A Dissertation Entitled Role of Bb-Elicited IL-10 in Suppression Of

Home , Nicotinamidase

A Dissertation

entitled

Role of Bb-elicited IL-10 in Suppression of Innate Immune Responses within Murine

Skin Tissue

Muhammed Saad Abdul Aziz Moledina

Submitted to the Graduate Faculty as partial fulfillment of the requirements for the

Doctor of Philosophy Degree in Biomedical Science

______Dr. R Mark Wooten, Committee Chair

______Dr. Andrea Kalinoski, Committee Member

______Dr. Kevin Pan, Committee Member

______Dr. Randall Worth, Committee Member

______Dr. Robert Blumenthal, Committee Member

______Dr. Cyndee Gruden, Dean College of Graduate Studies

The University of Toledo

August 2019

This document is copyrighted material. Under copyright law, no parts of this document may be reproduced without the expressed permission of the author.

An Abstract of

Role of Bb-elicited IL-10 in Suppression of Innate Immune Responses within Murine Skin Tissue

Muhammed Saad Abdul Aziz Moledina

Submitted to the Graduate Faculty as partial fulfillment of the requirements for the Doctor of Philosophy Degree in Biomedical Science

The University of Toledo August 2019

Borrelia burgdorferi (Bb) is an extracellular spirochetal bacterium and the causative agent of Lyme disease (LD). In vitro, Bb induces robust immune responses which suggests that Bb could be effectively cleared. However, neither innate nor adaptive immune responses are able to eliminate Bb in vivo resulting in persistent infection. Hence, there is a dysregulation of the immune response occurring in vivo, which cannot be recapitulated in an in vitro system due to the intricacy and complexity of the host tissue milieu. In this study, we utilized intravital microscopy (IVM) along with several transgenic mouse lines, to delineate and identify the immune dysregulation occurring within skin tissues in vivo, which permits this extracellular pathogen to maintain a persistent infection.

Our IVM studies identified that Bb undergo a massive proliferation between days

2 to 7 post-infection and peak bacterial numbers are achieved in skin tissues adjacent to the infection site around day 8 post-infection. During this period of acute infection, an activated innate response should be identifying and clearing this extracelluar pathogen, since Bb possess potent agonists. However, host innate responses fail to control the spirochete and Bb numbers undergo almost a 50-fold increase during this period.

iii Attainment of peak bacterial numbers is followed by a significant decrease in Bb numbers until day 14 post-infection; subsequently, this low level is maintained for over 2 years.

Based on this in vivo persistence data, the two major goals of our study were to 1) delineate whether adaptive immune responses were responsible for the significant decrease in Bb numbers after day 8 post-infection, and 2) determine whether Bb-elicited IL-10 is involved in the dysregulation of innate responses necessary to control Bb numbers during early infection.

To test the contribution of the host adaptive responses in reducing Bb numbers, we used several transgenic mice with B cell- and T cell-deficiencies, and compared Bb persistence between these and WT mice using IVM. Our studies revealed that, 1) T cells are necessary to promote an optimal Bb-specific antibody response, 2) both B cells and T cells play a role in controlling the kinetics associated with Bb persistence, but offer minimal contribution in controlling Bb persistence long-term, and 3) Bb-specific antibodies are not responsible for the decrease in Bb number after day 8 post-infection.

Previous investigation of Bb-elicited innate immune responses identified a rapid and potent interleukin-10 (IL-10) response that appears to affect Bb clearance. We hypothesized that Bb-elicited IL-10 was a cause of the innate immune dysregulation that promotes Bb persistence in vivo. IVM studies using an IL-10 reporter (tiger) mice indicated that macrophages and dermal dendritic cells are primary producers of IL-10 during active

Bb infection in skin. Use of LysM+ and Iaβ+ mice on an IL-10-/- background allowed us to visualize the effect of this Bb-elicited IL-10 on neutrophil and macrophage/dendritic cell

(DC) functions, respectively. LysM+ and Iaβ+ mice on a TLR2-/- background were used as negative controls, since TLR2 is essential for recognition of Bb surface lipoproteins. IL-

iv 10-/- LysM+ mice display significantly higher neutrophil infiltration and persistence at the site of infection up to day 7 post-infection, however these numbers reach basal levels by day 14 post-infection, indicative of additional IL-10-independent factors contributing to this suppression of neutrophil recruitment. TLR2-/- LysM+ mice had significantly diminished neutrophil infiltration. The resident immune cells in IL-10-/- Iaβ+ mice display significantly larger cell diameter and area within 6 hours post-infection whereas WT Iaβ+ cells take up to 6 days to attain similar levels. Increase in cell diameter and area was completely absent in TLR2-/- macrophages/dendritic cells, suggesting TLR2-/- immune cells are deficient in responding to Bb infection. IL-10-/- macrophages/DCs also achieve higher velocity and confinement ratios faster than WT cells, whereas TLR2-/- cells retained baseline characteristics. Hence, Bb-elicited IL-10 diminishes activation of resident immune cells, whereas loss of TLR2 almost completely abolishes innate immune cell activation in response to Bb infection.

Lastly, by using the IL-10-/- mouse as a hyperactive and TLR2-/- mouse as a hypoactive immune response model, we investigated the role of Bb velocity in causing persistent infection. Previous studies from our lab identified that Bb can achieve average velocities of ≥200 μm/min in vivo, which is ≥40x faster than any observed immune cell within the skin. Using a number Bb chemotactic and motility mutants, Bb persistence analyses in IL-10-/- mice and TLR2-/- mice revealed that even reduction of Bb 50%, which is still 20-fold faster than all immune cells in the skin, significantly hampers Bb persistence in vivo, perhaps at levels that are unable to maintain the enzootic cycle.

To summarize, these findings suggest that 1) presence of Bb in skin tissue elicits a rapid and potent IL-10 response from skin resident immune cells within 6 hours post-

v infection, 2) Bb-elicited IL-10 suppresses host immune activation which allows for enhanced Bb persistence, 3) there are additional mechanisms independent of IL-10 that further suppresses innate immune responses, and 4) it is essential for Bb to maintain this high velocity to achieve maximal long-term persistence. Overall this study provides a snapshot of immune dysregulation occurring during Lyme disease at the site of infection and identifies the factors which allow Bb to evade the host immune response.

Dedicated to the beautiful memory of my late grandmother, Mrs. Shamsunnia Venjara, and my grandmother, Ms. Zainab Moledina, who always remembered me in their unfeigned prayers. Their unfettered affection, love, and prayers provided me with comfort and security, empowering me to succeed in life. I dedicate my effort and hard work to them both, for helping me reach this milestone.

vii

Acknowledgements

I would like to start by thanking my advisor, Dr. R Mark Wooten for his faith, patience and guidance which was a cornerstone for my success. His mentorship has enabled me to expand my skillset and scientific acumen. His efforts and dedication have allowed me to improve my presentation skills and penmanship. I would like to express my gratitude to John Presloid, a colleague, friend, and guide, who always provided me with support and advice throughout my project.

I am grateful to all my committee members, Drs Andrea Kalinoski, Kevin Pan,

Randall Worth and Robert Blumenthal for their time, effort, comments and encouragement.

They have been an integral part of this project and their reviews helped me steer this project towards success.

I would like to extend my gratitude to my past and present lab mates, Irum Syed,

Caroline Lambert, Erin Sheehan, Dr. Laura Nejedlik, Dr. Padmapriya Sekar, all members of the MMI track and all my friends for their support and friendship. They are my family away from home.

