factors influences disease outcome. The current study evaluated the interaction of colitogenic and GBS patient-derived C. jejuni strains with bone-marrow derived DCs (BMDCs) from mice of C57BL/6 and BALB/c backgrounds, with and without IL-10, to examine potential differences in cellular uptake, killing ability, and cytokine production. C. jejuni 11168 was originally isolated from an enteritis patient in the United Kingdom, is LOS class C, and encodes cst-III sialyl transferase. This strain bears both GM1 and GM2 mimics,37 but has not been reported to be associated with GBS. C. jejuni 11168 has repeatedly produced colitis mimicking human disease with induction of Th1/Th17-associated anti-C. jejuni antibodies in our C57BL/6 IL-10-/- mouse models.6; 39; 40; 52 Absence of anti-inflammatory IL-10 has been shown to be critical for induction of colitis, as C. jejuni 11168 stably colonizes wild-type (WT) mice without causing disease.9; 40; 41 The second strain included in these studies is C. jejuni 260.94. This strain was isolated from a GBS patient in South Africa. It is LOS class A, possesses a GM1a ganglioside mimic, and encodes cst-II sialyl transferase.37 This strain has colonized well in our mouse models without producing colitis,6; 9; 39; 55 and, in contrast to C. jejuni 11168, has induced a Th2-mediated response, including production of anti- ganglioside antibodies, in C57BL/6 IL-10-/- mice.39 163 C57BL/6 and BALB/c mice are two commonly used laboratory strains. Reports in the literature suggest that mice of these genetic backgrounds vary in innate, adaptive, and mucosal immune responses to various stimuli. One classic example is experimental infection studies with Leishmania major, in which susceptible BALB/c mice fail to suppress and even display overabundant IL-4 production, while resistant C57BL/6 mice downregulate IL-4 production and instead produce IFN-γ.51 BALB/c mice also reportedly harbor greater proportions of immunosuppressive T regulatory (T reg) cells in the small intestine and lymphoid organs than C57BL/6 mice,10; 18 and exhibit comparatively enhanced vitamin A metabolism, IgA production, and resistance to chemically induced colitis.18; 62 In vitro studies have shown important differences between the responses of stimulated spleen cells, peritoneal macrophages, and DCs derived from C57BL/6 and BALB/c mice. C57BL/6-derived spleen cells stimulated with concanavalin A produced more IFN-γ, while those from BALB/c mice produced higher levels of IL-4.42 Similarly, peritoneal macrophages from C57BL/6 mice stimulated with LPS and MALP-2 (TLR-4 and TLR-2 ligands, respectively) produced more TNF-α and IL-12 compared to BALB/c- derived macrophages.59 DCs derived from the spleens of naïve C57BL/6 and BALB/c mice differed in expression of various TLRs, and when stimulated with TLR-4, -2, and -9 ligands, C57BL/6-derived DC produced more IL-12p40 while those from BALB/c mice produced more MCP-1.35 Studies comparing C57BL/6 and BALB/c mice utilizing in vivo infection models combined with assessment of DC responses help to more completely elucidate differences in immune responses and disease outcomes. Similar to L. major infection, C57BL/6 mice are resistant to infection with Listeria monocytogenes whereas BALB/c mice are susceptible. Infected C57BL/6 mice showed increased survival, enhanced pathogen clearance, higher baseline and infection-induced IL-12 production in splenic DC, and increased IFN-γ production by splenic T cells compared to BALB/c mice.36 Similarly, C57BL/6 mice infected with the respiratory pathogen Chlamydia muridarum showed reduced morbidity, increased pathogen clearance, and greater production of IFN-γ in the lung and spleen compared to infected 164 BALB/c mice, while BALB/c mice produced more IL-17.28 When BMDCs from C57BL/6 and BALB/c mice were infected with C. muridarum, no difference was shown in endocytosis but BMDCs from BALB/c mice produced less IL-12 and more IL-23, IL-6, IL-10, and TNF-α than those from C57BL/6 mice.28 These studies highlight discrepancies in immune responses between C57BL/6 and BALB/c mice to various infectious agents, and suggest that these differences may begin with DCs. We hypothesized that 1) BMDCs from WT and IL-10-/- mice on both C57BL/6 and BALB/c genetic backgrounds would efficiently internalize and kill both colitogenic and GBS-associated C. jejuni strains, and 2) cytokines produced by infected BMDCs derived from IL-10-/- mice would mirror immunogenicity and pathogenicity previously observed in our mouse models: specifically, C. jejuni 11168-infected BMDCs from both BALB/c IL-10-/- and C57BL/6 IL-10-/- mice, and C. jejuni 260.94-infected BMDCs from BALB/c IL-10-/- mice, would produce pro-inflammatory and Th1/Th17-polarizing cytokines, while C57BL/6 IL-10-/- BMDCs infected with C. jejuni 260.94 would produce Th2-polarizing cytokines. Differentiation of bone marrow cells from both WT and IL-10-/- mice of C57BL/6 and BALB/c backgrounds into DCs was compared by flow cytometry. The gentamicin killing assay (GKA) was used to identify broad patterns in invasion efficiency and intracellular survival of the two C. jejuni strains in experiments using BMDCs derived from WT and IL-10-/- mice on both C57BL/6 and BALB/c backgrounds. BMDCs derived from WT mice were used as a comparison group to assess the impact of IL-10 on ability of BMDCs to effectively kill internalized C. jejuni. Finally, production of pro-inflammatory and Th- polarizing cytokines, including TNF-α, IFN-γ, IL-12p70, IL-4, MCP-1, TGF-β, and IL-6, by infected BMDC derived from IL-10-/- mice was assessed. Our results show that DCs differentiated from bone marrow precursors of C57BL/6 mice comprise a more diverse population of maturing DCs and a higher proportion of macrophages than those from BALB/c mice. C. jejuni 11168 exhibited increased invasion efficiency and intracellular survival in BMDCs compared to C. jejuni 260.94, but mouse genetic background did not have a consistent effect in each replicate on C. jejuni uptake or killing. BMDC 165 cytokine profiles indicated that pro-inflammatory and Th-polarizing cytokine production depends on both infecting C. jejuni strain and host genetic background, and that DCs are important early contributors to adaptive immunity following C. jejuni infection. MATERIALS AND METHODS Experimental Animals. Wild-type (WT) C57BL/6J, B6.129P2-IL-10tm1Cgn/J (referred to as C57BL/6 IL-10-/-), WT BALB/cJ, and C.129P2(B6)-Il10tm1Cgn/J (referred to as BALB/c IL-10-/-) mice were originally purchased from the Jackson Laboratories (Bar Harbor, ME). Mice were bred and maintained in-house; husbandry has been previously described.40 Briefly, mice were housed in sterile cages on a rack with filtered air and received sterile chow and water. The colony is monitored for enteric pathogens, and is free of Helicobacter spp. All protocols were approved by the Michigan State University (MSU) Institutional Animal Care and Use Committee (approval numbers 06/12-107-00 and 06/15-101-00). All mice were humanely euthanized using an overdose of CO2 according to the guidelines of the American Veterinary Medical Association (https://www.avma.org/KB/Policies/Pages/Euthanasia-Guidelines.aspx). Generation of Bone Marrow-Derived Dendritic Cells (BMDCs). Bone marrow stem cells were derived from the femurs of WT C57BL/6, WT BALB/c, C57BL/6 IL-10-/-, and BALB/c IL-10-/- mice of 2—5.5 months of age. The cells were isolated and differentiated according to published methods49 with modifications. Briefly, both femurs were removed, soaked in RPMI 1640 medium (Gibco; Waltham, MA), cleaned of muscles and tendons, surface sterilized in 70% ethanol, and rinsed in Hank’s Balanced Salt Solution (HBSS; Sigma, St. Louis, MO). Cells were flushed from the femur and through a 70 µm cell strainer, pelleted, treated with ACK Lysing Buffer (Gibco) to lyse erythrocytes, and washed twice with HBSS. Trypan Blue (Sigma, St. Louis, MO) staining was used to assess viability and cell numbers were estimated using a hemocytometer. Finally, cells were seeded onto polystyrene 100 × 15 mm sterile Petri dishes 166 (VWR; Radnor, PA) at a density of approximately 2.5  106 cells in 10 mL of R10 medium (RPMI 1640 with L-glutamine (Gibco), supplemented with 10% fetal bovine serum (FBS), 20 mM HEPES, 50 µM 2- mercaptoethanol, 100 u/mL penicillin, and 100 µg/mL streptomycin with 20 ng/mL recombinant murine granulocyte-macrophage colony stimulating factor (GMCSF; Peprotech, Rocky Hill, NJ)) and incubated at 37°C with 5% CO2. On day 3, non-adherent cells were discarded and fresh medium was added. On day 6, half of the medium was exchanged for fresh medium and non-adherent cells were again discarded. Half of the medium was again exchanged for fresh medium on day 8, but without discarding any cells. Morphology of the cells, including size, shape, adherence, and presence and size of cellular processes was observed until they were collected on days 9, 10, or 11 for assessment by flow cytometry, use in GKAs, or for assessment of cytokine production. Collection of adherent and non-adherent populations was achieved by collecting the medium and incubating the cells remaining in the Petri dish in 10 mL of Accutase (Sigma) for 20 minutes at 37°C with 5% CO2. Flow Cytometry. Multiple batches of cultured cells were analyzed by flow cytometry on days 9, 10, or 11 for confirmation of CD11c and MHC II expression over the course of time in which GKAs were performed. Flow cytometry was performed on cells derived from C57BL/6 IL-10-/- mice (once on days 9 and 11 of culture, twice on day 10), C57BL/6 WT mice (once each on days 9 and 10), BALB/c IL-10-/- mice (once on days 9 and 11, twice on day 10), and BALB/c WT mice (once on day 9, twice on day 10). In one preliminary experiment, cells derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice on days 10 and 9 of culture, respectively, were analyzed with and without stimulation with lipopolysaccharide (LPS). Flow cytometry was performed on extra cells one day prior to the GKA in two experiments. Cells from C57BL/6 10-/- and BALB/c IL-10-/- mice used in cytokine production experiments on day 10 were also assessed for expression of the F4/80 macrophage marker, in addition to CD11c and MHC II, on the day of the experiment. 167 For flow cytometric analysis, cells were harvested, resuspended in flow staining buffer (phosphate buffered saline (PBS) containing 1% bovine serum albumin and 0.09% sodium azide), and incubated in cocktails including LIVE/DEAD Fixable Far-Red Stain (Life Technologies, Carlsbad, CA), rat anti-mouse CD16/CD32 (Mouse BD Fc Block, clone 2.4G2; BD Biosciences, San Jose, CA), and antibodies including PE-conjugated anti-mouse CD11c (N418; eBioscience, Waltham, MA), FITC-conjugated anti- mouse I-A/I-E (2G9; BD Biosciences), and Pacific Blue-conjugated anti-mouse F4/80 (BM8; BioLegend, San Diego, CA). Controls included incubation of cells in flow staining buffer only (unstained controls) or with appropriate antibody isotypes. Following incubation for 30 minutes covered on ice, the cells were washed twice with flow staining buffer and analyzed with a BD FACSCanto II (BD Biosciences) flow cytometer. Compensation matrices were established and populations were gated using FlowJo software version 10.2 (Becton, Dickinson & Company, Franklin Lakes, NJ). Preparation of Campylobacter jejuni Inocula. Frozen glycerol stock cultures of minimally passaged C. jejuni strain 11168 (ATCC 700819) and C. jejuni strain 260.94 (ATCC BAA-1234) were streaked onto Bolton agar (BA) plates and grown in microaerophilic conditions for approximately 48 (range: 41-49.5) hours at 37°C. The microaerophilic environment was achieved by evacuation of vented anaerobic jars to -25 in Hg and equilibration with a gas mixture comprising 80% N2, 10% CO2, and 10% H2. Colonies were suspended in tryptic soy broth (TSB) to an optical density of approximately 0.2-0.3 at 560 nm (OD560). Lawns (100 µL) made from this suspension were grown on BA plates for approximately 18-24 hours in microaerophilic conditions at 37°C. Immediately prior to use, bacterial growth was suspended in R10.1 medium (RPMI 1640 with L-glutamine, supplemented with 10% FBS and 20 mM HEPES) to an OD560 of ~0.120, corresponding to approximately 2 × 108 colony forming units (CFU)/mL. Wet mounts of both cultures were performed with each experiment to assess purity, morphology, and motility. CFU in the inocula were verified by limiting dilution in TSB and plating in single or duplicate on BA plates. 168 Gentamicin Killing Assays. GKAs were performed using published methods49; 58 with modifications. BMDCs were harvested on days 10 or 11 and resuspended in R10.1 medium at a concentration of 2 × 105 cells/mL. Cells were kept on ice until just prior to use, when 1 mL (2 × 105 cells) of cell suspension was added to each well of a 24-well cell culture plate. One hundred microliters (aiming to achieve approximately 2 × 107 CFU) of C. jejuni culture was immediately added to appropriate wells to achieve a multiplicity of infection (MOI) of approximately 100. Cells were incubated at 37°C with 5% CO2 throughout the assay. BMDCs were not removed from the original 24-well plate at any point during the assay until lysis. The experimental design for the GKAs was as follows. Each mouse genotype/C. jejuni strain combination was assayed in triplicate wells, with four experimental groups: 1) C57BL/6 BMDCs, C. jejuni 11168; 2) C57BL/6 BMDCs, C. jejuni 260.94; 3) BALB/c BMDCs, C. jejuni 11168; 4) BALB/c BMDCs, C. jejuni 260.94. Killing of both C. jejuni strains by gentamicin was verified by addition of 100 µL of C. jejuni 11168 or 260.94 inocula to separate wells containing only R10.1 medium, with and without subsequent gentamicin treatment. Additional duplicate control wells to monitor any cross-contamination during the assay included sham inoculation (R10.1 medium only) of wells containing BMDCs, with and without subsequent application of gentamicin. Aliquots of cell culture suspensions were spread onto BA plates at the beginning of the assay to screen for any initial contamination. The cells were visualized at multiple time points throughout the assay to assess morphology, viability, and adherence, and following lysis and harvesting to verify removal of well contents. A control experiment to test viability of both C. jejuni strains in 0.1% Triton X-100 (Sigma) in PBS in a closed Eppendorf tube at room temperature showed relatively constant viability at 30, 60, 90, and 120 minutes of incubation. To assess combined invasion efficiency and intracellular survival of the two C. jejuni strains, cells derived from IL-10-/- mice of both C57BL/6 and BALB/c genetic backgrounds were infected for 1, 2, 3, or 23 hours. Following incubation at 37°C with 5% CO2, medium was discarded and replaced with R10.1 169 medium containing 250 µg/mL of gentamicin (Life Technologies) for 1 hour. Medium was discarded again, and cells were washed twice with R10.1 medium to remove gentamicin before lysis. Each time point was assessed by individual experiments. To assess intracellular survival alone, cells derived from IL-10-/- or WT mice were infected for 1 hour. Medium was removed and replaced with R10.1 medium containing 250 µg/mL of gentamicin for 1 hour. Cells were washed twice and incubated with fresh R10.1 medium at 37°C with 5% CO2 until lysis at 24 hours post-infection (p.i.). In both invasion and survival assays, cells were lysed by incubation in 500 µL of PBS containing 0.1% Triton X-100 for 15 minutes at room temperature. Well contents were removed by scraping the bottom with a pipette tip, pipetting up and down, and transferring the contents to a microfuge tube. Undiluted lysate and/or appropriate dilutions made in R10.1 medium were then spread on BA plates. C. jejuni CFU were enumerated following approximately 72 hours incubation in microaerophilic conditions at 37°C. Assessment of Cytokine Production by C. jejuni-Infected BMDCs Derived From IL-10-/- Mice. Cells derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were harvested on day 10 of differentiation and resuspended in R10.1 medium to a concentration of 2 × 105 cells/mL. Cells were kept on ice until just prior to use, when 1 mL (2 × 105 cells) of cell suspension was added to each well of a 24-well cell culture plate. The experimental design was as follows: BMDCs from each mouse strain were infected with C. jejuni 11168 or 260.94 (MOI of approximately 100), or treated with 1 µg/mL LPS from Salmonella enterica serotype Typhimurium (Sigma) (positive control), or R10.1 medium (negative control). The eight groups were as follows: 1) C57BL/6 BMDCs, C. jejuni 11168; 2) C57BL/6 BMDCs, C. jejuni 260.94; 3) BALB/c BMDCs, C. jejuni 11168; 4) BALB/c BMDCs, C. jejuni 260.94; 5) C57BL/6 BMDCs, R10.1 medium; 6) C57BL/6 BMDCs, LPS; 7) BALB/c BMDCs, R10.1. medium; and 8) BALB/c BMDCs, LPS. Each mouse 170 genotype/C. jejuni strain combination was performed in triplicate, as were positive and negative controls. CFU in the inocula were verified by limiting dilution. Wet mount preparations to evaluate purity, morphology, and motility were performed with each experiment. Infected BMDCs were incubated for 1 hour at 37°C with 5% CO2. Medium was then removed, and replaced with 1 mL R10.1 medium containing 250 µg/mL of gentamicin. Following 1 hour of incubation at 37°C with 5% CO2, medium was again removed. The plate was washed once to remove gentamicin by adding and removing 1 mL of R10.1 medium to each well. Finally, 1 mL of fresh R10.1 was added to each well, and the plate was incubated at 37°C with 5% CO2 until approximately 24 hours p.i. Medium was collected from wells, aliquoted into cryovials, and stored at -80°C until analysis. Cytokine production by IL-10-/- BMDCs was assessed by a custom multiplex bead-based assay panel (LEGENDplex Custom Panel; BioLegend, San Diego, CA) according to the manufacturer’s instructions. The panel was designed to measure pro-inflammatory cytokines and those involved in polarization of Th1, Th2, and Th17 cells by stimulated dendritic cells. The seven analytes included TNF-α, IFN-γ, IL-12p70, IL-4, MCP-1, IL-6, and TGF-β. Data were acquired using a BD FACSCanto II flow cytometer and analyzed using LEGENDplex Data Analysis Software (BioLegend) according to the manufacturer’s guidelines. Statistical