Lastly, I have to acknowledge my parents, who’s unwavering faith, love, support and prayers played a vital contribution in my success. This achievement is equally theirs, as it is mine. My sisters for their wisdom and my wife for her love and patience through this challenging, yet exciting period of life.

viii

Table of Contents

Abstract ...... iii

Acknowledgements ...... viii

Table of Contents ...... ix

List of Tables ...... xiii

List of Figures ...... xiv

List of Abbreviations ...... xvi

1 Introduction and Literature Review ...... 1

1.1 Borrelia burgdorferi (Bb): The zoonotic spirochete ...... 1

1.1.1 Ticks, reservoirs and the enzootic lifecycle ...... 2

1.1.2 Bb genome: Extravagantly redundant yet metabolically frugal ...... 4

1.1.3 Bb motility and chemotaxis: Periplasmic nano-machinery ...... 6

1.1.3.1 Bb motility ...... 7

1.1.3.2 Bb chemotaxis ...... 10

1.1.3.3 Motility and chemotactic mutants ...... 12

1.2 Lyme borreliosis: An infection or an inflammatory disease? ...... 14

1.2.1 History...... 14

1.2.2 Prevalence and distribution ...... 15

1.2.3 Lyme Pathology ...... 18

1.2.3.1 Stages of Lyme disease ...... 19

1.2.3.2 Antibiotic-refractory Lyme disease ...... 21

1.2.3.3 Post-treatment Lyme disease syndrome ...... 22

1.2.4 Immune response in Lyme disease ...... 23

1.2.4.1 Innate immunity: The 1st responders ...... 27

1.2.4.1.1 Phagocytes and associated surface receptors ...... 27

1.2.4.1.2 Bb-elicited IL-10 response ...... 29

1.2.4.2 Adaptive immunity ...... 32

1.2.4.2.1 Role of B cells ...... 33

1.2.4.2.2 Role of T cells ...... 34

1.3 Intravital Microscopy: The live picture from within ...... 35

1.3.1 Intravital Microscopy to study Lyme disease ...... 36

1.3.1.1 Preliminary assessment of Bb in ear tissue ...... 38

1.3.1.1.1 Bb dissemination and persistence ...... 39

1.3.1.1.2 Bb numbers and Bb-specific antibodies ...... 41

1.3.1.1.3 Early neutrophil response to Bb infection ...... 42

1.4 Goals for dissertation ...... 43

2 Material and Methods ...... 46

2.1 Bacterial strains and growth conditions ...... 46

2.2 Mice ...... 47

2.3 Intravital microscopy ...... 50

2.4 Quantitative measurement of B. burgdorferi in murine ear skin ...... 54

2.5 Detection of B. burgdorferi-specific antibody by ELISA ...... 55 x

3 Role of adaptive immunity in controlling Bb persistence at the site of infection..57

3.1 Bb persistence in T cell-deficient mouse models ...... 59

3.1.1 Bb persistence in Nude mice...... 60

3.1.2 Bb persistence in TCR-/- mice ...... 62

3.2 Bb persistence in B cell-deficient mouse models ...... 64

3.2.1 Bb persistence in µMT mice ...... 64

3.2.2 Bb persistence in AID-/- mice ...... 66

4 Role of Bb-elicited IL-10 in suppression of innate immunity ...... 69

4.1 Visualization of Bb-elicited IL-10 using the tiger mouse model ...... 72

4.1.1 Identification of cell types associated with IL-10 production ...... 73

4.1.2 Kinetics associated with IL-10 production ...... 75

4.2 WT-GFP Bb persistence in WT, TLR2-/- and IL-10-/- mice ...... 77

4.3 Neutrophil response to Bb infection in WT, TLR2-/- and IL-10-/- mice .....81

4.4 Resident immune cell activity in WT, TLR2-/- and IL-10-/- mice ...... 84

4.4.1 Iaβ+ cell numbers in response to Bb infection ...... 86

4.4.2 Iaβ+ cell size in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice ...... 87

4.4.3 Dynamic behavior of Iaβ+ cells ...... 90

5 Role of Bb velocity in evasion of innate immune response...... 94

5.1 ΔcheY3-GFP Bb persistence in WT, TLR2-/- and IL-10-/- mice ...... 98

5.2 ΔflaA-GFP Bb persistence in WT, TLR2-/- and IL-10-/- mice ...... 100

6 Discussion ...... 103

6.1 Overview ...... 103

6.2 Role of Bb-elicited IL-10 on innate cell function ...... 105 xi

6.3 Role of adaptive immunity in Bb clearance ...... 111

6.4 Role of Bb motility in evading host immunity ...... 115

6.5 Concluding Remarks ...... 119

References ...... 121

A Permission to use Figure (I) in dissertation ...... 136

B Permission to use Figure (II) in dissertation ...... 137

C IVM images of Bb numbers in WT, IL-10 KO and TLR2 KO mice...... 138

D IVM images of LysM+ cells in WT, IL-10 KO and TLR2 KO mice ...... 139

E IVM images of Iaβ+ cells in WT, IL-10 KO and TLR2 KO mice ...... 140

xii

List of Tables

2.1 List of mice used in this study ...... 49

2.2 List of primers used in the study ...... 55

xiii

List of Figures

1 – 1 Enzootic cycle of Borrelia burgdorferi...... 3

1 – 2 Morphology and motor-flagella apparatus in B. burgdorferi ...... 9

1 – 3 Spread of Lyme disease in United States from 2001 - 2017 ...... 17

1 – 4 Overview of host immune response to B. burgdorferi infection ...... 19

1 – 5 B. burgdorferi persistence at site of infection in mice ...... 38

1 – 6 Correlation between Bb-specific antibody and bacterial numbers ...... 40

1 – 7 Early neutrophil response to B. burgdorferi at the injection site ...... 42

2 – 1 Ear template used for mouse infection...... 52

3 – 1 Bb persistence and antibody production in WT and nude mice ...... 61

3 – 2 Bb persistence and antibody production in WT and TCR-/- mice ...... 63

3 – 3 Bb persistence and antibody production in WT and µMT mice ...... 65

3 – 4 Bb persistence and antibody production in WT and AID-/- mice ...... 67

4 – 1 Cell types associated with IL-10 production in response to Bb infection ...... 74

4 – 2 Kinetics associated with IL-10 production in response to Bb infection ...... 76

4 – 3 WT Bb persistence in WT, IL-10-/- and TLR2-/- mice ...... 80

4 – 4 Bb-specific antibody production in WT, IL-10-/- and TLR2-/- mice ...... 81

4 – 5 LysM+ cell infiltration in WT, IL-10-/- and TLR2-/- mice ...... 84

4 – 6 Iaβ+ cell numbers in WT, IL-10-/- and TLR2-/- mice ...... 85

xiv

4 – 7 Iaβ+ cell size in during Bb infection ...... 89

4 – 8 Dynamic behavior of Iaβ+ cell mice during Bb infection ...... 92

5 – 1 Comparison of WT, ΔflaA, ΔcheY3 Bb velocity ...... 96

5 – 2 WT and ΔcheY3 Bb persistence in WT, IL-10-/- and TLR2-/- mice ...... 100

5 – 3 WT and ΔflaA Bb persistence in WT, IL-10-/- and TLR2-/- mice ...... 101

List of Abbreviations

2D ...... 2-Dimensional 3D ...... 3-Dimensional

Ab ...... Antibody AID ...... Activation – Induced Cytidine Deaminase AKT ...... (protein) Kinase B APC ...... Antigen Presenting Cells

Bb ...... Borrelia burgdorferi BL/6 ...... Black 6 BSK II ...... Barbour – Stoenner – Kelly II (media)

CCL ...... Chemokine (C – C motif) Ligand CCW ...... Counter-Clock(W)ise CD ...... Cluster (of) Differentiation CD40L...... CD40 Ligand CDC ...... Center (for) Disease Control cp ...... Circular Plasmid CSF ...... Cerebro Spinal Fluid CW ...... Clock(W)ise CXCL ...... Chemokine (C – X – C motif) Ligand

DBP ...... Decorin Binding Protein DC ...... Dendritic Cell

ECM ...... Extra-Cellular Matrix ELISA ...... Enzyme Linked Immunosorbent Assay EM...... Erythema Migrans

FBS ...... Fetal Bovine Serum

GFP ...... Green Fluorescence Protein

HRP ...... Horse Raddish Peroxidase

xvi

i.d ...... Intradermal i.p ...... Intraperitoneal IBD ...... Inflammatory Bowel Disease

IFN ...... Interferon Ig ...... Immunoglobulin IL ...... Interleukin IVM ...... Intra Vital Microscopy

Kb ...... Kilo Base KDa ...... Kilo Dalton

LD ...... Lyme Disease lp ...... Linear Plasmid LysM ...... Lysozyme M

MAP ...... Mitogen-Activated Protein (kinase) Mb ...... Mega Base MCP ...... Methyl accepting Chemotactic Protein MIP ...... Macrophage Inflammatory Protein MMP9 ...... Matrix Metallo Peptidase 9 MZB ...... Marginal Zone B cells

NAD ...... Nicotinamide Adenine Dinucleotide NF-κB ...... Nuclear Factor kappa-light-chain (enhancer of activated) B cells NO ...... Nitric Oxide

OSP ...... Outer Surface Protein

Pam3 ...... N-palmitoyl-S-dipalmitoylglyceryl PAMP ...... Pathogen Associated Molecular Pattern PBMC ...... Peripheral Blood Mononuclear Cell PBS ...... Phosphate Buffered Saline PCR ...... Polymerase Chain Reaction PI3K ...... Phosphatidyl Inositol 3 kinase PRR ...... Pathogen Recognition Receptor PTLDS ...... Post-Treatment Lyme Disease Syndrome qPCR ...... Quantitative (real time) PCR

RA ...... Rheumatoid Arthritis RAG ...... Recombination Activating Genes RANTES ...... Regulated (on) Activation Normal T cell Expressed (and) Secreted ROS ...... Reactive Oxygen Species

xvii

SCID ...... Severe Combined Immuno-Deficiency STM ...... Signature-Tagged Mutagenesis

TCR ...... T Cell Receptor TLR ...... Toll Like Receptor TNF ...... Tumor Necrosis Factor Treg ...... Regulatory T Cell

WT ...... Wild-Type

xviii

Chapter 1

Introduction and Literature Review

1.1 Borrelia burgdorferi (Bb): The zoonotic spirochete

Spirochetes comprise of a group of long cylindrical bacteria which are approximately a micron in diameter, but can range from 5 to 250 microns in length.[1]

Though a rather small group of bacteria with around 6 genera, these spirochetes have a tremendous impact on human lives, since most of them are pathogenic and cause diseases such as syphilis, yaws, leptospirosis, relapsing fever and Lyme disease.