Analyses. Analyses were performed and figures generated using commercial statistical software packages (SigmaStat 3.5, Systat Software, Inc., San Jose, CA; and GraphPad Prism 6, GraphPad Software, La Jolla, CA). Assumptions of normality and equal variance were tested prior to each analysis. P-values ≤0.05 were considered significant. Flow Cytometry. Statistical analyses were performed on data from cells used in the 3 independent cytokine production experiments, which included CD11c, MHC II, and F4/80 markers. t- 171 tests were performed to determine if there was a difference between C57BL/6 IL-10-/- and BALB/c IL-10-/- cells in %CD11c(+) cells; %CD11c(+)/MHC IIhigh cells; %CD11c(+)/MHC IIlow cells; %CD11c(+)/F4/80(+) cells; and the median fluorescence intensity (MFI) of MHC II in populations including all MHC II(+) cells and within CD11c(+)/F4/80(+) cells. For percentage data, t-tests were performed on Arcsin Square Root transformed data. Gentamicin Killing Assays. Statistical analyses were performed on data from individual experiments from each time point. In experiments with both C. jejuni strains and both mouse strains, two-way ANOVAs were performed. When normality or equal variance assumptions were not met or a significant interaction was identified, a Kruskal-Wallis ANOVA on ranks or one-way ANOVA was also performed. Post-hoc testing, including Holm-Sidak, Tukey, Dunn’s, Bonferroni t-test, and Student- Newman-Keuls methods for pairwise multiple comparisons, was performed as appropriate. Data from experiments with two mouse strains but only one C. jejuni strain were tested for normality and equal variance and subjected to a t-test or Mann Whitney rank sum test as appropriate. Cytokine Production. The seven analytes assessed in each experiment were analyzed individually by 2-way ANOVA. Holm-Sidak pairwise testing was implemented when overall statistical significance in mouse genotype (C57BL/6, or BALB/c), treatment group (positive or negative control, C. jejuni 11168, or C. jejuni 260.94), or an interaction was identified. One-way ANOVA followed by Holm-Sidak pairwise testing was additionally performed in some cases to further evaluate pairwise comparisons. Data not meeting equal variance assumptions were analyzed by Kruskal-Wallis one-way ANOVA on ranks followed by Tukey’s post-test. RESULTS Bone Marrow-Derived Dendritic Cell Cultures. Cells were analyzed by flow cytometry on days 9, 10, or 11 of culture, used in GKAs on day 10 (all experiments except 2, which were performed on day 11), or 172 used to measure cytokine production on day 10 (3 experiments). Cells were occasionally seeded at lower densities in 10 mL of medium at day 0 if sufficient numbers were not obtained from bone marrow. Cultures used in most GKAs analyzed statistically had a starting density of isolated bone marrow stem cells of approximately 2.5 × 106 in 10 mL medium, with a range of 1.5  106 - 3.2  106 (see Tables 4.2A- D and Tables 4.3A-B). A concentration of 2  105 cells/mL was always used on the day of the assay. Morphology was observed on days 3, 6, and 8 of differentiation and the day of the experiment. Subjectively, cells from C57BL/6 and BALB/c mice were typically observed to have slightly different morphology during differentiation, culminating in cells from C57BL/6 appearing to have a greater proportion of larger, firmly adherent cells, and a lower proportion of smaller, round, semi-adherent cells when fully differentiated than cells from BALB/c mice. Cells from BALB/c mice also had a layer of larger, firmly adherent cells, but these cells typically appeared smaller than the corresponding population of adherent cells in C57BL/6 mice. Cells from BALB/c mice also had a relatively greater proportion of smaller, round, semi-adherent cells than those from C57BL/6 mice. Presence or absence of IL-10 within a mouse strain did not obviously change cellular morphology. Criteria for Statistical Analyses. Flow Cytometry. Flow cytometry data generated from analysis of CD11c, MHC II, and F4/80 expression in BMDCs from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice used in the 3 independent cytokine production experiments were analyzed statistically (Table 4.1, Figure 4.2). Scatter plots from day 10 of differentiation of IL-10-/- and WT BMDCs from C57BL/6 and BALB/c mice are presented for descriptive purposes (Figure 4.1), but comparisons of differentiation in WT and IL-10-/- cells or differences over days 9-11 of culture were not performed due to insufficient replicates of these experiments for statistical analyses. 173 Gentamicin Killing Assays: Invasion and Intracellular Survival. A total of 17 experiments analyzed statistically had the following characteristics: performance on day 10 or 11 of BMDC culture; inclusion of BMDCs from both mouse genotypes, with starting densities (isolation from marrow on day 0) above 8  105 cells in 10 mL (the lowest starting density of cells assessed by flow cytometry in other experiments); and confirmation of MOIs by limiting dilution between approximately 60-175 calculated from a dilution factor of 10-6 (MOI ranges from limiting dilution plates given with data in Tables 4.2 and 4.3). When an MOI of ~60-175 was achieved for only one C. jejuni strain, data regarding the other C. jejuni strain were excluded from analysis. An exception to the MOI criterion was made regarding the two GKAs assessing 23 hours of invasion time, in which the MOI for C. jejuni 11168 was ~44 and ~33. The exception was made due to the prolonged incubation in medium with a protein source (10% FBS) and conditions (37°C, 5% CO2) favorable for C. jejuni growth. Masses of clumped structures exhibiting movement, interpreted to be C. jejuni organisms, were visualized in inoculated wells at the end of the 23-hour invasion period in both of these experiments. Additionally, C. jejuni 11168 exhibited a several-fold increase in growth in similar conditions over 8 hours.49 Cytokine Production. Eight independent experiments including BMDCs derived from both C57BL/6 IL-10-/- and BALB/c IL-10-/- mice and including both C. jejuni strains were performed; medium was collected and stored pending evaluation of MOI and flow cytometric data. The three experiments selected for measurement of cytokine production and statistical analysis all had confirmation of MOIs of both C. jejuni strains between 60-175 by limiting dilution and flow cytometric analysis on the day of the experiment for CD11c, MHC II, and F4/80 expression (Table 4.1, Tables 4.4-4.6). Flow Cytometry. CD11c and MHC II Expression in BMDCs Derived from WT and IL-10-/- Mice. Cells derived from WT and IL-10-/- mice on C57BL/6 and BALB/c genetic backgrounds were analyzed by flow cytometry 174 between days 9-11 of culture. Representative plots of CD11c and MHC II expression from one WT and one IL-10-/- mouse of each background, all on day 10 of differentiation, are shown in Figure 4.1. Presence or absence of IL-10 did not markedly influence patterns of differentiation. Percentage of CD11c(+) cells was generally similar between genotypes (WT or IL-10-/-, C57BL/6 or BALB/c). Within the CD11c(+) positive cells, C57BL/6 mice showed 4 distinct smaller populations, while BALB/c mice had 2 distinct larger populations. The Median Fluorescence Intensity of MHC II, in all MHC II positive cells (MFI MHC II), was consistently higher in BALB/c than C57BL/6 mice, with or without IL-10 (Figure 4.1, and Table 4.1). A preliminary experiment assessing LPS stimulation of cells derived from C57BL/6 IL-10-/- and BALB/c IL-10- /- mice on days 10 and 9 of culture, respectively, showed that LPS increased the percentage of CD11c(+)/MHC IIhigh cells in both mouse strains, although the increase was proportionately greater in cells from C57BL/6 IL-10-/- than BALB/c IL-10-/- mice (data not shown). CD11c, MHC II, and F4/80 Expression in BMDCs Derived from IL-10-/- Mice. Flow cytometry was performed on cells derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice used in the 3 independent cytokine experiments. Plots from one experiment are shown in Figure 4.2. Percentages of CD11c(+) cells, CD11c(+)/F4/80(+) cells, CD11c(+)/MHC IIhigh cells, CD11c(+)/MHC IIlow cells, and MHC II MFI were analyzed by t-test, following transformation of percentage data. Descriptive statistics (mean ± SD) and results of statistical analyses are given in Table 4.1. Cells derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice did not differ significantly in percentage of CD11c(+) cells (P = 0.111), or CD11c(+)/MHC IIhigh cells (P = 0.063). However, C57BL/6 IL-10-/- mice had significantly more CD11c(+)/MHC IIlow cells than BALB/c IL- 10-/- mice (57% versus 26% respectively; P = 0.03). Differences were also identified between mouse strains in proportion of macrophages in the differentiated populations, indicated by CD11c and F4/80 double positivity. Although not statistically significant, C57BL/6 IL-10-/- mice trended toward producing higher percentages of CD11c(+)/F4/80(+) cells than BALB/c IL-10-/- mice (26% versus 16% respectively; P = 0.054). Despite a trend toward lower numbers of macrophages in BALB/c IL-10-/- cultures, the 175 difference in MHC II MFI of CD11c(+)/F4/80(+) cells was significant, with BALB/c IL-10-/- mice having higher MHC II MFIs in this population than C57BL/6 IL-10-/- mice (1,169 versus 557 respectively; P = 0.05). Finally, a significant difference was found (P < 0.001) in the MFI of MHC II within all MHC II(+) cells, with BALB/c IL-10-/- cells exhibiting higher expression than C57BL/6 IL-10-/- populations (11,933 versus 3,202 respectively). These data show that there may be subtle differences in the differentiation of stem cells derived from bone marrow of C57BL/6 IL-10-/- and BALB/c IL-10-/- mice with regard to proportion of macrophages and expression of MHC II. The differentiated cells are considered to comprise a majority of DCs with fewer macrophages, and are referred to as BMDCs with acknowledgement that lower numbers of macrophages are also present. Invasion Assays: BMDCs Derived from IL-10-/- Mice. Individual experiments assessing combined invasion efficiency and intracellular survival of the C. jejuni strains within BMDCs derived from IL-10-/- mice on a C57BL/6 or BALB/c genetic background were performed with 1, 2, 3, and 23 hours of infection prior to application of gentamicin. Statistically analyzed experiments are described in Tables 4.2A-D. One representative, independent experiment from each time point that includes both C. jejuni strains is shown graphically (Figure 4.3). Because each time point was assessed in a separate experiment, differences in C. jejuni recovery over time could not be assessed statistically. C. jejuni Strain Differences. No statistically significant difference was identified between recovery of C. jejuni 11168 and C. jejuni 260.94 after 1 hour of infection time (P = 0.466, Table 4.2A). However, significantly more C. jejuni 11168 than C. jejuni 260.94 was recovered from BMDCs derived from both IL-10-/- mouse genotypes after 2, 3, and 23 hours of infection time (P ≤ 0.001 for each time point; repeatable between independent experiments; Tables 4.2B-D). The magnitude of difference between C. jejuni strain recoveries became more pronounced with increased infection time (Figure 4.3). 176 Mouse Genotype Differences. A statistically significant difference in C. jejuni recovery was identified between BMDCs derived from the two mouse genotypes after 1 hour of infection time, with more of both C. jejuni strains recovered from BMDCs obtained from C57BL/6 IL-10-/- than BALB/c IL-10-/- mice (P = 0.003; Table 4.2A). Significantly more C. jejuni 11168 but not 260.94 was similarly recovered from C57BL/6 IL-10-/- than BALB/c IL-10-/- BMDCs infected for 3 hours (P = 0.002; Table 4.2C). There was a trend of more C. jejuni 260.94 recovered from BALB/c IL-10-/- than C57BL/6 IL-10-/- BMDCs after 23 hours of infection (Table 4.2D), but the difference was not statistically significant in both experiments and repeatable mouse genotype differences in C. jejuni 260.94 were not identified at other time points. Collectively, these data show that C. jejuni 11168 has higher invasion efficiency and/or intracellular survival compared to C. jejuni 260.94, beginning at 2 hours of infection and becoming more marked with increased infection time. A mouse genotype difference was identified early in infection, but with the exception of differences in C. jejuni 11168 recovery between genotypes at 3 hours of infection, a clear difference in mouse genotypes was not found at later time points. Intracellular Survival Assays: BMDCs Derived from WT and IL-10-/- Mice. Experiments assessing intracellular survival of the C. jejuni strains within BMDCs derived from WT and IL-10-/- mice on a C57BL/6 or BALB/c genetic background were performed with 1 hour of infection prior to gentamicin application. Cells were then washed and incubated with fresh medium until lysis and enumeration of C. jejuni CFU at 24 hours p.i. Experiments involving BMDCs from WT and IL-10-/- mice were performed separately. Details of the experiments and statistical results are described in Table 4.3A (WT mice) and Table 4.3B (IL-10-/- mice). Results of one representative experiment with WT BMDCs and both C. jejuni strains are shown in Figure 4.4A, and results of one representative experiment with IL-10-/- BMDCs and both C. jejuni strains are shown in Figure 4.4B. The experiments are combined graphically in Figure 4.4C. 177 C. jejuni Strain Differences. In BMDCs derived from WT mice, significantly more C. jejuni 11168 was recovered than C. jejuni 260.94 (P < 0.001) within BMDCs derived from both mouse genotypes (Table 4.3A, Figure 4.4A). A similar pattern was identified in IL-10-/- mice, with significantly more C. jejuni 11168 recovered than C. jejuni 260.94 (P < 0.001); however, this C. jejuni strain difference was only significant in BMDCs from C57BL/6 IL-10-/- mice, while BALB/c IL-10-/- BMDCs showed a nonsignificant trend toward higher C. jejuni 11168 than 260.94 recovery (Table 4.3B, Figure 4.4B). Mouse Genotype Differences. No difference in C. jejuni recovery between cells derived from WT C57BL/6 and WT BALB/c mice was identified in the experiment shown in Figure 4.4A (P = 0.637; Table 4.3A). This experiment showing no difference in genotype was chosen as representative in part because of conflicting results for C. jejuni 11168 recovery between genotypes in the other two experiments (Table 4.3A). In contrast, a genotype effect was found to be significant (P < 0.001; Table 4.3B) in cells derived from IL-10-/- mice, with significantly more C. jejuni 11168 recovered from C57BL/6 IL-10-/- than BALB/c IL-10-/- derived BMDCs (Figure 4.4B). Recovery of C. jejuni 260.94 was not significantly different between mouse genotypes, as virtually no C. jejuni 260.94 CFU were recovered from BMDCs of either C57BL/6 IL-10-/- or BALB/c IL-10-/- mice in either experiment (Table 4.3B). Taken together, these data show that C. jejuni 11168 exhibits enhanced intracellular survival within BMDCs compared to C. jejuni 260.94. This was observed in cells derived from both WT and IL-10-/- mice on both C57BL/6 and BALB/c genetic backgrounds. A difference in C. jejuni recovery between mouse genotypes was observed in recovery of C. jejuni 11168 only in BMDC derived from IL-10-/- mice. Although C. jejuni survival cannot be compared statistically between WT and IL-10-/- derived cells, the patterns suggest that C. jejuni exhibits reduced survival capacity within IL-10-/- cells, a difference that was more marked in C. jejuni 260.94. 178 Cytokine Production: BMDCs Derived from IL-10-/- Mice. Three independent experiments were performed to measure pro-inflammatory and Th-polarizing cytokine production by BMDCs derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with C. jejuni 11168 or C. jejuni 260.94. Statistical results of the three experiments are detailed in Table 4.4 (Experiment 1), Table 4.5 (Experiment 2), and Table 4.6 (Experiment 3). Results of Experiment 2 are displayed graphically (Figure 4.5) and described below. Levels of neither IFN-γ nor TGF-β were significantly different between any treatment groups, including negative (medium only) and positive (LPS) controls, in BMDCs derived from either C57BL/6 IL- 10-/- or BALB/c IL-10-/- mice. Production of pro-inflammatory TNF-α was significantly increased in BMDCs derived from both C57BL/6 IL-10-/- and BALB/c IL-10-/- mice infected with C. jejuni 260.94, but not C. jejuni 11168, compared to the negative control. The Th1-polarizing cytokine IL-12 was not significantly increased in BMDCs derived from either mouse genotype infected with either C. jejuni 11168 or 260.94; however, BMDCs derived from C57BL/6 IL-10-/- mice exhibited a significantly higher baseline production of IL-12 in uninfected cells compared to BMDCs derived from BALB/c IL-10-/- mice. Similarly, BMDCs derived from C57BL/6 IL-10-/- mice exhibited a significantly higher baseline production of the Th2- polarizing cytokine IL-4, compared to BMDCs derived from BALB/c IL-10-/- mice. IL-4 production was significantly decreased in C57BL/6 IL-10-/- BMDCs infected with C. jejuni 11168 compared to the negative control; in contrast, C. jejuni 11168-infected BALB/c IL-10-/- BMDCs trended toward increased production of IL-4 compared to the negative control. BMDCs from C57BL/6 IL-10-/- mice showed significantly increased MCP-1 production following exposure to LPS and infection with C. jejuni 260.94 compared to BALB/c IL-10-/- BMDCs receiving the same treatments. Within C57BL/6 IL-10-/- BMDCs, production of MCP-1 in C. jejuni 260.94-infected cells was significantly increased compared to exposure to medium only or infection with C. jejuni 11168; infection with either C. jejuni strain did not lead to significantly increased MCP-1 levels in infected BALB/c IL-10-/- BMDCs. Finally, IL-6 production was more marked in 179 BALB/c IL-10-/- than C57BL/6 IL-10-/- BMDCs: BALB/c-IL-10-/--derived cells exposed to LPS or infected with C. jejuni 11168 showed significantly higher IL-6 levels than C57BL/6 IL-10-/- cells receiving these treatments. Additionally, production of IL-6 following infection with C. jejuni 11168 was significantly increased in BALB/c IL-10-/- BMDCs compared to the negative control, but the difference was not significant in C57BL/6 IL-10-/- cells. Collectively, these data indicate that production of pro-inflammatory and Th-polarizing cytokines by C. jejuni-infected IL-10-/- BMDCs is determined by both infecting C. jejuni strain and mouse genetic background. DISCUSSION This study was conducted to evaluate the role of dendritic cells in initiation of innate and adaptive immunity following C. jejuni infection. The model was designed to elucidate the potential impacts of both C. jejuni strain characteristics and host genetic background on early DC-mediated determinants of eventual disease outcome, such as colitis or development of GBS following C. jejuni infection. Results of this study suggest that differentiation of cultured bone marrow stem cells into dendritic cells varies between C57BL/6 and BALB/c mice, most notably in MHC II expression and proportion of macrophages present. The colitogenic C. jejuni 11168 strain displayed enhanced invasion efficiency and intracellular survival compared to the Guillain-Barré syndrome patient-derived strain C. jejuni 260.94 in cultured BMDCs, while differences in C. jejuni uptake or killing were less consistent between mouse genotypes. Cytokine production by infected IL-10-/- BMDCs varied by both mouse genotype and infecting C. jejuni strain: a more marked response in MCP-1 from C57BL/6 IL-10-/- cells was seen, especially following infection with C. jejuni 260.94, while IL-6 production was more pronounced in BALB/c IL-10-/- cells, especially when infected with C. jejuni 11168. Flow cytometric analysis of BMDCs on days 9, 10, and 11 of culture from WT and IL-10-/- mice on both C57BL/6 and BALB/c backgrounds was performed with the original intent of verifying the BMDC 180 phenotype by CD11c and MHC II expression. These analyses revealed unexpected variation in differentiated populations related to genetic background, and assessment of F4/80 expression was subsequently evaluated. While not different from BALB/c IL-10-/- cells in overall CD11c positivity, cells derived from C57BL/6 IL-10-/- mice trended toward fewer CD11c(+)/MHC IIhigh cells, with correspondingly more CD11c(+)/MHC IIlow cells, and a greater proportion of CD11c(+)/F4/80(+) cells in differentiated populations compared to BALB/c IL-10-/- cells. Macrophages can be found intermixed with immature and mature DCs in populations of bone marrow stem cells cultured with GM-CSF,24; 38 and the proportion of macrophages compared to DCs generated in culture can be increased with higher concentrations of GM- CSF.45 However, to our knowledge, differences in proportions of macrophages in DC populations from the bone marrow of C57BL/6 and BALB/c mice cultured by the same method has not been previously reported. Subjective observations of morphology during differentiation are consistent with discrepancies in MHC II and F4/80 expression identified by flow cytometry. Both WT and IL-10-/- C57BL/6 cells appeared to have a relatively greater proportion of larger, firmly adherent cells (macrophage type morphology) and a relatively smaller proportion of smaller, rounder, semi-adherent cells (dendritic cell type morphology),38 compared to both WT and IL-10-/- BALB/c cultures. Although CD11c expression was until recently or is still used to distinguish DCs from macrophages,44 data from the current study corroborate the relatively recent realization that CD11c expression is not limited to DCs; F4/80+ monocytes in the intestine also acquire CD11c expression during maturation to macrophages, while retaining F4/80 expression.29 The other notable difference between cultured populations derived from WT and IL-10-/- mice on both C57BL/6 and BALB/c backgrounds was the higher expression of MHC II in BALB/c populations. As MHC II expression increases with DC maturation, this likely reflects greater numbers of more mature DCs in BALB/c than C57BL/6 cultures. This finding is inconsistent with that reported for DCs derived from spleens of naïve C57BL/6 and BALB/c mice, in which C57BL/6 mice contained more mature DC subsets 181 based upon expression of CD40 and Stat4.35 This discrepancy may be attributable to isolation methods and origin of DC, as culture of hematopoietic precursors from the marrow is likely to produce DCs with a different phenotype than those already fully differentiated in a naïve spleen. Differences in diet or microbiome may also contribute. In the current study, a pattern in which four smaller populations can be distinguished within WT and IL-10-/- C57BL/6 CD11c(+) cells, while two distinct larger populations are evident in WT and IL-10-/- BALB/c CD11c(+) cells (Figure 4.1), was repeatedly observed. Considering the F4/80 and MHC II expression together with the patterns exhibited in representative scatter plots, we speculate that bone marrow precursors from the two mouse genotypes exhibit varying kinetics of maturation under the influence of GM-CSF. F4/80 and MHC II expression suggest that populations from C57BL/6 IL-10-/- mice develop a higher proportion of macrophages and immature DCs in varying stages of maturation, while BALB/c IL-10-/- cells differentiate with a higher proportion of mature DCs following 10 days of culture with GM-CSF. Haase et al (2002) similarly found a predominance of immature BMDCs derived from C57BL/6 mice in unstimulated day 8 cultures; these cells responded to LPS stimulation with a conversion to MHC II expression in nearly all cells.20 In the current study, results of a preliminary experiment showed relatively greater upregulation of MHC II expression in BMDCs derived from C57BL/6 IL-10-/- than BALB/c IL-10-/- mice following stimulation with LPS. This finding also supports a higher baseline of more mature DCs in cultures from BALB/c IL-10-/- mice, but further replicates of this experiment are warranted to confirm this result. In populations of differentiated but unstimulated cells from C57BL/6 and BALB/c mice, the presence or absence of IL-10 did not markedly affect CD11c or MHC II expression (Figure 4.1). Previous in vitro studies have demonstrated potent effects of IL-10 on DC maturation, cytokine secretion, and T- cell stimulating ability. Neutralization of IL-10 resulted in enhanced maturation and TNF-α and IL-12 production by LPS-stimulated human peripheral blood monocyte-derived DC.11 Furthermore, these DCs 182 exhibited spontaneous maturation in vitro in the absence of IL-10.11 Similarly, administration of IL-10 effectively inhibited LPS-induced DC maturation, production of pro-inflammatory cytokines, and ability to drive T cell proliferation in BMDCs derived from C57BL/6 mice.20 These studies indicate that IL-10 has potent inhibitory effects on DC maturation and T-cell stimulating ability following LPS stimulation. It is possible that a difference in MHC II expression between WT and IL-10-/- BMDCs could be induced with LPS or other stimulation, but in unstimulated cells the presence or absence of IL-10 did not affect MHC II expression. Because DCs are potent antigen presenting cells and play a major role in initiation of the adaptive immune response through T cell activation, we next wanted to determine the effects of C. jejuni strain characteristics and host genetic background in the initial C. jejuni-DC interaction. The GKA was used to test the original hypothesis that BMDCs from WT and IL-10-/- mice on both C57BL/6 and BALB/c genetic backgrounds would efficiently internalize and kill both C. jejuni strains; this method was chosen as a way to identify broad patterns in invasion and survival for further study in more detail. In assays with increasing infection time of IL-10-/- BMDCs, the colitogenic C. jejuni 11168 showed enhanced combined invasion efficiency and intracellular survival compared to GBS patient-derived C. jejuni 260.94, beginning at 2 hours of infection time and becoming more marked as infection time increased up to 23 hours (Figure 4.3). Recovery of viable intracellular C. jejuni 11168 from IL-10-/- BMDCs was not significantly different than that of C. jejuni 260.94 when the infection time of 1 hour was followed by incubation with gentamicin and immediate lysis. In addition to enhanced invasion and/or intracellular survival, the possibility that C. jejuni 11168 may survive and replicate more efficiently than C. jejuni 260.94 in the extracellular medium or in vivo in the host, producing more bacteria over a longer time frame capable of invading host cells, cannot be excluded. Intracellular survival assays were performed with gentamicin treatment following 1 hour of infection and enumeration of viable intracellular C. jejuni at 24 hours p.i. In BMDCs derived from both 183 C57BL/6 WT and BALB/c WT mice, both C. jejuni strains were viable intracellularly at 24 hours p.i. with C. jejuni 11168 exhibiting significantly increased intracellular survival compared to C. jejuni 260.94 (Figure 4.4A). Significantly higher recovery of C. jejuni 11168 than C. jejuni 260.94 was also found in C57BL/6 IL- 10-/- BMDCs (Figure 4.4B). There was not a significant difference in viable C. jejuni 11168 versus 260.94 strains in the 1 hour invasion assay in IL-10-/- cells (Figure 4.3). Therefore, higher recovery of C. jejuni 11168 after the same invasion time but at 24 hours p.i. suggests that intracellular survival and perhaps replication contributed to increased recovery of C. jejuni 11168 over time in the invasion assays. Assays of increasing infection time in WT cells, as performed with IL-10-/- BMDCs, would be required to further characterize C. jejuni invasion and survival in WT cells. In contrast to intracellular survival within WT cells, virtually no C. jejuni 260.94 survived until 24 hours p.i. in BMDCs from IL-10-/- mice of either genotype in either of two replicates (Table 4.3B). However, intracellular survival of C. jejuni 11168 was significantly higher than that of C. jejuni 260.94 only in C57BL/6 IL-10-/- derived cells (Figure 4.4B). While more C. jejuni 11168 than 260.94 was recovered from BALB/c IL-10-/- BMDCs, the difference was not significant. Possible explanations for the discrepancy in survival between C. jejuni strains in IL-10-/- BMDCs from C57BL/6 vs. BALB/c mice include enhanced killing of C. jejuni 11168 or relatively fewer internalized C. jejuni organisms during the invasion period in BALB/c IL-10-/- BMDCs. The latter explanation is supported by the 1 hour invasion efficiency assay, in which significantly more C. jejuni was recovered in BMDCs from C57BL/6 IL-10-/- than BALB/c IL- 10-/- mice (Figure 4.3). These observations are consistent with a higher proportion of mature DCs and fewer macrophages in BALB/c IL-10-/- populations, leading to decreased phagocytic efficiency. Representative experiments of one intracellular survival assay using WT cells, and the other using IL-10-/- cells, are graphed together in Figure 4.4C.The patterns of viable intracellular C. jejuni recovered suggest that IL-10 deficiency decreases C. jejuni survival within BMDCs, although statistical comparisons between WT and IL-10-/- cells could not made as experiments were performed 184 independently. A pattern of increased killing of C. jejuni in the absence of IL-10 is consistent with reported enhancement of DC function, including expression of costimulatory molecules and production of TNF-α and IL-12, in human LPS-treated DCs following IL-10 neutralization.11 Experiments designed specifically to test the impact of IL-10 on killing ability of BMDCs are warranted, especially considering the severity of disease in IL-10-/- compared to WT mice infected with C. jejuni 11168.40; 41 The possibility that enhanced killing of C. jejuni in vivo by DCs in IL-10-/- mice contributes to exacerbated immune responses warrants further exploration. While significant differences were consistently found between C. jejuni strain recoveries, a difference between C57BL/6 and BALB/c mice was not consistent in all replicates at all time points (Tables 4.2A-4.2D and Tables 4.3A-4.3B). Significantly more C. jejuni was recovered from C57BL/6 IL-10- /- than BALB/c IL-10-/- derived BMDCs following 1 hour of invasion. Other repeatable genotype differences were seen only following 3 hours’ invasion time and in intracellular survival assays involving IL-10-/- BMDCs; in both cases, significantly more viable C. jejuni 11168 was enumerated in C57BL/6 IL-10- /- than BALB/c IL-10-/- cells. Possible explanations for these results include the enhanced intracellular survival shown by C. jejuni 11168, combined with the higher proportion of macrophages and immature DCs in C57BL/6 IL-10-/-compared to BALB/c IL-10-/- cultures. While both macrophages and DCs are antigen presenting cells, macrophages are more efficient phagocytes while DCs specialize in processing the antigen in preparation for presentation to T cells. Indeed, bone marrow-derived MHC IIlowF4/80high macrophages generated along with MHC IIhighF4/80low DCs using GM-CSF demonstrated superior phagocytosis of latex beads compared to DCs.45 C57BL/6 IL-10-/- BMDCs with a higher proportion of macrophages may have enhanced phagocytosis, and BALB/c IL-10-/- BMDCs with more mature DCs may lead to less efficient phagocytosis but increased killing. Therefore, increased recovery of the C. jejuni strain with enhanced intracellular survival, and also perhaps higher invasion efficiency, from cultured cells with a higher proportion of cells characterized by higher phagocytosis but less efficient killing, 185 represents key differences in characteristics of both C. jejuni strains and of populations of cultured cells in this model system. These findings suggest that the original hypothesis that both C. jejuni strains would be efficiently internalized and killed by BMDCs from both mouse genotypes should be rejected, as complexities in the host-microbe interaction preclude this predicted clear-cut result. In vitro assays reported in the literature evaluating invasion and survival of C. jejuni within cultured cells have varied substantially, including differences in cultured cell types, C. jejuni strains, use and concentration of gentamicin, MOI, and infection times.4; 27 Therefore, unsurprisingly, a variety of results have been reported concerning ability of C. jejuni to survive within phagocytic cells derived from mice and humans. Several studies have reported efficient killing of C. jejuni by macrophages and DCs. C. jejuni 81-176 was killed by mouse bone marrow-derived macrophages within 24 hours,60 and by human monocyte-derived DCs over 24-48 hours.23 Similarly, no C. jejuni 11168 was recovered from BMDCs derived from WT C57BL/6 mice after 8 hours.49 The last finding is inconsistent with our results (Table 4.3A, Figure 4.4A). This difference could be attributed to methodological variations or the time in which BMDCs were lysed and viable intracellular C. jejuni enumerated. That C. jejuni was present intracellularly in undetectable amounts by culture at 8 hours, but replicated sufficiently to allow detection at 24 hours, is a viable explanation; C. jejuni surviving within human macrophages began to multiply after 8 hours in one study.58 Similar to results of the current study, C. jejuni also has been reported to avoid intracellular killing, even up to several days. The majority of macrophages derived from human peripheral blood monocytes killed multiple strains of phagocytized C. jejuni, although ineffective killing and intracellular multiplication in macrophages derived from approximately 10% of donors was observed.58 Despite efficient phagocytosis, C. jejuni enteritis strains persisted within both BALB/c murine macrophages and human blood monocytes for 6 days,31 a finding confirmed with C. jejuni 81-176 in the human monocyte 28SC line.22 Persistent intracellular survival has implications for transit to other organs in vivo. 186 While methodological variation should not be discounted, discrepant reports in the literature regarding the intracellular fate of C. jejuni used in in vitro infection studies undoubtedly also result from infecting C. jejuni strain characteristics and type of cultured cell. To our knowledge, BMDCs from IL-10-/- mice on different genetic backgrounds have not been used to assess invasion and survival differences between C. jejuni strains associated with different disease outcomes. In the current study, the GKA was employed in two different ways to identify broad patterns of C. jejuni invasion efficiency and intracellular survival in BMDCs. A secondary goal was to determine if invasion and survival correlate with in vivo murine C. jejuni infection models, as studies reporting associations between invasiveness and clinical disease have shown conflicting results. No correlation was found between C. jejuni invasion and survival in murine macrophages and development of intestinal lesions in piglets.34 While invasive but non-inflammatory diarrhea-associated C. jejuni and C. coli isolates did not differ from colitis strains in internalization into HeLa cells, a higher number of colitis strains exhibited transcytosis through polarized Caco-2 monolayers than non-inflammatory strains.14 Association of C. jejuni and C. coli strains with HeLa cells correlated with symptoms of febrile diarrhea in patients, although association was not related to blood in the feces.15 Interestingly, C. jejuni mutant strains with deficiencies in formic acid metabolism shown to have reduced invasion capacity in vitro were able to colonize mice in similar abundance to the parent WT strain B2, but did not induce immunopathology in the colon as did the WT strain.7 In the current study, GKA results indicate superior invasion efficiency and intracellular survival of colitogenic C. jejuni 11168 compared to GBS-associated C. jejuni 260.94. These results are consistent with identification of several virulence-associated genes in C. jejuni 11168 with putative functions involving adherence, colonization, invasion, acid resistance, LOS, and motility that are either divergent or altogether absent in C. jejuni 260.94.6 C. jejuni 11168 also showed increased invasion into cultured young adult mouse colon (YAMC) epithelial cells compared to C. jejuni 260.94, although the difference was relatively smaller compared to differences between these strains and C. jejuni HB-9313.39 The 187 current GKA results also support findings in a BALB/c IL-10-/- mouse infection model, in which immunohistochemical labeling of