Borrelia burgdorferi (Bb) belongs to this group of spirochetal pathogens and is the causative agent of Lyme disease.[2] Highly prevalent in the Northern hemisphere, the Lyme disease-causing Borrelia species primarily include Borrelia burgdorferi, Borrelia garinii and Borrelia afzelii. They are transmitted by 4 species of Ixodes tick vectors: Ixodes scapularis, Ixodes pacificus, Ixodes ricinus and Ixodes persulcatus.[3, 4] In the United

States, Borrelia burgdorferi is the only Lyme-causing pathogen to infect humans and is transmitted by Ixodes scapularis ticks in the Midwest regions, whereas in the west coast

Ixodes pacificus is the primary vector.[3, 5-7] The pathogen was initially isolated from the midgut of an Ixodes scapularis tick by Dr. Willy Burgdorfer [2] in 1982 and subsequently

named Borrelia burgdorferi. This bacterium is an obligate extracellular parasite, and in spite of eliciting strong immune responses, can cause a persistent infection which disseminates throughout the host and produce a wide variety of pathologies, collectively known as Lyme disease (discussed in section 1.2). Even after 4 decades of research and advancement, the exact mechanisms employed by Bb to evade all host immune defenses still remains unclear. In this section, we will discuss certain aspects of Bb which makes this pathogen unique.

1.1.1 Ticks, reservoirs and the enzootic lifecycle

While mice, squirrels, small rodents, a variety of small mammals and even several avian species constitute a natural reservoir for Bb, the white-footed mouse, Peromyscus leucopus, is considered to be the main reservoir in north-eastern United States.[8] There is no transovarial transmission of Bb from infected adult ticks, so the larvae hatch uninfected

(Figure 1 – 1). In summer, larvae acquire their 1st blood meal which allows them to molt into nymphs. If the blood meal was acquired from a reservoir animal infected with Bb, the larvae is highly likely to get colonized with the Lyme disease pathogen. These larvae harbor the spirochete in a dormant state within the midgut of the arthropod as they molt into nymphs the following spring. These infected nymphs can now transmit Bb to uninfected vertebrate hosts as they take their 2nd blood meal. This process allows the maintenance and sustenance of Bb reservoirs in the wild. The nymph molt into adult ticks the same year in fall, after their 2nd blood meal. The adult ticks feed on much larger mammals, such as deer, in search of a larger and final blood meal which would prepare

Figure 1 – 1: Enzootic cycle of Borrelia burgdorferi. Ixodes spp. ticks acquire one blood meal per stage of their three-stage life cycle — larva, nymph and adult. Larval ticks feed on small rodents. Bb infection is acquired by feeding on an infected reservoir animal. This infection is retained for entire lifespan of the tick. Nymphs feed on a similar range of hosts to larvae; transmitting the spirochete to a competent reservoir host by perpetuating the enzootic cycle for the next generation of larval ticks. Adult ticks are not generally important for maintenance of Bb in the wild, as they feed predominantly on larger animals such as deer, which are incompetent hosts for the spirochete. However, deer are essential for maintenance of the tick population because adult ticks mate on them. While all three stages of Ixodes scapularis can feed on humans, nymphs are responsible for the vast majority of spirochete transmission to humans. Humans are generally considered dead-end hosts and not part of the enzootic cycle. Hence humans are not essential for maintenance of Bb lifecycle and infection to humans is a mere accidental exposure.[3] Reproduced with permission from Springer Nature: Nat Rev Microbiol 10(2): 87–99. doi:10.1038/nrmicro2714

them to lay eggs, hence completing the enzootic lifecycle of Bb. Humans are not essential for Bb to maintain its life cycle and are considered as ‘dead-end hosts’ in the Bb lifecycle because there is no further propagation/transmission of the pathogen to any subsequently feeding tick.[3, 4, 9, 10] The inadvertent exposure of Bb to humans results in an inflammatory immune disease known as Lyme disease.

1.1.2 Bb genome: Extravagantly redundant yet metabolically frugal

Spirochetes within the Borrelia genus display profound complexity of their genomic material,[11] and Bb is no exception. The linear chromosome of Bb comprises of

~1 Mb of genetic material which contains most of the housekeeping genes and its constitution is relatively constant across all species within the Borrelia genus. However,

Bb also contains 21 extra-chromosomal plasmids, larger than any other bacterium. The 12 linear plasmids (lp) and 9 circular plasmids (cp) contribute an additional ~600 Kb DNA material,[12, 13] and over 90% of genes present on these plasmids are unique to Bb, suggesting that they are involved in specialized functions. These specialized functions have been implicated in Bb being able to support and maintain an elaborate and complex enzootic lifecycle, which involves constant shuttling being and persistence within vertebrate and arthropod hosts.

Most extrachromosomal plasmids are disposable for in vitro culture, yet essential for maintaining the enzootic life cycle.[14-16] lp25 consists of a gene pncA which encodes for nicotinamidase which is essential for survival of Bb within mammalian host.[17]

Nicotinamidase is required for NAD synthesis and introduction of DNA fragment containing pncA gene back to lp25-deficient Bb is sufficient to restore infectivity in mice. 4

Bb possesses both inner and outer membranes of a cell wall (i.e. diderm), similar to most gram-negative bacterium. However, it is completely devoid of lipopolysaccharides in the outer leaflet. Instead, it expresses more than 100 putative surface lipoproteins which are almost entirely encoded by extrachromosomal plasmids. [11] Many of these lipoproteins are also essential for Bb to maintain its enzootic life cycle. For example, ospA and ospB encoded by lp54 are genes which encode for outer surface proteins (OspA and OspB) that serve as adhesins essential for Bb colonization of tick midgut. Disruption of the ospA/B operon is accompanied by a defect in colonization of ticks by Bb.[18] lp54 also encodes for adhesins required for mammalian infection and persistence. The dbpA/B operon encodes for decorin-binding proteins (DbpA and DbpB) which facilitate attachment of the spirochete to collagen and eukaryotic cells, and are implicated in dissemination of Bb from the initial site of infection during the early colonization stage.[19-21] lp28 contains a 8Kb region (vlsE region) of adjoining silent cassettes involved in antigenic variation. This region repeatedly alters the outer membrane proteins exposed on the surface and plays a major role in evasion of mammalian adaptive immune response.[22, 23] These extrachromosomal plasmids also possess several genes which are essential for survival in ticks, including glpF, glpK, glpD guaB and guaA.[24]

From the above examples, it is evident that the plasmids which account for almost

40% of Bb genome [25], provide Bb with very specific features allowing it to adapt and persist as it shuttles between different hosts that possess extremely varied environments.

However, a hallmark of Bb genome is the extreme redundancy it exhibits. The redundancy is reflected not only by the presence of almost 160 paralogous gene families within the plasmids, but also the large stretches of almost identical DNA interspersed by sequence- 5

variable lipoprotein genes. Yet, the metabolic capability of Bb is extremely restricted.

Notably, it is an auxotroph for amino acids, nucleotides and fatty acids. It even lacks genes encoding enzymes for the tricarboxylic acid cycle and oxidative phosphorylation. Hence, the pathogen requires a highly complex, poorly defined, and nutrient-rich media to grow in vitro, and is an obligate parasite in vivo since it is completely dependent on its hosts for nutrients. It has no metabolic requirement of iron and the pathogen does not encode any known classical virulence factors, like toxins and secretion systems.[11] This supports the dogma that Lyme disease symptoms are inflicted largely due to the host immune response generated against the pathogen rather than the pathogen itself. Interestingly, the pathogen chooses to invest 6% of its genome on its motility and chemotaxis apparatus. Several factors in their enzootic cycle justify the need to invest a vast amount of resources in motility and chemotaxis. We will discuss these, in detail, in the next segment.