C. jejuni in the ileocecocolic junction of sham-inoculated mice and those infected with C. jejuni 11168 or 260.94 strains was used to discern invasiveness in vivo. In this study, 9/10 C. jejuni 11168-infected BALB/c IL-10-/- mice exhibited positive intracellular labeling within macrophage/dendritic cell types within the submucosa, in contrast to 3/10 mice in each of the sham- inoculated and C. jejuni 260.94 groups. Similarly, 4/10 C. jejuni 11168-infected BALB/c IL-10-/-mice, but no mice in other treatment groups, exhibited intracellular labeling within macrophage/dendritic cell types within the lamina propria. In addition to heightened intracellular labeling of C. jejuni 11168 in the lamina propria and submucosa, BALB/c IL-10-/- mice infected with C. jejuni 11168 exhibited the most severe colitis and lowest survivorship (Brudvig et al., unpublished; Chapter 3). These studies suggest that patterns observed in vitro mimic in vivo interactions, and suggest that BMDCs are a useful in vitro model for further study of the initial host-microbe interaction in C. jejuni infection. A final objective of this study was to address the hypothesis that cytokines produced by C. jejuni-infected BMDCs derived from IL-10-/- mice would mirror immunogenicity previously observed in our IL-10-/- mouse models: C. jejuni 11168-infected IL-10-/- BMDCs from both mouse genotypes and BALB/c IL-10-/- BMDCs infected with C. jejuni 260.94 would produce pro-inflammatory and Th1/Th17- polarizing cytokines, while C57BL/6 IL-10-/- BMDCs infected with C. jejuni 260.94 would produce Th2- polarizing cytokines. Production of pro-inflammatory and Th-polarizing cytokines was assessed by a multiplexed flow cytometry-based bead assay (Figure 4.5). The most striking findings included enhanced production of MCP-1 by C57BL/6 IL-10-/- cells, especially those infected with C. jejuni 260.94. Because the chemokine MCP-1 is thought to contribute to Th2 polarization,19 this finding supports the Th2-mediated immune response of C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 reported previously.39 It was previously shown that splenic-derived DC from WT BALB/c mice produced more MCP-1 than those from WT C57BL/6 mice when exposed to TLR-2, 188 -4, and -9 ligands.35 This finding contrasts those in the current study, in which neither C. jejuni strain elicited significant MCP-1 production from BALB/c IL-10-/- mice. This discrepancy may be due to different DC subsets and stimulating ligands between these studies, or the absence of IL-10 in the current study. Additionally, because monocytes and macrophages are a main cellular source of MCP-1, the increased production by C57BL/6 IL-10-/- cells may reflect the higher number of macrophages in the culture. BALB/c-derived IL-10-/- BMDCs infected with C. jejuni 11168 exhibited significantly higher production of IL-6 compared to the negative control, and also to C. jejuni 11168-infected C57BL/6 IL-10-/- derived cells. IL-6 has pro-inflammatory effects and is critical in polarization of Th17 cells.8 This finding thus suggests that early events including the initial interaction of C. jejuni with DCs in the gut contribute to Th17 responses seen in vivo, including production of plasma anti-C. jejuni IgG2b and increased colonic IL-17 and IL-22 in BALB/c IL-10-/- mice infected with C. jejuni 11168 (Brudvig et al, unpublished; Chapter 3). The enhanced production of TNF-α in C. jejuni 260.94, but not C. jejuni 11168, infected IL-10-/- BMDCs from either mouse genotype in this experiment was surprising. Enteritis isolates C. jejuni 81-176 and C. jejuni 11168 both induced robust TNF-α production in human monocyte-derived DC and murine BMDCs, respectively.23; 49 TNF-α is a pro-inflammatory cytokine and the lack of significant production with C. jejuni 11168 infection was unexpected, considering the heightened inflammatory responses induced in in vivo IL-10-/- mouse models by C. jejuni 11168 compared to C. jejuni 260.94.6; 39 Possible explanations for the lack of robust production in C. jejuni 11168-infected IL-10-/- BMDCs include timing of collection of supernatants following infection, as perhaps without IL-10, maximal increases in TNF-α production stimulated by C. jejuni 11168 occur earlier or later than 24 hours p.i. as assessed in the current study. Because C57BL/6 mice have been reported to exhibit a Th1 immunological bias, production of IL-12p70 was of particular interest. In contrast to previous studies indicating that C. jejuni elicits robust 189 IL-12p70 production from infected DCs,23; 49 in the current study neither LPS nor either C. jejuni strain elicited significant IL-12 production in IL-10-/- BMDCs from either mouse genotype at 24 hours p.i. IL- 12p70 concentration was significantly higher only in uninfected (negative control) C57BL/6 IL-10-/- than BALB/c IL-10-/- BMDCs, suggesting a higher baseline production of IL-12 in C57BL/6 IL-10-/-mice and also complicating interpretation of responses in other treatment groups. IL-12p70, a heterodimer comprising p35 and p40 subunits, was chosen for measurement as the bioactive form of IL-12. Interestingly, LPS and IFN-γ stimulated blood monocytes isolated from human sepsis patients showed higher production of regulatory IL-12p40 and lower IL-12p70 production; the authors concluded that the balance between the two forms of IL-12 may represent a check on excessive inflammation.13 A possible explanation for the unexpected lack of production of bioactive IL-12p70 by infected IL-10-/- BMDCs is that, in the absence of IL-10, kinetics of production are altered or other immunoregulatory checks such as enhanced regulatory IL-12p40 production are activated to prevent uncontrolled inflammation. Production of IL-4 was also of particular interest as BALB/c mice are reported to have a Th2 immunological bias. IL-4 is a signature cytokine produced by Th2 cells, but IL-4 itself is also thought to be a key differentiation factor for Th2 cells.57 However, IL-4 is inconsistently produced by DCs and there is now evidence that other factors, such as affinity of T cell receptor in antigen presentation and cytokines produced by cells other than DCs, contribute to Th2 polarization.57 Nevertheless, IL-4 was included in the panel to determine if IL-10-/- BMDCs infected with C. jejuni, especially strain 260.94, would elicit IL-4 production. As with IL-12p70 production, infection with neither C. jejuni strain stimulated significant IL-4 production in IL-10-/- BMDCs from either C57BL/6 or BALB/c mice at 24 hours p.i. Surprisingly, baseline production in uninfected IL-10-/- BMDCs was significantly higher in C57BL/6 than BALB/c cells. As with IL- 12p70, the higher IL-4 production in uninfected C57BL/6 IL-10-/- BMDCs complicates interpretation of the response in other treatment groups. The lack of IL-4 production in uninfected BALB/c IL-10-/- cells suggests that in this model, absence of IL-10 dampens the reported Th2 bias in BALB/c mice. In C57BL/6 190 IL-10-/- BMDCs, infection with C. jejuni 11168 resulted in significant suppression of IL-4 production, supporting the earlier finding that this strain preferentially drives Th1/Th17 instead of Th2 responses in C57BL/6 IL-10-/- mice.39 The original hypotheses concerning cytokine production were formulated based upon systemic and local colonic immune responses seen in previous mouse C. jejuni infection models. As C57BL/6 IL-10- /- mice infected with C. jejuni 260.94 exhibited primarily Th2 responses,39 the enhanced production of MCP-1 from BMDCs of C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 suggests that production of this cytokine by DCs may contribute to early polarization of the immune response. Similarly, Th17 responses were observed in BALB/c IL-10-/- mice infected with C. jejuni 11168 (Brudvig et al, unpublished, Chapter 3), suggesting that early production of IL-6 by infected DCs likewise contributes to polarization of adaptive immunity. Although novel in many aspects, this study has several limitations and presents numerous opportunities for future research. While no striking differences between differentiation of BMDCs in WT and IL-10-/- mice were seen, further replicates with flow cytometric analysis especially on days 9, 10, and 11 of culture when the cells are used experimentally would allow more complete evaluation. Furthermore, while no difference was obvious in unstimulated cells, treatment of WT and IL-10-/- cells from both mouse genotypes with LPS would further elucidate the effects of IL-10 on maturation, including expression of costimulatory CD40, CD80, and CD86 molecules and MHC II, and function including cytokine production. Increasing the number of replicate experiments assessing expression of CD11c, MHC II, and F4/80 in C57BL/6 and BALB/c cells, and including a wider time frame of differentiation such as days 6-12 of culture, would help confirm or refute the hypothesis that C57BL/6 cultures represent a more diverse population of differentiating cells. Finally, the realization that differentiated BMDCs comprise different populations of DCs and macrophages between mouse genotypes complicates comparison of invasion and survival data and cytokine production between 191 mouse genotypes in this model, yet likely also represents how host genetics combined with C. jejuni strain variations can induce different host responses. The GKAs used in this study were designed to detect major differences in invasion efficiency and intracellular survival, accounting for both C. jejuni strain characteristics and host genetic background. In order to statistically evaluate both C. jejuni strain and mouse genotype as factors, a comparable MOI had to be attained for both C. jejuni strains in the same assay. While spectrophotometry allows estimation of CFU/mL in culture, achieving the desired OD560 did not consistently yield the predicted CFU when serially diluted and plated. This may have been due to several factors, including clumping together of the C. jejuni cells by their flagella, despite vortexing. Additionally, C. jejuni cultures were prepared in RPMI-based medium. The blank measurement used in determining OD of the C. jejuni culture was observed to drift over repeated readings, likely due to color change of medium upon exposure to air, and this may have led to a skewed OD of the culture. Difficulty in achieving precisely 2  108 CFU/mL for both C. jejuni cultures in all assays for an MOI of 100 led to inclusion of a broader range of approximate MOI (60-175) for the results presented in the body of the dissertation (Tables 4.2A-D and Tables 4.3A- B). The number of replicates at each time point is also a limitation; additional replicates would have strengthened statistical conclusions and reinforced observed patterns. [Additional experiments assessing intracellular survival using BMDCs derived from both WT and IL-10-/- mice were performed following completion of experiments included in the body of the dissertation. Results from these additional replicates are not discussed, but are presented as supplemental data in Appendix B, Table A.1.] However, numbers of C. jejuni recovered between replicates were fairly consistent despite lack of an MOI of precisely 100, and use of the GKA served the purpose of identifying broad patterns in invasion and survival. Statistical analyses also were limited by the experimental designs, which were complicated by logistical limitations. Each individual time point was assessed by independent experiments; ideally, C. 192 jejuni recovery over time would have been more precisely evaluated in time course assays, eliminating the variability of different MOIs and batches of BMDCs between time points. However, an assay that large, including four C. jejuni strain/mouse genotype combinations over multiple time points with sufficient replicate wells, would have been logistically challenging with this method. Thus, statistical analysis of results at each time point and observation of patterns over time was chosen as the best possible method of evaluating invasion and survival. Experiments performed assessing one mouse genotype/C. jejuni strain combination at different time points would allow statistical comparisons within that combination to more clearly evaluate changes in intracellular C. jejuni survival. Similarly, an interesting pattern of enhanced killing of C. jejuni by IL-10-/- compared to WT BMDCs was observed, but could not be compared statistically because IL-10-/- and WT cells were not evaluated in the same assays. Therefore, further assessment of the impact of IL-10 on intracellular C. jejuni survival can be compared in future experiments including BMDCs from a WT and IL-10-/- mouse of the same genetic background; again, the logistics of including WT and IL-10-/- mice from different genetic backgrounds infected with two C. jejuni strains would be logistically prohibitive by this method. Finally, measurement of cytokine production by infected IL-10-/- BMDCs provided insight on putative Th-polarization induced by colitogenic and GBS-associated C. jejuni strains in mice of different genetic backgrounds. Further experiments should include co-culture of infected BMDCs with naïve CD4+ T cells, followed by measurement of cytokines such as IFN-γ, IL-4, and IL-17 to further confirm induction of Th1, Th2, or Th17 polarization. Additionally, inclusion of BMDCs derived from WT mice to investigate the impact of IL-10 on cytokine production and Th polarization would be informative. Inclusion of WT cells would help confirm or refute the hypothesis that DC-derived IL-10 directs Th2 polarization, and determine if IL-10 alters kinetics of TNF-α or IL-12p70 production. Measurement of IL-12p40 along with IL-12p70 production in IL-10-/- cells would be warranted to evaluate if a similar anti-inflammatory check by IL-12p40 as seen in human monocytic cells is also present in DCs. 193 The main objectives of this study included characterizing the initial host-microbe interaction, and determining the interplay of C. jejuni strain characteristics and host genetic background in initiation of the immune response. Results of this study suggest that DCs differentiated from bone marrow precursors are a useful model for studying invasion and intracellular survival of C. jejuni. Intracellular survival of colitogenic C. jejuni 11168 correlates with enhanced invasion and pathogenicity observed in vivo, and DC cytokine responses elicited by the different C. jejuni strains are in general agreement with adaptive immune responses observed in previous mouse models. Future studies, including more precise characterization of differences in BMDCs derived from C57BL/6 and BALB/c mice and the effect of IL-10 during differentiation, are warranted. Additional studies to expand upon patterns observed in the GKAs and further discern the relative contributions of invasion and intracellular survival, especially in C. jejuni 11168, by methods such as fluorescent or electron microscopy, would be informative. 194 APPENDICES 195 APPENDIX A: TABLES AND FIGURES Table 4.1. Mean ± Standard Deviation of parameters determined by flow cytometric analysis of three independent experiments, BMDCs. Stem cells derived from the bone marrow of C57BL/6 interleukin (IL)-10-/- and BALB/c IL-10-/- mice were differentiated for 10 days with granulocyte-macrophage colony stimulating factor. t-tests were performed on each parameter to determine if discrepancies in differentiation between C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were statistically significant; P-values are listed in bottom row for each parameter. Percentage data were subjected to Arcsine transformation prior to analysis. MFI, Median Fluorescence Intensity; BMDC, Bone Marrow-derived Dendritic Cell. C57BL/6 IL-10-/- BALB/c IL-10-/- P-value % CD11c Positive 90.4 ± 3.96 80.5 ± 7.27 0.111 % CD11c Positive, MHC IIhigh 35.9 ± 11.8 55.5 ± 6.1 0.063 MFI MHC II (All MHC II Positive) 3,202 ± 293 11,933 ± 344 <0.001 % CD11c Positive, MHC IIlow 56.8 ± 15.6 25.5 ± 5.7 0.03 % CD11c, F4/80 Double Positive 25.7 ± 5.8 16.4 ± 2.5 0.054 MFI MHC II (CD11c, F4/80 Double Positive) 557 ± 96 1,169 ± 369 0.05 196 Table 4.2A. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 1 hour invasion time followed by 1 hour exposure to gentamicin and immediate lysis. MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units; ns = non-significant. Exposure Time to C. jejuni, Prior to Gentamicin Treatment Replicate (*Indicates Presented in Graph) Starting Density (No. Cells in 10 mL) 1 2 C57 2.6 x 106 BALB 2.5 x 106 C57 2.5 x 106 BALB 2.4 x 106 1 Hour MOI (Approximate Range Derived from Limiting Dilution Results) 11168: 99-105 Mean ± SD, CFU/mL Statistical Analyses and Results C57 11168: 4.7 ± 0.35 x 103 BALB 11168: 4.3 ± 0 x 103 Assumption of equal variance not met. Mann-Whitney Rank Sum Test: ns (P = 0.333) 260.94: 86-165 C57 260.94: 2.88 ± 0.536 x 103 BALB 260.94: 1.5 ± 0.96 x 103 t-test; ns (P = 0.114) 3* Both 2.5 x 106 11168: 65-105 260.94: 75-190 C57 11168: 3.7 ± 1.5 x 103 BALB 11168: 1 ± 0.9 x 103 C57 260.94: 2.97 ± 0.585 x 103 BALB 260.94: 1.13 ± 0.071 x 103 Two-way ANOVA: Genotype significant (P = 0.003), but not C. jejuni strain (P = 0.466). Holm-Sidak pairwise comparisons: Genotype significant, [C57 > BALB]. One-way ANOVA was also performed to further examine pairwise comparisons. Significance overall (P = 0.017); significant Holm-Sidak pairwise comparisons: [C57 11168 > BALB 260.94]; [C57 11168 > BALB 11168]. Results of the two-way ANOVA are shown graphically. Conclusions for one hour invasion, IL-10-/- BMDCs: No C. jejuni strain difference detected, but a significant genotype difference was found (supported by pattern in the two replicates not graphed): More C. jejuni recovered from C57 than BALB mice. 197 Table 4.2B. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 2 hours invasion time followed by 1 hour exposure to gentamicin and immediate lysis. MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units. Exposure Time to C. jejuni, Prior to Gentamicin Treatment Replicate (*Indicates Presented in Graph) Starting Density (No. Cells in 10 mL) MOI (Approximate Range Derived from Limiting Dilution Results) Mean ± SD, CFU/mL Statistical Analyses and Results 1 2* C57: 3.2 x 106 BALB: 2.8 x 106 11168: 90-96 C57 11168: 1.66 ± 0.157 x 104 BALB 11168: 2.87 ± 0.465 x 104 t-test: significantly more C. jejuni 11168 recovered from BALB than C57 mice (P = 0.013) C57: 2.8 x 106 BALB: 2.1 x 106 11168: 60-63 260.94: 118-160 C57 11168: 5.0 ± 0.23 x 103 BALB 11168: 5.0 ± 1.1 x 103 C57 260.94: 1 ± 0.7 x 103 BALB 260.94: 1 ± 0.2 x 103 Two-way ANOVA: significance found in C. jejuni strain (P < 0.001), but not in Genotype (P = 0.867). Holm-Sidak pairwise comparisons: significantly more C. jejuni 11168 than 260.94 recovered. C57 11168: 8.0 ± 1.6 x 103 BALB 11168: 4.4 ± 1.9 x 103 C57 260.94: 1 ± 0.6 x 102 2 Hours 3 C57: 2.3 x 106 BALB: 2.5 x 106 11168: 70-92 260.94: 144-190 4 C57: 2.5 x 106 BALB: 2.1 x 106 11168: 104-170 260.94: 115-145 Two-way ANOVA: Genotype (P = 0.032), C. jejuni strain (P <0.001), and interaction (P = 0.037) all significant. The following Holm-Sidak pairwise comparisons were significant: [C57 11168 > C57 260.94]; [BALB 11168 > BALB 260.94]; [C57 11168 > BALB 11168]; recovery of C. jejuni 260.94 was not significantly different between mouse genotypes. Because the assumption of normality was not met, Kruskal-Wallis was also performed, with overall significance (P = 0.018). Dunn's and Tukey's post-hoc testing: [C57 11168 > BALB 260.94] was significant. Significance was found in the following comparisons using Student-Newman-Keuls method: [C57 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]; [BALB 11168 > BALB 260.94]; and [BALB 11168 > C57 260.94]. Recovery of C. jejuni 260.94 was not different between mouse genotypes. Two-way ANOVA: C. jejuni strain significant (P = 0.001), but not Genotype (P = 0.848). Holm-Sidak pairwise comparisons determined significantly more C. jejuni 11168 than 260.94 was recovered. Because the assumption of normality was not met, Kruskal-Wallis was also performed, with overall significance (P = 0.024). Tukey, Dunn's, and Student-Newman-Keuls post-hoc testing all found significance in only [C57 11168 > C57 260.94]. BALB 260.94: 70 ± 60 C57 11168: 7.1 ± 2.4 x 103 BALB 11168: 6.4 ± 3.9 x 103 C57 260.94: 70 ± 20 BALB 260.94: 2.4 ± 1.0 x 102 Conclusions for two hour invasion, IL-10-/- BMDCs: Consistent significant difference in C. jejuni strain: more 11168 than 260.94 was recovered. A mouse genotype difference was not consistent between replicates. 198 Table 4.2C. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 3 hours invasion time, followed by 1 hour exposure to gentamicin and immediate lysis. MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units; ns = non-significant. Exposure Time to C. jejuni, Prior to Gentamicin Treatment Replicate (*Indicates Presented in Graph) Starting Density (No. Cells in 10 mL) MOI (Approximate Range Derived from Limiting Dilution Results) Mean ± SD, CFU/mL Statistical Analyses and Results 1* C57: 2.5 x 106 BALB: 1.7 x 106 3 Hours C57 260.94: 5.3 ± 1.9 x 103 BALB 260.94: 4.8 ± 0.72 x 103 11168: 126-150 260.94: 95-192 C57 11168: 4.91 ± 0.436 x 104 BALB 11168: 3.36 ± 0.421 x 104 Two-way ANOVA: Genotype (P = 0.002), C. jejuni strain (P < 0.001) and interaction (P = 0.004) all significant. Holm-Sidak significant pairwise comparisons: [C57 11168 > C57 260.94]; [BALB 11168 > BALB 260.94]; [C57 11168 > BALB 11168]; recovery of C. jejuni 260.94 was not significant between mouse genotypes. Because of the significant interaction, one-way ANOVA also was performed. Significance found overall (P <0.001). Post-hoc testing included Bonferroni t-test, Holm-Sidak, Tukey, and Student-Newman-Keuls methods, with the following significant comparisons found with all methods: [C57 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]; [BALB 11168 > BALB 260.94]; [BALB 11168 > C57 260.94]. Recovery of C. jejuni 260.94 was not significantly different between mouse genotypes. Results of the two-way ANOVA shown graphically. 2 3 Both 2.5 x 106 260.94: 43-75 C57 260.94: 5.8 ± 1.6 x 102 BALB 260.94: 3.7 ± 0.44 x 102 t-test: ns (P = 0.092) Both 2.5 x 106 11168: 106-127 C57 11168: 6.7 ± 1.1 x 104 BALB 11168: 2.73 ± 0.318 x 104 t-test: significantly more C. jejuni 11168 recovered from C57 than BALB mice (P = 0.003) Conclusions for three hour invasion, IL-10-/- BMDCs: Significantly more C. jejuni 11168 than 260.94 was recovered from mice of both genotypes, and significantly more C. jejuni 11168 was recovered from C57 than BALB mice. C. jejuni 260.94 recovery was not different between mouse genotypes. 199 Table 4.2D. Results of gentamicin killing assays evaluating combined invasion efficiency and intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; 23 hours invasion time, followed by 1 hour exposure to gentamicin and immediate lysis. MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units. Exposure Time to C. jejuni, Prior to Gentamicin Treatment Replicate (*Indicates Presented in Graph) Starting Density (No. Cells in 10 mL) MOI (Approximate Range Derived from Limiting Dilution Results) Mean ± SD, CFU/mL Statistical Analyses and Results 1 C57: 2.5 x 106 BALB: 1.5 x 106 23 Hours 11168: 35-60 260.94: 109-150 C57 11168: 3.00 ± 0 x 105 BALB 11168: 3.00 ± 0 x 105 C57 260.94: 3.2 ± 1.0 x 103 BALB 260.94: 1.4 ± 0.51 x 104 Two-way ANOVA: Genotype (P = 0.006), C. jejuni strain (P <0.001), and interaction (P = 0.006) all significant. Significant Holm-Sidak pairwise comparisons: [C57 11168 > C57 260.94]; [BALB 11168 > BALB 260.94]; [BALB 260.94 > C57 260.94]. Because assumption of normality was not met and a significant interaction was present, Kruskal-Wallis was also performed. Overall significance was found (P = 0.014), but Tukey, Dunn's, and Student-Newman-Keuls post-hoc methods did not identify any significant pairwise comparisons. [Note: C. jejuni 11168 colony numbers were too numerous to count (>300) with a dilution of 10-3; for statistical purposes, these results were treated as 3.00 x 105 CFU/mL recovered for both C57 and BALB mice.] 2* Both 2.5 x 106 11168: 27-39 260.94: 142-185 C57 11168: 8.4 ± 3.5 x 105 BALB 11168: 1.1 ± 0.42 x 106 C57 260.94: 9 ± 10 x 102 BALB 260.94: 3.0 ± 2.8 x 103 Two-way ANOVA: C. jejuni strain was significant (P < 0.001), but not Genotype (P = 0.361). Holm-Sidak pairwise comparisons: significantly more C. jejuni 11168 than 260.94 was recovered. Because normality and equal variance assumptions were not met, Kruskal-Wallis was also performed with overall significance found (P = 0.030), followed by three post-hoc tests. Significant comparisons listed. Tukey: [BALB 11168 > C57 260.94]; Dunn's: none; Student-Newman-Keuls: [BALB 11168 > C57 260.94]; [BALB 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; [C57 11168 > BALB 260.94]. Results of the two-way ANOVA are shown graphically. Conclusions for 23 hour invasion, IL-10-/- BMDCs: Significantly more C. jejuni 11168 was recovered than 260.94. There was also a trend for more C. jejuni to be recovered from BALB than C57 mice, but this was not confirmed statistically in both replicates. The second replicate was chosen as representative since more precise enumeration of C. jejuni 11168 recovery was possible. 200 Table 4.3A. Results of gentamicin killing assays evaluating intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of wild-type (WT) mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; one hour of invasion followed by gentamicin treatment, with lysis at 24 hours post-infection. MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units. Mean ± SD, CFU/mL Statistical Analyses and Results Replicate (*Indicates Presented in Graph) Starting Density (No. Cells in 10 mL) MOI (Approximate Range Derived from Limiting Dilution Results) 1* Both 2.5 x 106 11168: 117-175 260.94: 146-175 2 Both 2.5 x 106 11168: 154-215 260.94: 103-152 C57 11168: 2.48 ± 0.603 x 104 BALB 11168: 2.13 ± 0.590 x 104 C57 260.94: 1.2 ± 0.58 x 103 BALB 260.94: 2.22 ± 0.673 x 103 C57 11168: 6.6 ± 1.2 x 103 Two-way ANOVA: Genotype was not significant (P = 0.637), but C. jejuni strain was significant (P <0.001). Holm-Sidak post-hoc testing: significantly more C. jejuni 11168 than 260.94 was recovered. One-way ANOVA was also run to further examine pairwise comparisons. There was significance overall (P <0.001), and Holm- Sidak testing indicated signficant differences in the following: [C57 11168 > C57 260.94]; [C57 11168 > BALB 260.94]; [BALB 11168 > C57 260.94]; [BALB 11168 > BALB 260.94]. Results of the two-way ANOVA are shown graphically. BALB 11168: 2.6 ± 0.92 x 103 C57 260.94: 4.2 ± 2.8 x 102 BALB 260.94: 80 ± 70 Two-way ANOVA: Significance in Genotype (P = 0.001), C. jejuni strain (P < 0.001), and interaction (P = 0.004) were found. Holm-Sidak pairwise testing showed signficance in the following: [C57 11168 > C57 260.94]; [BALB 11168 > BALB 260.94]; [C57 11168 > BALB 11168]. Recovery of C. jejuni 260.94 was not significantly different between mouse genotypes. Because a signficant interaction was identified, one-way ANOVA was also performed. Significance overall was found (P < 0.001). Holm-Sidak post-hoc testing revealed significance in the following comparisons: [C57 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]; [BALB 11168 > BALB 260.94]; [BALB 11168 > C57 260.94]. 3 Both 2.5 x 106 11168: 147-160 C57 11168: 1.46 ± 0.532 x 104 BALB 11168: 8.5 ± 1.8 x 104 t-test: Significantly more C. jejuni 11168 recovered from BALB than C57 mice (P = 0.003) WT mice: Conclusions for intracellular survival: Significantly more C. jejuni 11168 than 260.94 was recovered, but differences between mouse genotypes were not consistent. The first replicate was chosen as representative because no genotype difference was identified, and numbers of C. jejuni 11168 recovered were similar to those in the third replicate. 201 Table 4.3B. Results of gentamicin killing assays evaluating intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of interleukin (IL)-10-/- mice on C57BL/6 (C57) and BALB/c (BALB) genetic backgrounds; one hour of invasion followed by gentamicin treatment, with lysis at 24 hours post-infection. Wet mount preparations of inocula were not possible in the first replicate. MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units; ns = non-significant. Replicate (*Indicates Presented in Graph) Starting Density (No. Cells in 10 mL) MOI (Approximate Range Derived from Limiting Dilution Results) 1 C57: 2.5 x 106 BALB: 2.3 x 106 11168: 40-88 260.94: 50-107 Mean ± SD, CFU/mL Statistical Analyses and Results C57 11168: 4 ± 7 x 102 BALB 11168: 3 ± 3 x 102 C57 260.94: 0 ± 0 BALB 260.94: 3 ± 6 Two-way ANOVA: Genotype (P = 0.685), C. jejuni strain (P = 0.157) both ns. Because assumption of normality was not met, Kruskal-Wallis was also performed. Significance was found overall (P = 0.027), but Tukey, Dunn's and Student-Newman- Keuls post-hoc testing did not determine significance in any pairwise comparison. [Note: Uneven distribution of C. jejuni 11168 recovery between triplicate wells.] 2* Both 2.5 x 106 11168: 162-190 260.94: 58-61 Two-way ANOVA: Genotype (P <0.001), C. jejuni strain (P < 0.001), and interaction (P <0.001) all significant. Significant comparisons by Holm-Sidak testing: [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]. Because a significant interaction was found, one-way ANOVA was also performed and was significant overall (P < 0.001). Holm-Sidak post-hoc testing indicated significance in the following comparisons: [C57 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; and [C57 11168 > BALB 11168]. Because assumption of normality was not met, Kruskal-Wallis was also performed. Significance was found overall (P = 0.017). Pairwise testing with Tukey, Dunn's methods indicated [C57 11168 > BALB 260.94] was significant; Student-Newman Keuls pairwise comparisons indicated significance in the following: [C57 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]; [BALB 11168 > BALB 260.94]; [BALB 11168 > C57 260.94]. Results of the 2-way ANOVA are shown graphically. C57 11168: 1.41 ± 0.267 x 103 BALB 11168: 2 ± 0.9 x 102 C57 260.94: 10 ± 20 BALB 260.94: 0 ± 0 IL-10-/- mice: Conclusions for intracellular survival: In mice of either genotype, more C. jejuni 11168 than 260.94 was recovered, but this difference was only significant within C57 mice. Significantly more C. jejuni 11168 was recovered from C57 than from BALB mice, but the difference in C. jejuni 260.94 recovered between mouse genotypes was not significant. The second replicate was chosen as representative due to less variation between triplicate wells in recovery of C. jejuni 11168, in addition to identification of the same patterns as in the first replicate. 202 Table 4.4. Cytokine analysis, first independent experiment. This table depicts the results of statistical analyses of cytokine production by C. jejuni-infected BMDCs derived from IL-10-/- mice on a C57BL/6 or BALB/c genetic background, Experiment 1. Multiplicities of infection for this assay were: C. jejuni 11168: 144; C. jejuni 260.94: 115. C57, C57BL/6 IL-10-/-; BALB/c, BALB/c IL-10-/-; LPS, lipopolysaccharide (positive control); Med, medium only (negative control); 11168, C. jejuni 11168; 260.94, C. jejuni 260.94; Ns, non- significant (P > 0.05); Treatment group refers to LPS, C. jejuni strain, or medium. Significance? Significant Comparison(s)? No No No Yes None None None Mouse Genotype Treatment Group Interaction P-value (for Factor or Interaction) Ns Ns Ns <0.001 <0.001 0.003 MCP-1, significant pairwise comparisons (Holm-Sidak) Within 260.94: C57 > BALB/c Within LPS: C57 > BALB/c None Treatment Group Interaction Within LPS: BALB/c > C57 None Ns 0.002 0.05 Ns IL-6, significant pairwise comparisons (Holm-Sidak) Within C57: Within BALB/c: Analyte Statistical Test 2-way ANOVA 2-way ANOVA 2-way ANOVA Normality/Equal Variance Passed Both Passed Both Failed Normality 2-way ANOVA Failed Normality IFN-γ IL-12p70 TGF-β MCP-1 Within C57: LPS > Med LPS > 11168 LPS > 260.94 260.94 > Med 260.94 > 11168 None IL-4 IL-6 2-way ANOVA Failed Normality 2-way ANOVA Failed Normality TNF-α 2-way ANOVA Failed Normality Within BALB/c: LPS > Med LPS > 11168 LPS > 260.94 No Yes LPS > Med LPS > 260.94 LPS > 11168 No 203 Table 4.5. Cytokine analysis, second independent experiment. This table depicts the results of statistical analyses of cytokine production by C. jejuni-infected BMDCs derived from IL-10-/- mice on a C57BL/6 or BALB/c genetic background, Experiment 2. Multiplicities of infection for this assay were: C. jejuni 11168: 170; C. jejuni 260.94: 138. Graphs of these results are shown in Figure 4.5. C57, C57BL/6 IL- 10-/-; BALB/c, BALB/c IL-10-/-; LPS, lipopolysaccharide (positive control); Med, medium only (negative control); 11168, C. jejuni 11168; 260.94, C. jejuni 260.94; Ns, non-significant (P > 0.05); Treatment group denotes LPS, C. jejuni strain, or medium. Significance? Comparison(s)? Interaction) Significant P-value (for Factor or No Yes None Interaction Ns 0.039 Ns <0.001 <0.001 <0.001 0.029 Analyte IFN-γ Statistical Normality/Equal Test Variance 2-way ANOVA Passed Both IL-12p70 2-way ANOVA Failed Normality None TGF-β 2-way ANOVA Failed Normality MCP-1 2-way ANOVA Failed Normality Within C57: LPS > Med LPS > 11168 LPS > 260.94 260.94 > Med 260.94 > 11168 IL-4 2-way ANOVA Failed Normality IL-12p70: significant pairwise comparisons (Holm-Sidak) Within C57: Within BALB/c: Within Medium: None No Yes C57 > BALB/c None Mouse Genotype Treatment Group Interaction MCP-1: significant pairwise comparisons (Holm-Sidak) Within 260.94: C57 > BALB/c Within LPS: C57 > BALB/c Within BALB/c: LPS > Med LPS > 260.94 LPS > 11168 Yes Interaction IL-4: significant pairwise comparisons (Holm-Sidak) Within C57: Med > 11168 Within BALB/c: Within Medium: None C57 > BALB/c 204 Table 4.5 (cont’d) Analyte Statistical Test Normality/Equal Variance 2-way ANOVA Failed Normality Within C57: LPS > Med LPS > 11168 LPS > 260.94 IL-6 TNF-α 2-way ANOVA Failed Normality Significance? Comparison(s)? Interaction) Significant P-value (for Factor or Yes Mouse Genotype Treatment Group Interaction <0.001 <0.001 0.001 IL-6: significant pairwise comparisons (Holm-Sidak) Within BALB/c: LPS > Med LPS > 260.94 LPS > 11168 11168 > Med Yes Within 11168: BALB/c > C57 Within LPS: BALB/c > C57 Treatment Group <0.001 TNF-α: significant pairwise comparisons (Holm-Sidak) (2-way ANOVA) All within Treatment Group: LPS > Med LPS > 11168 LPS > 260.94 260.94 > Med 1-way ANOVA Failed Normality Yes Overall <0.001 BALB/c LPS > BALB/c Med TNF-α: significant pairwise comparisons (Holm-Sidak) (1-way ANOVA) BALB/c LPS > BALB/c 260.94 BALB/c LPS > BALB/c 11168 C57 LPS > C57 11168 C57 LPS > BALB/c 260.94 C57 LPS > BALB/c 11168 BALB/c LPS > C57 260.94 BALB/c LPS > C57 Med BALB/c LPS > C57 11168 C57 LPS > BALB/c Med C57 LPS > C57 Med 205 Table 4.6. Cytokine analysis, third independent experiment. This table depicts the results of statistical analyses of cytokine production by C. jejuni-infected BMDCs derived from IL-10-/- mice on a C57BL/6 or BALB/c genetic background, Experiment 3. Multiplicities of infection for this assay were: C. jejuni 11168: 107; C. jejuni 260.94: 148. C57, C57BL/6 IL-10-/-; BALB/c, BALB/c IL-10-/-; LPS, lipopolysaccharide (positive control); Med, medium only (negative control); 11168, C. jejuni 11168; 260.94, C. jejuni 260.94; Ns, non- significant (P > 0.05); Tx Grp, Treatment Group (LPS, C. jejuni strain, medium). Analyte Statistical Test Normality/Equal Variance IFN-γ 2-way ANOVA Passed Both Within C57: Med > 260.94 Med > 11168 Med > LPS IL-12p70 2-way ANOVA Failed Normality Significance? Comparison(s)? Interaction) Significant P-value (for Factor or Yes Treatment Group Interaction 0.012 <0.001 IFN-γ: significant pairwise comparisons (Holm-Sidak) Within BALB/c: None Within Med: C57 > BALB/c Within 11168: BALB/c > C57 Yes Interaction 0.009 TGF-β 2-way ANOVA Passed Both IL-12p70: significant pairwise comparisons (Holm-Sidak) Tx Grp w/in C57: Tx Grp w/in BALB/c: None None Yes Within Med: C57 > BALB/c Within LPS: BALB/c > C57 TGF-β: significant pairwise comparisons (Holm-Sidak) (2-way ANOVA) Genotype <0.001 C57 > BALB/c; 1-way ANOVA performed for further comparison 1-way ANOVA Passed Both Yes Overall 0.003 TGF-β: significant pairwise comparisons (Holm-Sidak) (1-way ANOVA) C57 260.94 > BALB/c 260.94 C57 260.94 > BALB/c 11168 206 Table 4.6 (cont’d) Analyte Statistical Test Variance Significance? Comparison(s)? Interaction) Normality/Equal Significant P-value (for Factor or MCP-1 2-way ANOVA Failed Normality Yes Mouse Genotype Treatment Group Interaction <0.001 <0.001 <0.001 MCP-1: Significant pairwise comparisons (Holm-Sidak) Genotype within Treatment Groups: C57 > BALB/c for: Med, 11168, 260.94, LPS Within C57: LPS > Med LPS > 11168 LPS > 260.94 260.94 > Med 260.94 > 11168 11168 > Med Within BALB/c: LPS > Med LPS > 11168 LPS > 260.94 IL-4 IL-6 2-way ANOVA Failed Normality 2-way ANOVA Failed Both No Yes None Mouse Genotype Treatment Group Interaction Ns <0.001 <0.001 <0.001 0.003 <0.001 <0.001 Kruskal-Wallis IL-6: significant pairwise comparisons (Holm-Sidak) (2-way ANOVA) Within C57: LPS > Med LPS > 260.94 LPS > 11168 Within BALB/c: LPS > Med LPS > 260.94 11168 > Med 11168 > 260.94 LPS > 11168 Yes Within 11168: BALB/c > C57 Within LPS: BALB/c > C57 Overall 0.002 IL-6: significant pairwise comparisons (Tukey) (Kruskal-Wallis) BALB/c LPS > C57 Med BALB/c LPS > BALB/c Med TNF-α 2-way ANOVA Failed Normality Yes Mouse Genotype Treatment Group Interaction TNF-α: significant comparisons (Holm-Sidak) Within C57: LPS > Med 11168 > Med LPS > 260.94 Within BALB/c: 11168 > Med 11168 > 260.94 LPS > Med LPS > 260.94 11168 > LPS 207 Within 11168: BALB/c > C57 Figure 4.1. Analysis of CD11c and MHC II expression by flow cytometry. This figure represents scatter plots of bone marrow stem cells isolated from (A) one C57BL/6 wild-type (WT) mouse, (B) one C57BL/6 IL-10-/- mouse, (C) one BALB/c WT mouse, and (D) one BALB/c IL-10-/- mouse each after 10 days of culture with granulocyte-macrophage colony stimulating factor. Figure (A) is representative of two analyses of C57BL/6 WT mice, one each from days 9 and 10 of culture; the Median Fluorescence Intensity of MHC II, in all MHC II-positive cells (MHC II MFI), of the C57BL/6 WT mouse was 5,640. Figure (C) is representative of three analyses of BALB/c WT mice, one from day 9 and the other two from day 10 of culture; the MHC II MFI of the BALB/c WT mouse was 14,175. (B) and (D) are also shown in Figure 4.2, and each figure is from one experiment representative of three independent experiments. The MHC II MFI of the C57BL/6 IL-10-/- mouse (B) was 2,987, while that of the BALB/c IL-10-/- mouse (D) was 11,554. 