1.1.3 Bb motility and chemotaxis: Periplasmic nano-machinery

Bb has almost 50 putative genes, approximately 6% of its genome, encoding proteins associated with motility and chemotaxis based on homology to other bacteria.[11]

The proteins are involved in construction of the motor and stator ring, the periplasmic flagella and the chemotactic sensing element (methyl-accepting chemotactic proteins;

MCP) along with the chemotaxis two-component system. Several facts explain the need for Bb to invest vast amounts of resources on its motility and chemotaxis apparatus. For one, the spirochete utilizes its sophisticated but powerful motor machinery for dissemination within dense vertebrate host tissues. However, it also has to translocate to

and from particular tissues within both an arthropod vector and vertebrate hosts to complete its enzootic lifecycle.

Both chemotaxis and motility are essential for Bb to sense and adapt to the environmental changes while it migrates between mammalian and arthropod hosts.

Needless to say, deletion of key motility and chemotaxis genes render the pathogen non- infections.[26-30] A global signature-tagged mutagenesis (STM) study identified 10 of the

14 chemotaxis genes and 4 of the 7 genes involved in flagella assembly as essential for infectivity.[26] The machinery also allows this pathogen to traverse at velocities of ≥ 200

μm/min in dense murine tissues, approximately 40-times faster than the fastest- documented immune cells.[31, 32] (Lavik et al, unpublished data). This supports the belief that Bb utilizes its motility as an immune evasion strategy by effectively outrunning the cellular immune response.

1.1.3.1 Bb motility

Bb possess 7-11 periplasmic flagella (i.e. endoflagella) which are sub-terminally attached to either ends of the spirochete (Figure 1 – 2).[33] Each flagellum consists of a filament, hook and motor. While the individual motors are 90-120 nm apart and are inserted within the inner membrane along the long axis of the cell on either ends, the filaments are wrapped around the inner membrane within the periplasmic space, forming a ribbon-like pattern such that filaments originating from either end of the poles overlap in the center.[34,

35] In this section, we will discuss the structural properties of each of these components and importance of various proteins involved in their assembly.

Filament: The flagellar filament is comprised of two proteins: a 41KDa FlaB protein which forms the majority of the filament and a 38KDa minor FlaA protein which forms the flagellar sheath. Both FlaA and FlaB share sequence identity with other spirochetes.[36]

FlaB forms the core of the filament and is essential for Bb persistence and infectivity. ΔflaB mutants of Bb are completely non-motile and lose their flat wave-like morphology; appear rod-like in vitro. Also, ΔflaB Bb are unable to establish and infection in both ticks and mice.[28, 36] While the functional role of FlaA is still undetermined, FlaA is known to form a sheath around the FlaB core in Bb and several other spirochetes, and is believed to provide additional rigidity to the structure. Recent assessment utilizing ΔflaA mutants of

Brachyspira hyodysenteriae, which have unsheathed periplasmic flagella, demonstrated that these spirochetes are still motile, but move slower than wild-type.[37] In our study, similar observations were made while comparing wild-type Bb and ΔflaA Bb velocity; this will be discussed later in depth. ΔflaA mutants of Brachyspira hyodysenteriae also displayed lesser rigidity than wild-type. Since the Bb FlaA sheath is localized adjacent to the flagellar hook[38], the data suggests flagellar rigidity provided by FlaA around the hook could be required for optimal flagellar rotation and motility.

Hook: In most bacteria, the hook acts as a universal connector between the flagellar filament and motor.[39] In Bb, it is a 61 nm long hollow tube made up of approximately 120 units of FlgE protein units.[40] While most bacteria have hydrophobic interactions holding the FlgE complex together, making it readily dissociable by denaturing agents, the FlgE complex in Bb appears to be cross-linked by covalent interactions, forming a very stable, high molecular weight complex.[40, 41] This evidence supports the line of thought that spirochetal filament and hook are under greater amount of stress, requiring a stronger 8

complex to withstand the force. Similar to ΔflaB, ΔflgE mutants also lose their morphology and motility in vitro.[40]

Figure 1 – 2: Morphology and motor-flagella apparatus in B. burgdorferi. (a) Schematic model of Bb displaying general morphology and periplasmic flagella located between the outer membrane (OM) and the inner membrane (IM). (b) Components of the flagellar motor. PG, peptidoglycan layer; EXP, export apparatus. Both panels reproduced with permission from Elsevier: Current Opinion in Microbiology 28 (2015): 106-113. 10.1016/j.mib.2015.09.006

The Motor: Bb motor is a remarkable and complex nano-machine (Figure 1-2). It is comprised of more than 20 different proteins, together which make up the individual components of the motor: the MS-ring, the C-ring, the rods, the export apparatus and the stator.

• The MS ring consists of the M-ring embedded in the inner membrane and the S- ring in the periplasmic space. The two rings are made up of FliF proteins. The export 9

apparatus is situated beneath the MS-ring and comprises of 9 proteins: 6 transmembrane

(FlhA, FlhB, FliO, FliP, FliQ and FliR) and three soluble (FliH, FliI and FliJ) proteins.

Soluble proteins FliH and FliI have been identified as essential for Bb infectivity.[26]

• The C-ring is present in the cytoplasmic compartment and consists of bulky top with a Y-shaped bottom. The C-ring plays an important role in export, torque generation, and directional switching. It is 54-57 nm in diameter and comprises of 3 proteins: FliG,

FliM and FliN[42, 43]. FliG is the only motor protein to have 2 genes (fliG1 and fliG2). Loss of FliG2 resulted in non-motile, rod-shaped spirochetes. While FliG1 mutants retain their morphology, they were non-motile in viscous media resulting in loss of infectivity. [26, 27]

• The 22 nm rod is made up of five different proteins: FlgG, FliF, FlgB, FlgC and

FlgF. It acts as a driveshaft and its function is to transfer the torque generated by the motor to the hook and the filament.[42]

• Surrounding the MS-ring are the stator complexes. Two proteins, MotA and MotB make up this complex. While MotA is embedded in the inner membrane, only the N- terminal of MotB is in the cytoplasmic membrane whereas the C-terminal is in the periplasmic space. The stator complex is the component involved in generation of torque required for motility. They do so by transforming the flow of H+ ions into energy which is then transferred to the hook/filament complex with assistance from the C-ring.[43] MotB is essential for Bb infectivity as ΔmotB mutants are non-motile both in vitro and in vivo, and get cleared from the murine system within 24-36 hrs.[29]

1.1.3.2 Bb chemotaxis

Chemotaxis is defined as a movement towards or away from a chemical stimulus.

Chemotaxis proteins are involved in sensing of chemo-attractant or chemo-repellent 10

molecules resulting in a biased movement of organism either towards or away from chemical stimulus. It does so by modulating both, the directionality of motor rotation as well as the speed of motor rotation. The general paradigm of a bacterial chemotaxis system, as seen in E. coli, consists of a 2-component system: a histidine kinase CheA and a response regulator CheY.[44, 45] CheA is associated with the polarly located sensor element: methyl- accepting chemotaxis protein (MCP) receptor. A change in directionality of flagellar rotation is initiated when the MCP binds to a ligand. CheA is associated to the MCP by an adaptor molecule CheW. Activation of the MCP results in activation of CheA via an MCP- specific methyl transferase: CheR. An activated CheA undergoes auto-phosphorylation and goes on to transfer the phosphate group to CheY. CheY-P binds to the FliM protein of the flagellar motor resulting in the motor switching its direction of rotation. This prevents the bacteria from translocating (e.g. performing a directed run) and initiates a tumbling process where the bacteria reorient. Removal of the ligand from MCP results in dephosphorylation of CheY-P by a phosphatase (CheZ), detaching CheY from the FliM protein and the motor regains its original directionality of rotation. Now the bacteria can again translocate

(perform a run), however it moves in a different direction due to the reorientation introduced during the tumbling process. This is how a bacterium move towards or against a gradient of attractants or repellents by using a series of runs and tumbles, all orchestrated by the chemotaxis system.[44, 46]

As most systems, the chemotaxis system in spirochetes is more complex than most motile bacteria. The flagellar motors and MCP are present on either pole of the spirochete and motors at each end need to rotate in opposite directions for the spirochete to move forward. The steady state direction of flagellar motor rotation in E. coli is counter 11

clockwise (CCW) while in B. subtilis it is clockwise (CW). Since Bb possesses motors on both ends, one end needs to rotate CW whereas the other one rotates CCW for Bb to translocate (perform a run).[47] It is not well understood how the motors and receptors on either ends of Bb communicate with each other to synchronize their direction of motor rotation and coordinate a series of tumbles and runs. Diffusion of CheY-P from across the long bacterial cell would be greatly time consuming, thus making it highly unlikely. More studies need to be performed to understand the exact mechanism of chemotaxis.

While the signaling mechanism in Bb is similar to E. coli, Bb contains the phosphatase CheX instead of CheZ[48] and it also contains an additional protein CheD, which can bind both the chemoreceptor and the phosphatase, regulating their activity.[45]

The complexity of the chemotaxis system in Bb is enhanced even further due to the presence of multiple homologs of key chemotactic genes. This includes: 5 mcp (2 lack membrane spanning regions), 2 cheA, 3 cheY, 2 cheB, 2 cheR, and 3 cheW. [49] Research from our lab along others have provided evidence that these genes perform varied function and affect spirochetal motility, dissemination and persistence based on the environment in which Bb resides. [50, 51] (unpublished data, Moledina et al)

1.1.3.3 Motility and chemotactic mutants: Genetic tools for Lyme research

From this section, it is evident that both motility and chemotaxis are essential systems for Bb infectivity. Previous work from our lab have determined that Bb traverses within dense host tissue at a rate 40-times faster than neutrophils (e.g. fastest immune cells in humans) (unpublished data; Lavik et al). Construction of single deletion mutants of Bb

which lack specific proteins associated with the motility and chemotaxis apparatus has allowed us to determine the function of these proteins in Bb infectivity.

ΔmotB Bb is devoid of the stator protein and are completely non-motile. These spirochetes cannot persist in vivo and get cleared within 24 hours.[29] ΔcheY3 Bb lack the chemotaxis response regulator CheY3. Loss of CheY3 does not affect Bb velocity and these spirochetes retain normal velocities.[51] However, loss of CheY3 prevents Bb from reversing its directionality, preventing it from performing its characteristic back-and-forth motion. Thus, while WT-Bb can use the back-and-forth motion to burrow through dense host tissue, this mutant spirochete eventually becomes sessile. Predictably, ΔcheY3 Bb gets cleared from the host within 3 to 4 days post-infection. ΔflaA Bb are devoid of the flagellar sheath protein FlaA. Loss of FlaA also allows Bb to retain its motility, however these spirochetes move much slower than WT-Bb (around half the speed of Bb) (Unpublished data, RM Wooten and Md Motaleb lab).

Since one of the aims of this dissertation study is to assess the importance of Bb velocity to evade host immune clearance, these mutants serve as important tools to perform

푛표푟푚푎푙 푠푝푒푒푑 this study. By using WT-Bb (normal speed), ΔflaA Bb ( ), ΔcheY3 (lose motile 2 in vivo) we have 3 strains of Bb that traverse at 3 different velocities, thus providing a perfect tool to tease apart the importance and relevance of not only Bb motility but also the significance of why Bb has to maintain such high speeds while in a mammalian host.

1.2 Lyme borreliosis: An infection or an inflammatory disease?

1.2.1 History

The etiology of LD was first noted in 1970, when an abnormally high number of children were being diagnosed with juvenile rheumatoid arthritis in Old Lyme,

Connecticut. On investigation, physicians confirmed an elevated level of arthritis incidence in this region as compared to rest of USA and also identified adults experiencing similar symptoms, yet the sequellae were different from juvenile rheumatoid arthritis. Since, the arthritis events were accompanied with neurological and cardiological deficits, the disorder was named as Lyme disease. [52] Further examination of cases revealed that several patients developed an erythematous cutaneous lesion days before arthritis onset. These lesions were characterized to be similar to lesions associated with the bite of Ixodes ricinus, called

‘erythema migrans’.[53]

Around the same time frame, an increase in I. scapularis tick population on the eastern side of the Connecticut River was associated with higher LD cases compared to the west side.[54-56] Eventually, Dr. Willy Burgdorfer at the Rocky Mountain Lab was successful in isolating a spirochete from these I. scapularis ticks in 1982 and confirmed this spirochete to be the causative agent of LD.[2] Peromyscus leucopus mice from that area also tested positive for this spirochete and were suspected to be a natural reservoir for the organism.[57] In 1983, the spirochete was named Borrelia burgdorferi, after Dr. Willy

Burgdorfer at the 1st international Lyme disease meeting.

I. ricinus was identified as the vector for transmitting similar spirochetes in Europe.

These spirochetes were named as B. afezelli and B. garnii, and were attributed to causing

LD-like symptoms, as well as acrodermatitis chronica atrophicans skin etiology that 14

appears unique to the European strains. Since then, the original Bb isolate from the USA was termed B. burgdorferi sensu stricto, whereas the European isolates were collectively termed as B. burgdorferi sensu lato.[8]

1.2.2 Prevalence and distribution

LD became a nationally reportable disease in 1991 and the number of reported cases has been rising ever since.[58] A total of 42,743 confirmed and probable cases of Lyme disease were reported to the CDC in 2017, a 17% increase since 2016. However, studies conducted by the CDC report this number to be grossly underestimated and in reality almost 0.3 million people currently contract the disease each year.[59] The increase in the number of reported cases can be attributed to several factors: 1) enhanced surveillance by physicians, 2) heightened awareness about the disease amongst the public and 3) increase in the rate of infection associated with the spread of the tick population across geographical regions.[60]

The increase in reported cases has been accompanied with an increase in number of counties with an incidence of ≥10 confirmed cases per 100,000 persons; from 324 in

2008 to 454 in 2017. Being a zoonotic disease, LD is highly prevalent in regions where appropriate ticks and reservoir species are present. In the United States, 14 states are considered to be endemic for LD and account for 96% of the cases. These states include

Connecticut, Delaware, Maine, Maryland, Massachusetts, Minnesota, New Hampshire,

New Jersey, New York, Pennsylvania, Rhode Island, Vermont, Virginia and Wisconsin.[59]

I. scapularis (deer tick or eastern blacklegged tick) is the primary vector for Bb in the

Northeast and Midwestern states. The emergence and spread of I. scapularis to non- 15

endemic regions in recent times[61, 62] has resulted in 25-50% increase in LD cases in these regions.[63] Hence, a drastic change has been observed over the years, where this regionally- endemic disease has spread rapidly to surrounding areas, increasing the number of LD cases (Figure 1 – 3 ).

(a) Reported cases of Lyme disease – United States, 2001

(b) Reported cases of Lyme disease – United States, 2017

Figure 1 – 3: Spread of Lyme disease in United States from 2001 – 2017; CDC

1.2.3 Lyme Pathology

Bb gains access to the dermal layer of the skin during tick feeding.[3, 8, 64] The tick has to feed on the host for at least 36 to 48 hours for Bb to successfully get transmitted from the infected tick to the naïve mammalian host. Within the skin, resident immune cells, mainly macrophages (MØs) and dendritic cells (DCs), are continuously surveilling the region for foreign materials. In vitro experiments from our lab as well as other groups have confirmed that these resident professional antigen-presenting cells (APCs) are capable of recognizing Bb surface lipoproteins via Toll-like receptor 2 (TLR2) present on the immune cell surface, leading to immune cell activation.[65, 66] . In vitro, activated APCs respond by

1) directly killing the internalized Bb by producing nitric oxide (NO) and reactive oxygen species (ROS) , 2) secrete cytokines which should activate neighboring APCs to assist in

Bb clearance at the infection site, and 3) secrete chemokines which should recruit additional immune cells from the periphery to the site of infection [67-69].

These activated APCs also migrate to the germinal center of lymph nodes to prime an adaptive immune response against Bb. Although delayed, a Bb-specific T cells and antibody (activated B cells differentiate into plasma cells to produce Bb-specific antibodies) is generated against Bb.[70-72] To summarize, both the innate and adaptive immune components are capable of recognizing and responding to a Bb infection (Figure

1 – 4). However, these responses fail to clear the spirochete during natural infection, and in the absence of a therapeutic intervention, Bb proceeds to disseminate throughout the body. The stage of dissemination and persistence correlates with the symptoms and severity of LD. In humans, LD can be roughly categorized into three different stages.[59, 60, 64]

Figure 1 – 4: Overview of host immune response to B. burgdorferi infection

1.2.3.1 Stages of Lyme disease

1. Early-localized stage: This is the 1st stage of LD and can be observe anywhere from

3 to 30 days post-transmission. Around 75% of LD patients at this stage develop erythema migrans (EM), a characteristic expanding skin rash at the site of tick bite. In certain cases, the erythematous center is surrounded by an erythematous ring with a zone of clearance dividing the two rashes, giving it a ‘bull’s-eye like’ appearance. The rash is precipitated by the activation of immune response to Bb. As mentioned previously, activation of the resident immune cells results in pro-inflammatory cytokine secretion by MØs and DCs.