208 Figure 4.2. Analysis of CD11c, MHC II, and F4/80 expression by flow cytometry. This figure represents scatter plots of bone marrow stem cells isolated from (A) one C57BL/6 IL-10-/- mouse, and (B) one BALB/c IL-10-/- mouse following 10 days of culture with granulocyte-macrophage colony stimulating factor. The left figure in each panel shows the percentage of CD11c+ cells within the single viable cell population. The right figure in each panel shows the percentage of F4/80+ cells within the CD11c+ population. Results of one experiment representative of three independent experiments are shown. Details regarding these and additional parameters in the three independent experiments are shown in Table 4.1. The plots depicting CD11c and MHC II expression are also shown in Figure 4.1. 209 Figure 4.3. Differences in invasion efficiency and intracellular survival of two Campylobacter jejuni strains in bone marrow-derived dendritic cells (BMDCs) from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice assessed by gentamicin killing assay. Cells were infected with either colitogenic C. jejuni 11168, or Guillain-Barré syndrome patient-derived C. jejuni 260.94, for 1, 2, 3, or 23 hours prior to gentamicin treatment to kill extracellular bacteria. BMDCs were then lysed, and lysates were spread (diluted when needed) and cultured on Bolton agar plates for 72 hours to enumerate viable intracellular C. jejuni, presented as colony forming units (CFU)/mL. Data are from three-wells in an experiment, expressed as mean ± SEM. One experiment representative of the 2-4 independent experiments performed at each time point is shown. Colored, dotted lines connect the data between time points for each treatment group and are added to help identify pattern over time; no statistical analyses were performed between time points, since experiments at each time point were performed separately. Statistical significance, determined by two-way ANOVA followed by Holm-Sidak pairwise comparisons, denoted in this graph represents repeatable differences observed between replicates. Boxes around data at 1 and 3 hours of infection indicate significant differences between mouse genotypes: at 1 hour, more C. jejuni was recovered from C57BL/6 IL-10-/- than BALB/c IL-10-/- mice, and at 3 hours, significantly more C. jejuni 11168 was recovered from C57BL/6 IL-10-/- than BALB/c IL-10-/- mice. Asterisks above the 2, 3, and 23 hour time points indicate a statistically significant difference in C. jejuni strain recovery: beginning at 2 hours of infection, significantly more C. jejuni 11168 was recovered than C. jejuni 260.94. For details including CFU/mL results and statistical analyses including the experiments graphed and other replicates, see Tables 4.2A-2D. 210 Figure 4.4. Differences in intracellular survival of two C. jejuni strains in bone marrow-derived dendritic cells (BMDCs) from wild-type (WT) and IL-10-/- mice on both C57BL/6 and BALB/c backgrounds as assessed by gentamicin killing assay. Cells were infected with either C. jejuni 11168 or 260.94 strains for 1 hour, treated with gentamicin to kill extracellular bacteria, washed, and incubated until 24 hours post-infection. BMDCs were lysed, and lysates (diluted when needed) were spread and cultured on Bolton agar plates for 72 hours to enumerate viable intracellular C. jejuni, presented as colony forming units (CFU)/mL. Data are from three-wells in an experiment, expressed as mean ± SEM. (A) One representative experiment of three, WT mice; (B) One representative experiment of two, IL-10-/- mice. C. jejuni 260.94 was not recovered from BALB/c IL-10-/- cells and does not show up on the graph; (C) Experiments from (A) and (B) combined in same graph for comparison purposes. Killing ability of WT mice could not be compared statistically to IL-10-/- mice, as the experiments were performed separately. Asterisks indicate statistically significant differences; 2-way ANOVA, with Holm-Sidak pairwise comparisons. For details including CFU/mL results and statistical analyses including the experiments graphed and other replicates, see Tables 4.3A-B. 211 Figure 4.5. Cytokine production in C. jejuni-infected bone marrow-derived dendritic cells (BMDCs) derived from C57BL/6 IL-10-/- and BALB/c IL-10-/- mice. BMDCs were collected on day 10 of differentiation, seeded into 24-well culture plates, and infected with C. jejuni 11168 (multiplicity of infection [MOI]: 170), C. jejuni 260.94 (MOI: 138), or incubated with lipopolysaccharide (LPS; positive control) or medium only (Med; negative control) for 1 hour. Cells were then treated with gentamicin, washed, and incubated with fresh medium until supernatants were collected at approximately 24 hours p.i. Pro-inflammatory cytokines and those directing Th polarization into Th1, Th2, or Th17 cells were measured using a multiplexed flow cytometry-based bead assay. Data are from 3-wells, and mean ± SD is shown. One experiment representative of three independent experiments is shown. Brackets indicate statistically significant differences between treatment groups (2-way ANOVA, followed by Holm-Sidak pair-wise comparisons; Table 4.5). For details regarding statistical analyses of the other two replicates, see Table 4.4 and Table 4.6. 212 APPENDIX B: SUPPLEMENTAL DATA This appendix comprises Table A.1, showing results of additional gentamicin killing assay (GKA) replicates evaluating intracellular survival. These experiments were performed after completion of those included in the dissertation and are presented as supplemental data. Table A.1. Results of additional GKAs assessing intracellular survival of C. jejuni 11168 and 260.94 strains in cultured dendritic cells derived from the bone marrow of wild-type or interleukin-10-/- mice on C57BL/6 and BALB/c genetic backgrounds. Stem cells isolated from bone marrow were seeded at 2.5 × 106 cells/10 mL medium (except 1 mouse in 1 experiment, seeded at 2.3 × 106 cells/10 mL) and differentiated for 10 days. Dendritic cells were infected for 1 hr, treated with gentamicin, washed, and lysed at 24 hr to enumerate viable intracellular C. jejuni. WT = wild-type; IL-10-/-= interleukin-10-deficient; MOI = multiplicity of infection; SD = standard deviation; CFU = colony forming units; C57 = C57BL/6; BALB = BALB/c; ns = non-significant. Additional Replicate MOI (Approx. Range Derived from Limiting Dilution) WT #1 11168: 104-150 260.94: 102-335 Mean ± SD, CFU/mL Statistical Analyses and Results C57 11168: 2.1 ± 2.6 × 104 BALB 11168: 3.12 ± 1.67 × 104 C57 260.94: 3.3 ± 1.5 × 103 BALB 260.94: 4.7 ± 3.1 × 103 2-way ANOVA (failed equal variance): Only C. jejuni strain factor significant (P = 0.031). Holm-Sidak post-test: more C. jejuni 11168 than 260.94 recovered. Kruskal-Wallis (due to failing equal variance): ns (P = 0.168). (NOTE: one of 3 plates from the C57 11168 group was excluded due to too numerous to count colonies.) WT #2 11168: 30-105 260.94: 60-115 C57 11168: 6.21 ± 3.01 × 104 BALB 11168: 6.53 ± 1.16 × 104 C57 260.94: 1.7 ± 1.3 × 104 BALB 260.94: 1.82 ± 1.70 × 103 IL-10-/- #1 11168: 190-340 260.94: 219-270 C57 11168: 6.14 ± 1.60 × 104 BALB 11168: 2.04 ± 1.79 × 104 C57 260.94: 2.4 ± 2.0 × 103 BALB 260.94: 3.8 ± 4.5 × 102 IL-10-/- #2 260.94: 144-450 IL-10-/- #3 11168: 159-250 260.94: 203-215 C57 260.94: 4.4 ± 1.9 × 103 BALB 260.94: 6 ± 6 × 102 C57 11168: 1.5 ± 1.9 × 104 BALB 11168: 8 ± 11 × 102 C57 260.94: 1 ± 2 × 103 BALB 260.94: 3 ± 6 2-way ANOVA: Only C. jejuni strain factor significant (P < 0.001). Holm-Sidak post-test: more C. jejuni 11168 than 260.94 recovered. 2-way ANOVA (failed normality, passed equal variance): both factors significant (C. jejuni strain, P <0.001; Genotype, P = 0.015) and significant interaction (P = 0.023) present. Significant pairwise comparisons by Holm-Sidak post-test: [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]. 1-way ANOVA: P <0.001. Significant pairwise comparisons (Holm-Sidak): [C57 11168 > BALB 260.94]; [C57 11168 > C57 260.94]; [C57 11168 > BALB 11168]. Kruskal-Wallis (due to failing normality): ns (P = 0.078). t-test: Significantly more C. jejuni 260.94 recovered from C57 than from BALB mice (P = 0.031). 2-way ANOVA (failed normality): both factors and interaction all ns. Kruskal-Wallis (due to failing normality): ns (P = 0.186). 213 REFERENCES 214 REFERENCES Allos BM, Lippy FT, Carlsen A, Washburn RG, Blaser MJ. 1998. Campylobacter jejuni strains from patients with Guillain-Barre syndrome. Emerging infectious diseases 4:263-268. Ang CW, Jacobs BC, Laman JD. 2004. The Guillain-Barre syndrome: a true case of molecular mimicry. Trends in immunology 25:61-66. Ang CW, Laman JD, Willison HJ, Wagner ER, Endtz HP, De Klerk MA, Tio-Gillen AP, Van den Braak N, Jacobs BC, Van Doorn PA. 2002. Structure of Campylobacter jejuni lipopolysaccharides determines antiganglioside specificity and clinical features of Guillain-Barre and Miller Fisher patients. Infection and immunity 70:1202-1208. Backert S, Hofreuter D. 2013. Molecular methods to investigate adhesion, transmigration, invasion and intracellular survival of the foodborne pathogen Campylobacter jejuni. Journal of microbiological methods 95:8-23. Bax M, Kuijf ML, Heikema AP, van Rijs W, Bruijns SC, Garcia-Vallejo JJ, Crocker PR, Jacobs BC, van Vliet SJ, van Kooyk Y. 2011. Campylobacter jejuni lipooligosaccharides modulate dendritic cell-mediated T cell polarization in a sialic acid linkage-dependent manner. Infection and immunity 79:2681-2689. Bell JA, Jerome JP, Plovanich-Jones AE, Smith EJ, Gettings JR, Kim HY, Landgraf JR, Lefebure T, Kopper JJ, Rathinam VA, St Charles JL, Buffa BA, Brooks AP, Poe SA, Eaton KA, Stanhope MJ, Mansfield LS. 2013. Outcome of infection of C57BL/6 IL-10(-/-) mice with Campylobacter jejuni strains is correlated with genome content of open reading frames up- and down-regulated in vivo. Microbial pathogenesis 54:1-19. Bereswill S, Fischer A, Plickert R, Haag LM, Otto B, Kuhl AA, Dasti JI, Zautner AE, Munoz M, Loddenkemper C, Gross U, Gobel UB, Heimesaat MM. 2011. Novel murine infection models provide deep insights into the "menage a trois" of Campylobacter jejuni, microbiota and host innate immunity. PloS one 6:e20953. Bettelli E, Carrier Y, Gao W, Korn T, Strom TB, Oukka M, Weiner HL, Kuchroo VK. 2006. Reciprocal developmental pathways for the generation of pathogenic effector TH17 and regulatory T cells. Nature 441:235-238. Brooks PT, Brakel KA, Bell JA, Bejcek CE, Gilpin T, Brudvig JM, Mansfield LS. 2017. Transplanted human fecal microbiota enhanced Guillain Barre syndrome autoantibody responses after Campylobacter jejuni infection in C57BL/6 mice. Microbiome 5:92. Chen X, Oppenheim JJ, Howard OM. 2005. BALB/c mice have more CD4+CD25+ T regulatory cells and show greater susceptibility to suppression of their CD4+CD25- responder T cells than C57BL/6 mice. Journal of leukocyte biology 78:114-121. 215 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. Corinti S, Albanesi C, la Sala A, Pastore S, Girolomoni G. 2001. Regulatory activity of autocrine IL-10 on dendritic cell functions. Journal of immunology 166:4312-4318. Edwards LA, Nistala K, Mills DC, Stephenson HN, Zilbauer M, Wren BW, Dorrell N, Lindley KJ, Wedderburn LR, Bajaj-Elliott M. 2010. Delineation of the innate and adaptive T-cell immune outcome in the human host in response to Campylobacter jejuni infection. PloS one 5:e15398. Ethuin F, Delarche C, Gougerot-Pocidalo MA, Eurin B, Jacob L, Chollet-Martin S. 2003. Regulation of interleukin 12 p40 and p70 production by blood and alveolar phagocytes during severe sepsis. Laboratory investigation; a journal of technical methods and pathology 83:1353- 1360. Everest PH, Goossens H, Butzler JP, Lloyd D, Knutton S, Ketley JM, Williams PH. 1992. Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. Journal of medical microbiology 37:319-325. Fauchere JL, Rosenau A, Veron M, Moyen EN, Richard S, Pfister A. 1986. Association with HeLa cells of Campylobacter jejuni and Campylobacter coli isolated from human feces. Infection and immunity 54:283-287. Fox JG, Rogers AB, Whary MT, Ge Z, Taylor NS, Xu S, Horwitz BH, Erdman SE. 2004. Gastroenteritis in NF-kappaB-deficient mice is produced with wild-type Camplyobacter jejuni but not with C. jejuni lacking cytolethal distending toxin despite persistent colonization with both strains. Infection and immunity 72:1116-1125. Godschalk PC, Kuijf ML, Li J, St Michael F, Ang CW, Jacobs BC, Karwaski MF, Brochu D, Moterassed A, Endtz HP, van Belkum A, Gilbert M. 2007. Structural characterization of Campylobacter jejuni lipooligosaccharide outer cores associated with Guillain-Barre and Miller Fisher syndromes. Infection and immunity 75:1245-1254. Goverse G, Olivier BJ, Molenaar R, Knippenberg M, Greuter M, Konijn T, Cook EC, Beijer MR, Fedor DM, den Haan JM, Napoli JL, Bouma G, Mebius RE. 2015. Vitamin A metabolism and mucosal immune function are distinct between BALB/c and C57BL/6 mice. European journal of immunology 45:89-100. Gu L, Tseng S, Horner RM, Tam C, Loda M, Rollins BJ. 2000. Control of TH2 polarization by the chemokine monocyte chemoattractant protein-1. Nature 404:407-411. Haase C, Jorgensen TN, Michelsen BK. 2002. Both exogenous and endogenous interleukin-10 affects the maturation of bone-marrow-derived dendritic cells in vitro and strongly influences T- cell priming in vivo. Immunology 107:489-499. Hadden RD, Karch H, Hartung HP, Zielasek J, Weissbrich B, Schubert J, Weishaupt A, Cornblath DR, Swan AV, Hughes RA, Toyka KV, Plasma Exchange/Sandoglobulin Guillain-Barre Syndrome Trial G. 2001. Preceding infections, immune factors, and outcome in Guillain-Barre syndrome. Neurology 56:758-765. 216 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Hickey TE, Majam G, Guerry P. 2005. Intracellular survival of Campylobacter jejuni in human monocytic cells and induction of apoptotic death by cytholethal distending toxin. Infection and immunity 73:5194-5197. Hu L, Bray MD, Osorio M, Kopecko DJ. 2006. Campylobacter jejuni induces maturation and cytokine production in human dendritic cells. Infection and immunity 74:2697-2705. Inaba K, Inaba M, Romani N, Aya H, Deguchi M, Ikehara S, Muramatsu S, Steinman RM. 1992. Generation of large numbers of dendritic cells from mouse bone marrow cultures supplemented with granulocyte/macrophage colony-stimulating factor. The Journal of experimental medicine 176:1693-1702. Jacobs BC, Koga M, van Rijs W, Geleijns K, van Doorn PA, Willison HJ, Yuki N. 2008. Subclass IgG to motor gangliosides related to infection and clinical course in Guillain-Barre syndrome. Journal of neuroimmunology 194:181-190. Jacobs BC, Rothbarth PH, van der Meche FG, Herbrink P, Schmitz PI, de Klerk MA, van Doorn PA. 1998. The spectrum of antecedent infections in Guillain-Barre syndrome: a case-control study. Neurology 51:1110-1115. Janssen R, Krogfelt KA, Cawthraw SA, van Pelt W, Wagenaar JA, Owen RJ. 2008. Host-pathogen interactions in Campylobacter infections: the host perspective. Clinical microbiology reviews 21:505-518. Jiang X, Shen C, Yu H, Karunakaran KP, Brunham RC. 2010. Differences in innate immune responses correlate with differences in murine susceptibility to Chlamydia muridarum pulmonary infection. Immunology 129:556-566. Joeris T, Muller-Luda K, Agace WW, Mowat AM. 2017. Diversity and functions of intestinal mononuclear phagocytes. Mucosal immunology 10:845-864. Keithlin J, Sargeant J, Thomas MK, Fazil A. 2014. Systematic review and meta-analysis of the proportion of Campylobacter cases that develop chronic sequelae. BMC public health 14:1203. Kiehlbauch JA, Albach RA, Baum LL, Chang KP. 1985. Phagocytosis of Campylobacter jejuni and its intracellular survival in mononuclear phagocytes. Infection and immunity 48:446-451. Koga M, Yuki N, Hirata K, Morimatsu M, Mori M, Kuwabara S. 2003. Anti-GM1 antibody IgG subclass: a clinical recovery predictor in Guillain-Barre syndrome. Neurology 60:1514-1518. Kuroki S, Saida T, Nukina M, Haruta T, Yoshioka M, Kobayashi Y, Nakanishi H. 1993. Campylobacter jejuni strains from patients with Guillain-Barre syndrome belong mostly to Penner serogroup 19 and contain beta-N-acetylglucosamine residues. Annals of neurology 33:243-247. Law BF, Adriance SM, Joens LA. 2009. Comparison of in vitro virulence factors of Campylobacter jejuni to in vivo lesion production. Foodborne pathogens and disease 6:377-385. 