Not only does this activate surrounding innate immune cells, but also recruits T cells and plasma cells to the site of infection. The inflammatory cascade results in hyperemia in the capillaries, leading to this characteristic rash.[73] Histopathological investigations have identified neutrophils, eosinophils, T cells and occasionally abundant plasma cells as primary infiltrates associated with the rash.[74] Along with the rash, the disease also presents itself with flu-like symptoms which include: fever, fatigue, headache, lymphadenopathy and myalgia. If correctly diagnosed at this stage, the disease is extremely treatable with antibiotics with good prognosis of complete recovery.

2. Early-disseminated stage: If left untreated, the disease progresses to a more disseminated stage. This can take days to months after the initial infection. At this stage,

Bb have migrated from the initial site of infection to multiple distant target tissues. Multiple

EM lesions can occur which are believed to be a result of host immune cells “chasing” disseminating Bb.[75] Patients can develop cardiac and neurological complications at this stage which include Bell’s palsy, heart palpitations and dizziness due to changes in the normal electrical conduction within heart tissues, and neck stiffness due to meningitis. It is noteworthy, patients experiencing neuroborreliosis have been reported to have high levels of T cells [76, 77], T cell recruitment chemokine CXCL11 [78] and B cell recruitment chemokine CXCL13 [79] in their cerebrospinal fluids (CSF).

3. Late-disseminated stage: As emphasized earlier, it is important to receive an early diagnosis and start antibiotic treatment for LD. If left untreated, Bb will persist and disseminate all over the body. It can take months to years for LD to reach the late disseminated stage. This stage is characterized by severe Lyme arthritis, with intermittent bouts of joint pain and swelling. Arthritis can be monoarticular or oligoarticular, and 20

primarily affects large joints [52, 80]. Fluid in the joint contains high levels of pro- inflammatory cytokines which cause severe inflammation, pain and recruit more immune cells at the site causing swelling. Even at this stage, the primary symptoms (arthritis, swelling, joint pain) is not precipitated by the spirochete itself, but our own hyperactive immune response against bacterial agonists, which results in varying tissue damage.

1.2.3.2 Antibiotic-refractory Lyme disease

Early detection and appropriate treatment with 1 to 2 weeks of oral antibiotic therapy, or 2 to 4 weeks of i.v. therapy for cases diagnosed late during infection, should promote successful resolution of Lyme disease.[81, 82] However, in rare cases, patients continue to experience proliferative synovitis or additional symptoms in other tissues for months, even years after complete antibiotic therapy. This condition is termed as antibiotic- refractory Lyme disease. Several factors dictate the rate and extent of resolution of Lyme arthritis after a course of antibiotics. The amount of synovial pathology present and the degree of tissue damage occurred prior of start of treatment plays a major role and can extend resolution time for Lyme-associated sequellae by years. However, the primary factor contributing to extended periods of Lyme sequelae is the extent of immune dysregulation induced and tissue damage that occurred during the infection.

Similarities between mechanisms of rheumatoid arthritis (RA - autoimmune disease) and some cases of Lyme arthritis have been drawn due to the prevalence of RA- associated alleles HLADRB1*0401, *0101 in these patients.[83] Auto-reactive B-cell and

T-cell responses have been identified to play a role in Lyme arthritic pathology and may be associated in perpetuating the arthritis if left unregulated. Elevated levels of antibodies

to endothelial cell growth factor are more pronounced in patients with antibiotic-refractory

Lyme arthritis.[84] This is coupled with reduced frequencies of D25+FoxP3+ T cells and invariant natural killer (NK) T cells in synovial fluid of antibiotic-refractory Lyme arthritis patients.[85, 86] Hence, autoimmunity may underlie the major pathogenesis of antibiotic- refractory Lyme arthritis in a minority of patients.

1.2.3.3 Post-treatment Lyme disease syndrome (PTLDS)

About 10% of patients with Lyme disease present with persistent musculoskeletal pain and cognitive dysfunction even after recommended antibiotic treatment. Unlike antibiotic-refractory Lyme arthritis patients, PTLDS patients do not complain of swollen joints and joint pain. While the exact mechanism for pathogenesis of PTLDS still remains a conundrum, several theories have been postulated for this condition. These include residual tissue damage, persistence of an inflammatory state due to presence of inflammatory B. burgdorferi components and/or a form of cytokine-induced sickness behavior due to previously high levels of circulating cytokines. Systemic presence of inflammatory cytokines can induce neurobehavioral dysfunction in other conditions, leading to persistent cognitive defects. On the other hand, nonspecific symptoms experienced in PTLDS such as fatigue and pain are common in the general population as well and cannot be successfully attributed to a single cause.[87] Since no test can definitively prove the clearance of Bb infection in humans, several animal models have been developed for testing. Studies using two–photon intravital microscopy have been developed to investigate the fate of B. burgdorferi in mice after antibiotics. [88] While doxycycline therapy rapidly cleared viable Bb, Bb-associated inflammatory products (antigens and 22

DNA) were detected for extended periods adjacent to cartilage and in certain tissues.

Xenodiagnosis performed by feeding uninfected ticks on these mice occasionally detected

Borrelia DNA as well. This study provided visual evidence that Bb-associated inflammatory debris can persist in host tissue for extended periods after antibiotic therapy, which could be a potential cause for persistent symptoms even after antibiotic treatment for Lyme disease.

The key point to note in this entire section is this: all symptoms of Lyme disease, irrespective of the stage of the disease, are a result of an overactive host immune response causing the damage. The pathogen itself does not secrete any harmful toxin causing tissue degradation or induce cell death. In fact, as previously mentioned in section 1.1.1, humans are dead-end hosts and do not contribute at all to maintenance of the Bb enzootic cycle.

The pathogen gains no benefit from infecting humans. It is a stochastic event where humans acquire this pathogen and elicit a dysregulated immune response resulting in Lyme disease pathology. Hence, while Lyme disease is an infectious disease with Bb as the causative agent, the pathology is precipitated by the inflammatory response generated by the host against the pathogen. In the next section, we will discuss the various aspects of the host immune response against Bb.

1.2.4 Immune response in Lyme disease

The host immune response to any pathogen is classically segregated into two major branches: innate immunity and adaptive (or acquired) immunity. While both, the innate and adaptive immunity are critical for immune clearance of any pathogen, their mode of activation and the kinetics associated with their activation vary considerably.[89-92] In this 23

section we will discuss the general paradigm of innate and adaptive immunity and their specific role during Lyme borreliosis.

The innate immune system comprises of cellular elements which include macrophages, dendritic cells, mast cells, neutrophils, eosinophils, Natural killer (NK) cells and NK T cells. These cellular defenses are augmented by a vast repertoire of humoral components which include well-characterized components, such as complement proteins,

LPS binding protein, C-reactive protein and other pentraxins, collectins, and antimicrobial peptides, including defensins.[93, 94] While cellular components of innate immunity are responsible for directly killing the pathogen, circulating innate immune proteins act by sensing of microbes and effector mechanisms to facilitate clearance of the infection. The innate immune system utilizes a limited repertoire of pattern recognition receptors (PRRs) to detect invading pathogens, but compensates for this limited number of invariant receptors by recognizing conserved microbial domains or pathogen-associated molecular patterns (PAMPs) that are shared by large groups of pathogens. This lack of specificity towards a pathogen is also supplemented with a fast-acting response. Innate immunity can activate a response to a pathogen within minutes of exposure and begins to elicit a protective inflammatory response. This inflammatory response generated by the innate immunity (initially by resident immune cells) enables 1) activation of surrounding resident immune cells, 2) recruitment of granulocytes and antigen-presenting cells (APCs) from the vasculature, and 3) the activated APCs prime the adaptive immune response by presenting various pathogenic epitopes to T cells and secrete cytokines necessary for T cell proliferation.

The innate immune response is capable of effectively combating many pathogens, however, as discussed earlier, they are reliant on germline-encoded receptors to recognize microorganisms that can evolve more rapidly than the hosts they infect. [92] As expected, several pathogens have developed and evolved capsules which allow them to obscure common PAMPs and thereby avoid being recognized by professional phagocytes.

Similarly, intracellular pathogens can avoid clearance from macrophages and neutrophils by residing within host cells. To counteract these challenges, the adaptive immune response has evolved its recognition mechanisms which enables recognition of an almost infinite diversity of microbial molecules (i.e. antigens), so that each different pathogen can be targeted specifically. [91, 95, 96] While these responses are slow to develop, they are targeted against antigenic segments or epitopes which are specific for the invading pathogen.

Specific antigenic epitopes present on pathogen surfaces are presented to T cells by various

APCs within in the lymph nodes. While B cells are a component of adaptive immunity, they can also act as APCs and stimulate T cell activation. These cells present the antigen on MHCII molecules which allows for antigen-specific T cell receptor (TCR) development on T cells. These activated T cells are broadly classified into two categories based on their function. While Cytotoxic T cells (express CD8) will directly kill any pathogen-infected cells, helper T cells (express CD4) assist in either further activating innate immune cells or promoting B cell and plasma cell development. B cells can directly get activated independent of T cell help by B cell receptor (BCR) cross-linking (by the antigen directly binding to BCR).[91, 92, 95] However, within the lymph nodes, CD40 receptor on B cells binds to CD40L on T cells which allow T cell-dependent activation of B cells. B cell activation allows for development of a specific humoral response against a pathogen. B 25

cells secrete antibodies which are specific for a particular epitope on the pathogen. These antibodies (secreted forms of BCR) bind to pathogenic epitopes and function by 1) neutralizing the pathogen directly and reduce their infectivity by either interfere with the organism's attachment to host tissue or kill the organism by inducing pores in the outer membrane, 2) opsonize the bacterial surface which allows for enhanced phagocytosis of the pathogen by professional phagocytes, or 3) enhance binding of complement proteins which can directly kill the pathogen as well as opsonize pathogenic surface for phagocytosis.[97]

Though an extracellular pathogen, the susceptible hosts are incapable of clearing

Bb from its system without therapeutic intervention. Hence, Bb can cause a persistent infection, which can last the entire lifespan of a susceptible host. The development of EM

(phagocytes/innate immunity) and the emergence of a robust Bb-specific antibody (Ab) response in infected host (B cells/adaptive immunity) is indicative of the fact that Bb has mechanisms to deceive and evade both components of the host immune response, which allows this spirochete to maintain long-term infections. To develop any novel therapeutic measures to prevent or control Bb infection, it is important to understand the immune modulating capability of this pathogen. Since a major portion of our study is dedicated to visualizing this immune response generated against Bb at the initial site of infection, it is important to discuss the current paradigms which describe the host immune response in

Lyme disease.

1.2.4.1 Innate immunity: The first responders

Like any other microbe, the innate immune response is the first line of defense against this pathogen. Since adaptive immunity is a specific response generated by the host against a particular pathogen, it requires time to be primed (by APCs) and develop specificity against the pathogen, making it slow to initiate. During the early infection phase

(24-72 hours post-infection) Bb is confined locally at the infection site, where it replicates and host-adapts, changing its outer surface protein repertoire, a mechanism it uses to evade the developing adaptive immune response. During this phase, it is the innate immunity which is responsible for controlling Bb numbers right at the site of infection. These phagocytes utilize various pathogen recognition receptors (PRRs) on their surface to recognize different pathogen associated molecular patterns (PAMPs), thus eliciting an inflammatory response. In this section we will discuss the role and function of these phagocytes in controlling Bb numbers and the various receptors associated with it.

1.2.4.1.1 Phagocytes and associated surface receptors

The innate immune response plays an essential role in controlling the bacterial population at early times post-infection.[98] Bb possesses more than 150 putative triacylated lipoproteins on its outer membrane.[11, 99] A common motif, N-palmitoyl-S- dipalmitoylglyceryl (Pam3) Cys-Ser, that is endogenous on all of these lipoproteins can be recognized by host TLR1/2. [100, 101] TLR2 acts as the primary PRR for Bb recognition and interaction of Bb with phagocyte surface TLR2 results in potent induction of cytokines like

IL-6, TNFα, IL-12 and IL-1, and chemokines like IL-8, MIP-1β and RANTES.[102-104]. This

recognition also allows for Bb phagocytosis, subsequent to which TLR2 gets internalized along with the phagosomal vacuole. The internalized TLR2 works in conjunction with

TLR8 to mediate production of both pro-inflammatory cytokines (TNF-α, IL-1β and IL-6) and the anti-inflammatory cytokine IL-10 [105].

The importance of TLR2 in Bb recognition was initially identified by using TLR2- transfected U373 human cell line which displayed 10-fold higher sensitivity to Bb lipoproteins than LPS, as measured by NF-κB translocation and cytokine secretion [106].

Further studies conducted using TLR2-/- MØs and mice confirmed TLR2 is required for significant inflammatory responses. TLR2-/- MØs failed to respond to Bb lipoprotein stimulation and exhibited impaired production of TNF-α, IL-6 and nitric oxide. As expected, the diminished inflammatory response in TLR2-/- mice is accompanied by significantly higher Bb levels in infected tissues [65, 66, 107], indicating that the innate immune responses are critical for Bb clearance in a murine host. TLR2-/- mice also present severe arthritis and carditis, which directly correlates with their increased bacterial load.

While most studies have assessed the cytokine profile and bacterial bioburden in TLR2-/- mice, direct assessment of TLR2-deficiency on phagocytic cell activity has not been conducted. In our study, we focused on visualizing the importance of TLR2 has on resident and infiltrating cells in vivo, in real-time, at the site of infection. This will be discussed further in subsequent chapters.

Several other PRRs also play a crucial role in surmounting an inflammatory response against Bb. TLR1 and TLR6 form heterodimers with TLR2 and are essential for recognition of triacylated and dyacylated lipoproteins respectively. [108, 109] TLR1 is required along with TLR2 for optimal NF-κB translocation and IL-6 production. Similar 28

to TLR2 deficient mice, TLR1-/- mice also display higher Bb burdens in bladder and skin compared to wild-type mice [110]. TLR6 on the other hand, appears to play a lesser role since Bb does not express abundant diacylated lipoproteins on its surface. [111, 112] CD14 acts as a TLR2 co-receptor, where its binding to lipoproteins elicits both pro- and anti- inflammatory cytokine production [113-115]. CD14-/- mice display higher Bb persistence in joints compared to WT-mice, suggesting a role for CD14 in Bb clearance. CD14-dependent activation of matrix metallopeptidase 9 (MMP9) is required for neutrophil penetration of infected tissue, confirming a role for CD14 in Bb clearance [116]. This result is bolstered by the fact that increased neutrophil infiltration leads to decreased Bb numbers.[117]

These studies confirm that different phagocytes like monocytes, macrophages, microglia and neutrophils can efficiently kill Bb in vitro and certainly play a vital function in controlling Bb infection in vivo. Yet, during the course of an infection, the innate immune system is not able to contain the infection, much less clear it. Left unchecked, Bb overcomes the innate immune mechanisms and disseminates throughout the host system.

After entering the dermis, Bb is recognized by different immune cells, like macrophages, neutrophils and basophils. However, these cells are quickly replaced by eosinophils and macrophages, subduing the inflammatory response and preventing Bb clearance. [117] One explanation for this attenuation could be an early induction of the anti-inflammatory cytokine Interlukin-10 (IL-10) observed during B. burgdorferi infection.

1.2.4.1.2 Bb-elicited IL-10 response

Interlukin-10 (IL-10) is a member of an IL-10 cytokine family which includes IL-

20 and its sub-family members IL-19, IL-22, IL-24 and IL-26; and type III interferons 29

(IFNs). IL-10 and IL-20 subfamilies are involved in reducing tissue damage and protect tissue integrity during an infection. While IL-22 imposes its effect directly on tissue epithelial cells, promoting tissue regeneration and wound healing, IL-10 functions as a repressor for inflammatory responses, consequently minimizing tissue damage caused by excess inflammation. Hence, IL-10 is the only anti-inflammatory cytokine in this group and can exert its immunosuppressive effects on both innate and adaptive immune responses. It primarily functions as a switch-off signal for pro-inflammatory immune responses, thus preventing tissue damage caused by an exacerbated and uncontrolled inflammatory response, especially during the resolution phase of infection and inflammation.[118] IL-10 exerts its effect on myeloid cells by inhibiting antigen-presenting cells (APCs) function and pro-inflammatory cytokine production. It can also directly inhibit memory Th17 and Th2 cells while enhancing Foxp3+ regulatory T cells (Tregs) survival and activity, allowing tissue repair and restoration of homeostasis.

Deficiency in IL-10 signaling is associated with inflammatory diseases, such as inflammatory bowel disease (IBD), by allowing excessive inflammatory pathology during infections; conversely, enhanced IL-10 production may contribute to chronic infection.[119]

Thus, IL-10 can act as a double-edged sword and its production requires extreme spatial and temporal regulation to allow efficient pathogen clearance followed by tissue repair.