217 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Liu T, Matsuguchi T, Tsuboi N, Yajima T, Yoshikai Y. 2002. Differences in expression of toll-like receptors and their reactivities in dendritic cells in BALB/c and C57BL/6 mice. Infection and immunity 70:6638-6645. Liu T, Nishimura H, Matsuguchi T, Yoshikai Y. 2000. Differences in interleukin-12 and -15 production by dendritic cells at the early stage of Listeria monocytogenes infection between BALB/c and C57 BL/6 mice. Cellular immunology 202:31-40. Louwen R, Horst-Kreft D, de Boer AG, van der Graaf L, de Knegt G, Hamersma M, Heikema AP, Timms AR, Jacobs BC, Wagenaar JA, Endtz HP, van der Oost J, Wells JM, Nieuwenhuis EE, van Vliet AH, Willemsen PT, van Baarlen P, van Belkum A. 2013. A novel link between Campylobacter jejuni bacteriophage defence, virulence and Guillain-Barre syndrome. European journal of clinical microbiology & infectious diseases : official publication of the European Society of Clinical Microbiology 32:207-226. Lutz MB, Kukutsch N, Ogilvie AL, Rossner S, Koch F, Romani N, Schuler G. 1999. An advanced culture method for generating large quantities of highly pure dendritic cells from mouse bone marrow. Journal of immunological methods 223:77-92. Malik A, Sharma D, St Charles J, Dybas LA, Mansfield LS. 2014. Contrasting immune responses mediate Campylobacter jejuni-induced colitis and autoimmunity. Mucosal immunology 7:802- 817. Mansfield LS, Bell JA, Wilson DL, Murphy AJ, Elsheikha HM, Rathinam VA, Fierro BR, Linz JE, Young VB. 2007. C57BL/6 and congenic interleukin-10-deficient mice can serve as models of Campylobacter jejuni colonization and enteritis. Infection and immunity 75:1099-1115. Mansfield LS, Patterson JS, Fierro BR, Murphy AJ, Rathinam VA, Kopper JJ, Barbu NI, Onifade TJ, Bell JA. 2008. Genetic background of IL-10(-/-) mice alters host-pathogen interactions with Campylobacter jejuni and influences disease phenotype. Microbial pathogenesis 45:241-257. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. 2000. M-1/M-2 macrophages and the Th1/Th2 paradigm. Journal of immunology 164:6166-6173. Moran AP, Annuk H, Prendergast MM. 2005. Antibodies induced by ganglioside-mimicking Campylobacter jejuni lipooligosaccharides recognise epitopes at the nodes of Ranvier. Journal of neuroimmunology 165:179-185. Murray PJ, Wynn TA. 2011. Protective and pathogenic functions of macrophage subsets. Nature reviews Immunology 11:723-737. Na YR, Jung D, Gu GJ, Seok SH. 2016. GM-CSF Grown Bone Marrow Derived Cells Are Composed of Phenotypically Different Dendritic Cells and Macrophages. Molecules and cells 39:734-741. Nachamkin I, Allos BM, Ho T. 1998. Campylobacter species and Guillain-Barre syndrome. Clinical microbiology reviews 11:555-567. 218 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. Prendergast MM, Lastovica AJ, Moran AP. 1998. Lipopolysaccharides from Campylobacter jejuni O:41 strains associated with Guillain-Barre syndrome exhibit mimicry of GM1 ganglioside. Infection and immunity 66:3649-3655. Prendergast MM, Moran AP. 2000. Lipopolysaccharides in the development of the Guillain- Barre syndrome and Miller Fisher syndrome forms of acute inflammatory peripheral neuropathies. Journal of endotoxin research 6:341-359. Rathinam VA, Hoag KA, Mansfield LS. 2008. Dendritic cells from C57BL/6 mice undergo activation and induce Th1-effector cell responses against Campylobacter jejuni. Microbes and infection 10:1316-1324. Rees JH, Gregson NA, Hughes RA. 1995. Anti-ganglioside GM1 antibodies in Guillain-Barre syndrome and their relationship to Campylobacter jejuni infection. Annals of neurology 38:809- 816. Reiner SL, Locksley RM. 1995. The regulation of immunity to Leishmania major. Annual review of immunology 13:151-177. Samuelson DR, Eucker TP, Bell JA, Dybas L, Mansfield LS, Konkel ME. 2013. The Campylobacter jejuni CiaD effector protein activates MAP kinase signaling pathways and is required for the development of disease. Cell communication and signaling : CCS 11:79. Scallan E, Hoekstra RM, Angulo FJ, Tauxe RV, Widdowson MA, Roy SL, Jones JL, Griffin PM. 2011. Foodborne illness acquired in the United States--major pathogens. Emerging infectious diseases 17:7-15. Sheikh KA, Nachamkin I, Ho TW, Willison HJ, Veitch J, Ung H, Nicholson M, Li CY, Wu HS, Shen BQ, Cornblath DR, Asbury AK, McKhann GM, Griffin JW. 1998. Campylobacter jejuni lipopolysaccharides in Guillain-Barre syndrome: molecular mimicry and host susceptibility. Neurology 51:371-378. St Charles JL, Bell JA, Gadsden BJ, Malik A, Cooke H, Van de Grift LK, Kim HY, Smith EJ, Mansfield LS. 2017. Guillain Barre Syndrome is induced in Non-Obese Diabetic (NOD) mice following Campylobacter jejuni infection and is exacerbated by antibiotics. Journal of autoimmunity 77:11-38. van den Berg B, Walgaard C, Drenthen J, Fokke C, Jacobs BC, van Doorn PA. 2014. Guillain- Barre syndrome: pathogenesis, diagnosis, treatment and prognosis. Nature reviews Neurology 10:469-482. 57. Walsh KP, Mills KH. 2013. Dendritic cells and other innate determinants of T helper cell polarisation. Trends in immunology 34:521-530. 58. Wassenaar TM, Engelskirchen M, Park S, Lastovica A. 1997. Differential uptake and killing potential of Campylobacter jejuni by human peripheral monocytes/macrophages. Medical microbiology and immunology 186:139-144. 219 59. Watanabe H, Numata K, Ito T, Takagi K, Matsukawa A. 2004. Innate immune response in Th1- and Th2-dominant mouse strains. Shock 22:460-466. 60. Watson RO, Galan JE. 2008. Campylobacter jejuni survives within epithelial cells by avoiding delivery to lysosomes. PLoS pathogens 4:e14. 61. WHO. [Internet]. 2016. Campylobacter Fact Sheet. Available at: http://www.who.int/mediacentre/factsheets/fs255/en/. 62. 63. Yang F, Wang D, Li Y, Sang L, Zhu J, Wang J, Wei B, Lu C, Sun X. 2017. Th1/Th2 Balance and Th17/Treg-Mediated Immunity in relation to Murine Resistance to Dextran Sulfate-Induced Colitis. Journal of immunology research 2017:7047201. Yuki N, Taki T, Inagaki F, Kasama T, Takahashi M, Saito K, Handa S, Miyatake T. 1993. A bacterium lipopolysaccharide that elicits Guillain-Barre syndrome has a GM1 ganglioside-like structure. The Journal of experimental medicine 178:1771-1775. 220 CHAPTER 5: SUMMARY AND FUTURE DIRECTIONS SUMMARY Campylobacter jejuni is an important cause of bacterial gastroenteritis worldwide and is associated with numerous post-infectious sequelae. C. jejuni strains vary widely in factors determining pathogenicity, including invasive potential, ability to stably colonize a host, and expression of ganglioside mimics in the lipooligosaccharide. Similarly, host factors including microbiota and genetic background predisposing to immunological biases are highly variable. In people, C. jejuni most typically causes a severe but self-limiting gastroenteritis.5 Development of post-infectious complications, including the peripheral neuropathy Guillain-Barré syndrome (GBS), occurs in a small proportion of cases.12 The immunopathogenesis and factors determining susceptibility to GBS are incompletely understood, but both infecting C. jejuni strain characteristics and host factors such as immunogenetic background are thought to contribute. The complex interplay of pathogen and host, beginning with the initial host-microbe interaction, determines innate and adaptive immunity and corresponding disease outcomes. The overarching aim of this study was to determine how these interactions underlie contrasting immune responses to C. jejuni in mice. C. jejuni strains associated with colitis (C. jejuni 11168) and GBS (C. jejuni 260.94) have resulted in contrasting adaptive immune responses within C57BL/6 IL-10-/- mice.15 BALB/c and C57BL/6 are two commonly studied laboratory mouse strains, chosen specifically for this study because of documented immunological biases. In many models, C57BL/6 mice exhibit a pro-inflammatory and Th1 bias. In contrast, BALB/c mice are reportedly Th2-biased, in addition to harboring a greater proportion of immunomodulatory T regulatory (Treg) cells with greater suppressive potential on responding T cells.6; 9; 18; 20; 25 Thus, this system was designed to evaluate the interplay of infecting C. jejuni strain and host 221 characteristics, beginning with the initial interaction of C. jejuni with dendritic cells, and assessing the impact on development of innate and adaptive immunity and disease outcome. Dendritic cells (DCs) are sentinel antigen presenting cells located throughout the body. These cells continually surveil the local environment and upon interaction with antigen, undergo a maturation process culminating in migration to local lymph nodes and interaction with naïve T cells. This interaction involves presentation of antigen through peptide-MHC II complexes with the T cell receptor, costimulatory molecule expression for T cell activation, and cytokine production. Depending upon the signals generated during this interaction, the Th cell differentiation is polarized toward certain subsets, including Th1, Th2, Th17, or Treg cells; mixed responses are also possible. DCs are thus a vital bridge between innate and adaptive immunity, driving initial polarization of immunity that subsequently determines disease outcomes. An in vitro culture system was employed to evaluate host genetic background, in addition to differences in C. jejuni strain invasion efficiency and intracellular survival capability, in the initial host- microbe interaction. Granulocyte-macrophage colony-stimulating factor (GM-CSF) was used to obtain bone marrow-derived DCs (BMDCs) from hematopoietic precursors. BMDCs from wild-type (WT) and IL- 10-/- mice on both C57BL/6 and BALB/c backgrounds were used in this study. Confirmation of BMDC phenotype in differentiated cells was performed by flow cytometric analysis of CD11c, MHC II, and F4/80 expression. Interestingly, populations differentiated from stem cells of C57BL/6 IL-10-/- and BALB/c IL-10-/-mice have a high percentage of cells expressing CD11c but differ in proportions of immature DCs, mature DCs, and macrophages within final populations. BMDCs from C57BL/6 IL-10-/- mice reflected a more mixed population of immature and mature DCs along with approximately 25% macrophages, while BALB/c IL-10-/- populations comprised a higher proportion of more mature DCs and approximately 16% macrophages. Presence or absence of IL-10 did not appear to impact differentiation 222 within either C57BL/6 or BALB/c mice, but may affect maturation and function following stimulus with C. jejuni or LPS. Invasion efficiency and intracellular survival of C. jejuni strains 11168 and 260.94 in BMDCs derived from C57BL/6 and BALB/c mice were assessed by gentamicin killing assay. In invasion assays using BMDCs derived from IL-10-/- mice, C. jejuni 11168 exhibited higher combined invasion efficiency and intracellular survival compared to C. jejuni 260.94, with the difference becoming more pronounced with increased infection time. Similarly, C. jejuni 11168 exhibited higher intracellular survival in BMDCs derived from both WT and IL-10-/- mice, compared to C. jejuni 260.94. The enhanced invasion and survival of C. jejuni 11168 compared to C. jejuni 260.94 observed in vitro correlates with previous comparison of genes associated with invasion, adherence, and acid resistance in these two strains,1 and with immunohistochemical labeling of the two C. jejuni strains in the proximal colon of BALB/c mice (Brudvig, Chapter 3). Thus, BMDCs provide a useful model for studying C. jejuni strain characteristics in vitro. In these gentamicin killing assays, differences in C. jejuni recovery between mouse genotypes were less consistent than differences identified between C. jejuni strains. The most consistent pattern that emerged was higher intracellular viability of C. jejuni 11168 in C57BL/6-derived BMDCs. This pattern likely reflects a combination of both enhanced invasion and survival of C. jejuni 11168 compared to C. jejuni 260.94, and also a relatively higher proportion of immature DCs and macrophages in C57BL/6 compared to BALB/c populations. These findings emphasize the important interplay between host cells and C. jejuni strains during the initial interaction. Cytokine production by DCs is an important signal for Th cell polarization. The final characterization of the host-microbe interaction was assessed by cytokine production of BMDCs from IL- 10-/- mice infected with either C. jejuni 11168 or C. jejuni 260.94. Heightened production of the chemokine MCP-1 in BMDCs from C57BL/6 IL-10-/- mice, especially following infection with C. jejuni 223 260.94, reflects both host cell and C. jejuni strain characteristics. Macrophages are primary producers of MCP-1, and C57BL/6 IL-10-/- cells contained a higher proportion of macrophages than BALB/c IL-10-/- cells. MCP-1 was shown to be an important determinant for Th2 polarization,10 and Th2-mediated immunity was previously reported in C57BL/6 IL-10-/- mice infected with C. jejuni 260.94.15 Therefore, early production of MCP-1 by DCs may contribute to Th2 polarization in some models. Similarly, BMDCs from BALB/c IL-10-/- mice exhibited enhanced production of IL-6, especially following infection with C. jejuni 11168. As strong Th1/Th17 immunity was seen in vivo in BALB/c IL-10-/- mice following infection with C. jejuni 11168 (Brudvig, Chapter 3), these results further support the early contributions of DCs in determining adaptive immunity and disease outcome. In addition to in vitro methods to assess how initial host-microbe interactions help determine immunity, in vivo mouse models were also used to assess both immune response and pathology mimicking human disease. In the first study, WT and IL-10-/- BALB/c mice were infected with C. jejuni 260.94 to investigate the potential for a new GBS model. Considering the previous Th2-mediated responses in C57BL/6 IL-10-/- mice infected with C. jejuni 260.9415 and the reported relative Th2-bias in BALB/c compared to C57BL/6 mice, BALB/c mice were expected to mount strong Th2-mediated immunity. Production of anti-ganglioside antibodies, a hallmark of GBS, and development of neurological deficits and accompanying inflammatory lesions in dorsal root ganglia were expected. Instead, both WT and IL-10-/- BALB/c mice infected with C. jejuni 260.94 exhibited systemic Th1/Th17 responses, manifested by significant increases in C. jejuni-specific IgG2a, IgG2b, and IgG3 plasma antibodies. Infected mice also did not develop anti-ganglioside antibodies or peripheral nerve lesions as induced with C. jejuni 260.94 in other mouse strains.15; 23 Systemic Th1/Th17 responses were exacerbated by IL-10 deficiency, but even in WT mice no Th2 response was seen. Results of this study further support the impact of host genetic background in determining adaptive immunity and susceptibility to GBS. With the results of this study, different immune responses had now been shown 224 between C. jejuni strains 11168 and 260.94 within C57BL/6 IL-10-/- mice,15 and between BALB/c IL-10-/- and C57BL/6 IL-10-/- mice infected with C. jejuni 260.94 (Malik, et al. 201415 and Brudvig, Chapter 2). Therefore, the final study aimed to further investigate the in vivo interplay between host genetic background and C. jejuni strain characteristics in a single system. C57BL/6 IL-10-/- and BALB/c IL-10-/- mice were chosen due to contrasting immune responses to C. jejuni strains in our models, and to respective Th1 and Th2 immunological biases reported in the literature. C. jejuni strains 11168 and 260.94 are associated with colitis and GBS, respectively, and were shown to vary in colitogenic potential and magnitude of immune response.1 In the current 4-week study, BALB/c IL-10-/- mice infected with C. jejuni 11168 showed the lowest survivorship, most severe colitis, marked systemic and local Th1/Th17-mediated immunity reflected by C. jejuni-specific plasma antibodies and increased IFN-γ, TNF-α, and IL-22 production in the proximal colon, and robust mucosal immunity reflected by C. jejuni-specific IgA in fecal supernatants. C. jejuni 260.94-infected BALB/c IL-10-/- mice also mounted systemic Th1/Th17-mediated immune responses, but did not develop colitis or marked colonic or mucosal immunity. Interestingly, significant production of anti-GM1 and anti-GD1a ganglioside antibodies was seen only in BALB/c IL-10-/- mice infected with C. jejuni 11168, a strain associated with enteritis and not GBS, while infection with C. jejuni 260.94 did not induce anti- ganglioside antibody production. The anti -GM1 and -GD1a plasma antibodies produced were Th1/Th17- associated IgG2a and IgG2b isotypes, and thus the possibility that the marked systemic and local immunity induced by C. jejuni 11168 in BALB/c IL-10-/- mice was necessary for anti-ganglioside antibody production in this model should be considered. For BALB/c IL-10-/- mice, results of the first in vivo study were confirmed, in that C. jejuni 260.94 induced Th1/Th17 systemic responses, but without colitis or anti-ganglioside antibody production, and no Th2 component was seen. BALB/c IL-10-/- mice infected with C. jejuni 11168 exhibited strong Th1/Th17-mediated immunity and colitis, and provide an avenue for studying C. jejuni pathogenesis in mice of an additional genetic background. 225 Contrary to the striking immune response and colitis seen in C. jejuni 11168-infected BALB/c IL- 10-/- mice, C57BL/6 IL-10-/- mice infected with either C. jejuni strain demonstrated atypical results compared to previous studies by our group.1; 2; 15-17; 21 Despite colonization and an enlarged regional lymph node in many infected mice, virtually no systemic, mucosal, or local colonic immune response was seen following infection with either C. jejuni strain, and the mice did not develop the colitis or anti- ganglioside antibodies following C. jejuni 11168 or 260.94 infection, respectively, as in other models.1; 15- 17 Interestingly, Lactobacillus murinus was cultured from all C57BL/6 IL-10-/- mice, but no BALB/c IL-10-/- mice, at the end of the study. The presence of L. murinus was unexpected and the source is unknown, as is why it was cultured only from C57BL/6 IL-10-/- and not BALB/c IL-10-/- mice. Both genotypes of mice received the same C. jejuni inocula, which were pure and motile C. jejuni cultures based upon wet mount and Gram stain preparations. Lactobacilli comprise one group of bacteria currently being investigated for potential probiotic effects. Different Lactobacillus spp. have demonstrated protective effects against development of spontaneous colitis in IL-10-/- mice.14; 19 Furthermore, both in vitro and in vivo studies also have demonstrated antagonism of Lactobacillus spp. against C. jejuni. 3; 8; 22; 24 Infection of C57BL/6 IL-10-/- mice with C. jejuni 11168 has provided a repeatable model of colitis with few exceptions.1; 15-17; 21 Thus, the relative protection exhibited by C57BL/6 IL-10-/- mice in this study, with concurrent carriage of L. murinus, warrants further study. The current study adds to the relatively limited knowledge of the combination of host factors and infecting C. jejuni strain in determining disease outcome. The relative protection of C57BL/6 IL-10-/- mice from C. jejuni-induced pathology, compared to BALB/c IL-10-/- mice in the current study and to C57BL/6 IL-10-/- mice in previous studies, indicates a potential probiotic effect of L. murinus. Overall, our results indicate that BALB/c mice, with and without IL-10, respond to C. jejuni infection with Th1/Th17- but not Th2-mediated responses. The magnitude of the immune response and induction of colitis is 226 dependent upon infecting strain, with C. jejuni 11168 showing higher pathogenicity. In vitro studies showed that the colitogenic C. jejuni 11168 exhibits enhanced invasion and intracellular survival in dendritic cells compared to GBS-associated C. jejuni 260.94, a finding reflected by in vivo immunohistochemical labeling of these two C. jejuni strains in the ileocecocolic junction. While developing an in vitro system for assessment of the initial C. jejuni-dendritic cell interaction, important differences were found between C57BL/6 and BALB/c mice in differentiation of hematopoietic stem cells into dendritic cells. Functional differences in cytokine production and ability to kill intracellular C. jejuni were consistent with the different BMDC phenotypes. Cytokines produced by infected IL-10-/- BMDCs correlate with in vivo responses in the current study (Brudvig, Chapter 3) and reported previously,15 indicating that DCs play an important role in polarizing the immune response. This study provides critical insight into the complex interplay of C. jejuni strain characteristics and host factors, beginning with the contribution of dendritic cells in the initial host-microbe interaction. 227 FUTURE DIRECTIONS What is the effect of IL-10 on DC maturation and function between BMDCs derived from C57BL/6 and BALB/c mice? Presence or absence of IL-10 did not markedly affect the differentiation of hematopoietic stem cells derived from either C57BL/6 or BALB/c mice as assessed by CD11c and MHC II expression. At the end of the 9-11 day differentiation period, cells derived from both WT and IL-10-/- mice on their respective C57BL/6 or BALB/c backgrounds exhibited similar proportions of CD11c(+) cells and did not differ markedly in MHC II expression with or without IL-10. IL-10 has been shown to affect both spontaneous and induced DC maturation, in addition to pro-inflammatory cytokine production and T- cell stimulating ability.7; 11 Therefore, while IL-10 apparently does not substantially impact differentiation of stem cells into BMDCs, it would be informative to determine if there is a differential effect exerted by IL-10 on DC maturation and function in C57BL/6 compared to BALB/c mice. To test this, BMDCs derived from WT and IL-10-/- on C57BL/6 and BALB/c backgrounds could be cultured as before for 10 days and stimulated with LPS. Expression of MHC II in addition to costimulatory molecules CD80 and CD86 analyzed by flow cytometry would indicate differences in maturation. Differences in pro-inflammatory cytokine production and putative T-cell stimulating ability determined by measurement of cytokines including TNF-α, IFN-γ, IL-6, IL-12, and IL-4 would reflect effects of IL-10 on DC function. These experiments could also include infection with different C. jejuni strains, to determine if C. jejuni-induced DC maturation and function is variable depending upon presence or absence of IL-10. Another informative experiment would be to assess the intracellular survival of C. jejuni strains 11168 and 260.94 in BMDCs derived from WT and IL-10-/- on the same genetic background. In the current study, BMDCs from WT C57BL/6 and BALB/c mice were compared in single experiments, and BMDCs from IL-10-/- C57BL/6 and BALB/c mice were compared in single experiments. Killing ability of WT versus IL-10-/- cells could not be compared statistically for this reason, but the patterns suggest that IL- 228 10-/- cells exhibit enhanced killing of C. jejuni. This would be consistent with increased maturation and production of pro-inflammatory cytokines induced by neutralization of IL-10 in human DC studies.7 To further evaluate the effect of IL-10 on ability of DCs to kill different C. jejuni strains, BMDCs from WT and IL-10-/- C57BL/6 mice can be compared, and BMDCs from WT and IL-10-/- BALB/c mice can be compared, in separate experiments. It is logistically prohibitive to test four genotypes of mice with two C. jejuni strains in a single gentamicin killing assay. Thus, the question of effect of IL-10 within the separate genotypes should be tested independently. Further experiments can then be designed based on these results. Is increased viable intracellular C. jejuni 11168 compared to C. jejuni 260.94 in BMDCs, determined by gentamicin killing assay, reflecting enhanced survival more than invasion? In assays using BMDCs from IL-10-/- mice, viable intracellular C. jejuni 11168 was not different from C. jejuni 260.94 after 1 hour of invasion, gentamicin treatment, and immediate lysis. However, C. jejuni 11168 was recovered in significantly higher numbers than C. jejuni 260.94 from C57BL/6 IL-10-/-- derived BMDCs with the same infection time of 1 hour followed by gentamicin treatment, but without lysis until 24 hours p.i. This implies that the increased intracellular C. jejuni 11168 over time may be more due to intracellular survival and replication than invasion. In order to further delineate the relative contributions of invasion and survival leading to the marked increase in C. jejuni 11168 recovery over time, time-course experiments performed with 1 hour of infection, gentamicin treatment, and then lysis of the BMDCs at various time points such as 2, 4, 8, and 12 hours would further characterize the survival and replication over time. It would be logistically prohibitive to test all mouse genotype/C. jejuni strain combinations with sufficient replicate wells in a single assay by this method; thus individual assays testing single combinations could be performed, and identification of broad patterns can direct further studies. 229 What genes are involved in enhanced invasion and survival of C. jejuni 11168 compared to C. jejuni 260.94? Multiple genes with putative functions including adherence, colonization, invasion, acid resistance, the LOS, and motility present in C. jejuni 11168 are divergent or altogether absent in C. jejuni 260.94.1 Differences between these C. jejuni strains in invasion and intracellular survival in vitro may be due to any one gene or a combination. To further determine if one of the divergent or absent genes is enhancing C. jejuni 11168 survival or invasion compared to C. jejuni 260.94, mutant C. jejuni 11168 strains lacking one of these genes could be generated and tested in vitro for a reduction in invasion or survival; rescue of the phenotype by complementation would further confirm a role for the particular gene. What effect does colonization with Lactobacillus murinus have on development of spontaneous colitis in IL-10-/- mice? In the second in vivo model (Chapter 3), Lactobacillus murinus was cultured from all C57BL/6 IL- 10-/- mice that also appeared to be protected from C. jejuni-induced pathology. L. reuteri and L. salivarius ssp. salivarius have demonstrated protective effects against development of spontaneous colitis in 129 Sv/Ev IL-10-/- and C57BL/6 IL-10-/- mice.14; 19 IL-10-/- mice can develop spontaneous colitis as early as 3-4 weeks of age, with variable severity depending upon housing, and with more severe lesions in BALB/c than C57BL/6 mice.4; 13 It is worth noting that in the current study, 3/10 sham-inoculated BALB/c mice exhibited mild spontaneous colitis and 1/10 sham-inoculated C57BL/6 mice had mild colitis. The mice were inoculated at a younger age than in the first in vivo experiment (Brudvig, Chapter 2) in order to avoid the confounding spontaneous colitis seen with older age. Mice in the second in vivo model (Brudvig, Chapter 3) were euthanized at approximately 10 weeks of age, and the higher rate of spontaneous colitis in BALB/c mice may only reflect the previously reported pattern of increased 230 severity compared to C57BL/6 mice. However, as the C57BL/6 mice harboring L. murinus may have also been protected from C. jejuni-induced disease, the possibility that L. murinus also conferred protection against spontaneous colitis should be considered. Standardized experiments specifically designed to examine the prevalence of spontaneous colitis in IL-10-/- mice of both BALB/c and C57BL/6 backgrounds with and without L. murinus carriage would be necessary to further test the possibility that L. murinus abrogated spontaneous colitis in C57BL/6 IL-10-/- mice. For descriptive purposes, C57BL/6 IL-10-/-mice in the colony could be periodically tested for L. murinus by culture, monitored over time for clinical signs, and assessed histologically at sacrifice to determine if there was a correlation between L. murinus carriage and delayed onset or less severe spontaneous colitis. Controlled studies could also be performed using BALB/c IL-10-/- mice verified to be free of L. murinus in the beginning of the study. A placebo-controlled trial could then be conducted19 to determine if L. murinus was associated with reduced spontaneous colitis. Is there an inhibitory, strain-specific interaction between L. murinus and C. jejuni strains 11168 and 260.94 in vitro? Prior to in vivo experiments, in vitro assays should be conducted as further confirmation of an inhibitory effect of L. murinus on C. jejuni growth and invasiveness. The gentamicin killing assay performed in this study could incorporate pre-treatment with L. murinus prior to infection as previously described,24 followed by assessment of C. jejuni invasion and survival in BMDCs. In vivo immunohistochemical labeling of C. jejuni showed no C. jejuni 11168 was present beneath the surface epithelium in C57BL/6 IL-10-/- mice, while infected BALB/c IL-10-/- mice showed intracellular labeling in macrophages/dendritic cells within both the lamina propria and submucosa (Brudvig, Chapter 3). Therefore, using an epithelial line such as young adult mouse colon cells (YAMC) pretreated with L. 231 murinus prior to invasion assays would further characterize the interplay of C. jejuni and L. murinus at the epithelium. Testing antimicrobial activity of L. murinus against C. jejuni strains 11168 and 260.94 by previously reported methods22 would also be informative. L. murinus cultures can be spotted onto agar plates and allowed to grow, with subsequent overlaying of broth and agar inoculated with C. jejuni. Zones of inhibition would represent inhibitory effects of L. murinus on C. jejuni growth. Is Lactobacillus murinus protective against C. jejuni-mediated immunopathology in vivo? Antagonism of pathogens by probiotic bacteria can depend upon the pathogen, the probiotic, and the environment. If results of in vitro invasion and inhibition assays indicate specificity of L. murinus against C. jejuni, in vivo experiments specifically designed to determine if L. murinus ameliorates C. jejuni-induced colitis can be performed. Because BALB/c IL-10-/- mice were shown to be free of L. murinus and susceptible to severe C. jejuni 11168-induced colitis (Brudvig, Chapter 3), these mice provide an ideal system. Following confirmation of L. murinus negativity, BALB/c IL-10-/- mice can be either prophylactically or therapeutically treated with L. murinus, with and without C. jejuni inoculation. Effects of L. murinus on C. jejuni colonization and invasiveness and effects on systemic and mucosal immunity and local cytokine production can then be used to further support or refute that L. murinus protected C57BL/6 IL-10-/- mice from C. jejuni-induced pathology in the current study. 232 REFERENCES 233 REFERENCES Bell JA, Jerome JP, Plovanich-Jones AE, Smith EJ, Gettings JR, Kim HY, Landgraf JR, Lefebure T, Kopper JJ, Rathinam VA, St Charles JL, Buffa BA, Brooks AP, Poe SA, Eaton KA, Stanhope MJ, Mansfield LS. 2013. Outcome of infection of C57BL/6 IL-10(-/-) mice with Campylobacter jejuni strains is correlated with genome content of open reading frames up- and down-regulated in vivo. Microbial pathogenesis 54:1-19. Bell JA, St Charles JL, Murphy AJ, Rathinam VA, Plovanich-Jones AE, Stanley EL, Wolf JE, Gettings JR, Whittam TS, Mansfield LS. 2009. Multiple factors interact to produce responses resembling spectrum of human disease in Campylobacter jejuni infected C57BL/6 IL-10-/- mice. BMC microbiology 9:57. Bereswill S, Ekmekciu I, Escher U, Fiebiger U, Stingl K, Heimesaat MM. 2017. Lactobacillus johnsonii ameliorates intestinal, extra-intestinal and systemic pro-inflammatory immune responses following murine Campylobacter jejuni infection. Scientific reports 7:2138. Berg DJ, Davidson N, Kuhn R, Muller W, Menon S, Holland G, Thompson-Snipes L, Leach MW, Rennick D. 1996. Enterocolitis and colon cancer in interleukin-10-deficient mice are associated with aberrant cytokine production and CD4(+) TH1-like responses. The Journal of clinical investigation 98:1010-1020. Blaser M, Engberg J. 2008. Clinical Aspects of Campylobacter jejuni and Campylobacter coli Infections. Campylobacter, 3rd ed. Washington, D.C.: ASM Press. Chen X, Oppenheim JJ, Howard OM. 2005. BALB/c mice have more CD4+CD25+ T regulatory cells and show greater susceptibility to suppression of their CD4+CD25- responder T cells than C57BL/6 mice. Journal of leukocyte biology 78:114-121. Corinti S, Albanesi C, la Sala A, Pastore S, Girolomoni G. 2001. Regulatory activity of autocrine IL-10 on dendritic cell functions. Journal of immunology 166:4312-4318. Ding W, Wang H, Griffiths MW. 2005. Probiotics down-regulate flaA sigma28 promoter in Campylobacter jejuni. Journal of food protection 68:2295-2300. Goverse G, Olivier BJ, Molenaar R, Knippenberg M, Greuter M, Konijn T, Cook EC, Beijer MR, Fedor DM, den Haan JM, Napoli JL, Bouma G, Mebius RE. 2015. Vitamin A metabolism and mucosal immune function are distinct between BALB/c and C57BL/6 mice. European journal of immunology 45:89-100. Gu L, Tseng S, Horner RM, Tam C, Loda M, Rollins BJ. 2000. Control of TH2 polarization by the chemokine monocyte chemoattractant protein-1. Nature 404:407-411. Haase C, Jorgensen TN, Michelsen BK. 2002. Both exogenous and endogenous interleukin-10 affects the maturation of bone-marrow-derived dendritic cells in vitro and strongly influences T- cell priming in vivo. Immunology 107:489-499. 234 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. Keithlin J, Sargeant J, Thomas MK, Fazil A. 2014. Systematic review and meta-analysis of the proportion of Campylobacter cases that develop chronic sequelae. BMC public health 14:1203. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. 1993. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 75:263-274. Madsen KL, Doyle JS, Jewell LD, Tavernini MM, Fedorak RN. 1999. Lactobacillus species prevents colitis in interleukin 10 gene-deficient mice. Gastroenterology 116:1107-1114. Malik A, Sharma D, St Charles J, Dybas LA, Mansfield LS. 2014. Contrasting immune responses mediate Campylobacter jejuni-induced colitis and autoimmunity. Mucosal immunology 7:802- 817. Mansfield LS, Bell JA, Wilson DL, Murphy AJ, Elsheikha HM, Rathinam VA, Fierro BR, Linz JE, Young VB. 2007. C57BL/6 and congenic interleukin-10-deficient mice can serve as models of Campylobacter jejuni colonization and enteritis. Infection and immunity 75:1099-1115. Mansfield LS, Patterson JS, Fierro BR, Murphy AJ, Rathinam VA, Kopper JJ, Barbu NI, Onifade TJ, Bell JA. 2008. Genetic background of IL-10(-/-) mice alters host-pathogen interactions with Campylobacter jejuni and influences disease phenotype. Microbial pathogenesis 45:241-257. Mills CD, Kincaid K, Alt JM, Heilman MJ, Hill AM. 2000. M-1/M-2 macrophages and the Th1/Th2 paradigm. Journal of immunology 164:6166-6173. O'Mahony L, Feeney M, O'Halloran S, Murphy L, Kiely B, Fitzgibbon J, Lee G, O'Sullivan G, Shanahan F, Collins JK. 2001. Probiotic impact on microbial flora, inflammation and tumour development in IL-10 knockout mice. Alimentary pharmacology & therapeutics 15:1219-1225. Reiner SL, Locksley RM. 1995. The regulation of immunity to Leishmania major. Annual review of immunology 13:151-177. Samuelson DR, Eucker TP, Bell JA, Dybas L, Mansfield LS, Konkel ME. 2013. The Campylobacter jejuni CiaD effector protein activates MAP kinase signaling pathways and is required for the development of disease. Cell communication and signaling : CCS 11:79. Santini C, Baffoni L, Gaggia F, Granata M, Gasbarri R, Di Gioia D, Biavati B. 2010. Characterization of probiotic strains: an application as feed additives in poultry against Campylobacter jejuni. International journal of food microbiology 141 Suppl 1:S98-108. St Charles JL, Bell JA, Gadsden BJ, Malik A, Cooke H, Van de Grift LK, Kim HY, Smith EJ, Mansfield LS. 2017. Guillain Barre Syndrome is induced in Non-Obese Diabetic (NOD) mice following Campylobacter jejuni infection and is exacerbated by antibiotics. Journal of autoimmunity 77:11-38. 24. Wine E, Gareau MG, Johnson-Henry K, Sherman PM. 2009. Strain-specific probiotic (Lactobacillus helveticus) inhibition of Campylobacter jejuni invasion of human intestinal epithelial cells. FEMS microbiology letters 300:146-152. 235 Yang F, Wang D, Li Y, Sang L, Zhu J, Wang J, Wei B, Lu C, Sun X. 2017. Th1/Th2 Balance and Th17/Treg-Mediated Immunity in relation to Murine Resistance to Dextran Sulfate-Induced Colitis. Journal of immunology research 2017:7047201. 25. 236