Almost all subsets of leukocytes, including dendritic cells (DCs), macrophages, T cells, natural killer (NK) cells, and B cells can elicit an IL-10 response.[120] In myeloid cells such as macrophages and DCs, IL-10 production is signaled by various pattern-recognition receptors (PRRs), including Toll like receptors (TLR). TLR signaling can synergize with other co-stimulatory molecules, such as CD40, to enhance IL-10 production. The signaling 30

cascade can utilize multiple pathways including MAP kinases (MAPKs), extracellular signal-regulated kinase (ERK), PI3K and AKT pathways and NF-kB pathways; all downstream of various PRRs and play a role in fine-tuning Il10 gene regulation. Regarding

Lyme disease, IL-10 signaling partially occurs via TLR2. Loss of TLR2 significantly reduces IL-10 levels, however it does not result in complete abolishment of IL-10 production (unpublished data, Zhang et al). Irrespective of the surface receptor, NF-κB acts as the nuclear factor downstream, controlling Il-10 gene transcription during Bb infection, since inhibition of NF-kB results in loss of an IL-10 response (unpublished data, Zhang et al).

In spite of a stringent and complex signaling cascade, certain pathogenic infections can dysregulate the IL-10 response, impeding pathogen clearance and contribute to chronic disease. [121] For example, increased IL-10 expression in various mouse strains lead to chronic mycobacterial infection. Conversely, IL-10-/- mice display better control of the mycobacterial load during infection with M. tuberculosis. [122, 123] This fact holds true in clinical disease as well; higher virulence of the M. tuberculosis CH strain is associated with its ability to elicit enhanced expression of IL-10 from human monocyte-derived macrophages. Assessment of tissue and sera of patients with advanced tuberculosis consistently demonstrates elevated levels of IL-10. Hence, pathogens can use the host IL-

10 response to their own advantage.[122, 124]

Bb stimulation can elicit a potent IL-10 response from PMBCs, MØs, DCs and T cells. [67-69, 125-128] In fact, purified lipoprotein OspA alone was sufficient to induce substantial amounts of IL-10 from human and rhesus monkey PBMCs. [125] While the anti- inflammatory activity of IL-10 should diminish the pro-inflammatory cascade after an 31

infection has been successfully cleared by the initial inflammatory response, Bb infection elicits a potent IL-10 response at significant levels within 24h of infection, which is much earlier than most infectious agents.[67] Hence, this IL-10 response was hypothesized as an immune evasion strategy utilized by Bb to curb the Bb-induced early inflammatory response.

To test this hypothesis, Bb persistence was evaluated in IL-10-deficient mice. Loss of IL-10 significantly enhanced Bb clearance however this was accompanied by increased

[129] -/- inflammation in joint tissues. IL-10 mice demonstrated ~ 8-fold higher ID50 when compared to WT mice. Further in vitro investigations revealed IL-10 modulated Bb clearance by suppressing the production of pro-inflammatory cytokines (TNF-α, IFN-γ and

IL-6) [68]. Viable Bb elicited higher levels of IL-10 from MØs than lipoproteins alone. Bb- induced IL-10 in MØs potentially altered gene regulation, and modulated cytokine and chemokine production. [130] In vitro assessment also identified that Bb-induced IL-10 can suppress MØ/DC phagocytosis and activation [69]. These studies confirmed the role of IL-

10 as a potent immune-evasion strategy employed by Bb to abrogate the early immune response. Hence, in our study, we aim to visualize the effect on IL-10 on phagocytic cells and its effect on Bb persistence in vivo, in real-time, at the site of infection.

1.2.4.2 Adaptive immunity

Bb infection of susceptible hosts does elicit a robust adaptive immune response, however, the importance of the adaptive immune response in controlling Bb numbers and

Lyme-associated pathology is not completely understood. Mice lacking either B cells

(μMT) [71] or lacking both B and T cells (SCID, RAG-/-) [131, 132] show increased bacterial

burden, as well increased arthritis. Passive transfer of B cells, but not T cells, controlled disease progression in these deficient mouse lines, suggesting B cells have a greater effect than T cells in controlling Bb infection. In this section, we will discuss the role of adaptive immunity in Lyme disease.

1.2.4.2.1 Role of B cells

In comparison to most bacterial infections, the humoral response elicited against Bb is quite delayed. Bb infection does induce a strong B cell/humoral response, albeit this response is slow to develop and effective primarily during the early stages of infection [70, 133, 134]. The antibodies generated can protect a naïve mouse from infection if passively transferred before or soon after infection. However, these antibodies are ineffective at conferring protection if the passive transfer is carried out >5 days post- infection. [135] This highlights the fact that the humoral response loses its potency against host-adapted Bb and play a role only during the initial phase of the infection (before the spirochete is presumed to have undergone host-adaptation).

The observed early protective role of immune serum is T cell-independent, since immune serum isolated from CD40L or CD4+ T cells deficient mice also confer some protection during Bb challenge. The antibody response is multi-variant as seen by presence of antibodies against multiple Bb antigens. The initial B cell response mainly consists of

IgM-secreting cells, largely against T–independent antigens induced early in the lymph nodes and later in bone marrow. While IgM is effective at clearing the spirochete in blood, its large size may hinder this immunoglobulin from effectively clearing the bacteria from skin and other tissues in the absence of inflammation (which is shut down because of IL-

10 induction by Bb). [70] Marginal zone B cells (MZB) are the main source of antibodies against T-independent Bb antigens and MZB-deficient mice display reduced production of

Bb-specific IgM and IgG antibody, increased Bb burden in blood and exacerbated arthritis.[136] B-cell accumulation in lymph nodes also seems to be independent of MyD88 and T cells. This data suggests that B cells are important not only for controlling Bb infection, but also for reducing arthritis severity.

1.2.4.2.2 Role of T cells

The impact of T cells in Lyme disease is still a controversial topic. While T cells are generally considered not to play a major role in controlling Bb numbers, they have been implicated to be primarily associated in controlling Lyme pathology.

Human T cell responses during the development of Lyme disease, especially the

Th1 type, are commonly associated with persistent arthritis even after antibiotic treatment.

A skewed Th1 response by adoptive transfer of CD4+ T cells leads to the resolution of carditis [71], but causes arthritis via blockade of B7/CD28 costimulatory signaling [137, 138].

IL-4 and IL-6 are typical Th2 response cytokines, which if blocked can skew the host to a

Th1 response. Loss of IL-6 in mice is accompanied by increased incidence of arthritis [139] whereas IL-4 deficient mice exhibited unaltered ankle swelling and arthritis severity score

[140, 141]. Contrary to this, blockage of IFN-γ and IL-12 which are typical Th1 response cytokines, demonstrated either inconsistent results or resulted in reduced incidence of arthritis [140-142]. Taken together, there is no clear trend that can be concluded based on the above studies. However, in general, a skewed Th1 response is mostly related to an increased incidence of arthritis. A recent study identified bystander activation of CD4+ and 34

CD8+ T cells (independent of Borrelia-specific T cell-receptor interaction) leads to arthritis-promoting IFN-γ, creating an inflammatory environment seen in the synovial tissue of patients with post-treatment Lyme disease. Whiteside et al identified high levels of TLR2 expression on these bystander T cells. Cell transfer experiments revealed TLR2 as a critical mediator of T cell activation following Bb infection, which results in enhanced

IFN-γ production and Lyme arthritis.[143]

Although T cells are activated, the T-dependent germinal center responses are not maintained.[70] T follicular cells isolated from the lymph nodes of mice infected with B. burgdorferi rapidly differentiate B cells to antibody secreting plasma cells, but appear defective in promoting proliferation of these B cells. This leads to production of a strong, but short-lived antibody response rather than a protective long-lasting memory response regarding antibody production. Thus, T cells, though not critical, are still important for controlling Lyme disease manifestations.[71]

1.3 Intravital Microscopy: The live picture from within

Intravital microscopy (IVM) is a microscopy technique which allows visualization of various cellular activities within intact tissues of live organisms within its natural environment in vivo. This is a major advantage in the field of biomedical science as it provides us the means to study a live cellular process within a live animal, in real-time (as the process is occurring in vivo) as oppose to ex vivo or in vitro techniques which require tissue sections obtained from euthanized animals and supposedly should mimic the natural milieu. Several kinds of microscopes can be utilized to conduct IVM: widefield, confocal,

multiphoton, etc. For our study we utilized laser confocal microscopy for IVM to visualize

Bb in intact mouse ear skin.

1.3.1 Intravital Microscopy to study Lyme disease

Bb, the spirochetal causative agent of Lyme disease is an obligate extracellular parasite. Bb shuttles between the arthropod vector and mammalian hosts during its lifecycle, and subsequently modulates its outer surface protein expression profile based on the environment it currently resides. For instance, Bb changes expression of OspA, OspC and VlsE on its surface depending on whether it is in a tick or vertebrate host.[144, 145] This makes replication of the Bb natural infectious environment in vitro extremely difficult.

Many virulence mechanisms employed by Bb are not functionally activated in an in vitro setup and hence can only be studied in its in vivo milieu. For example, 1) Bb motility and chemotaxis can only be truly evaluated in vivo, since the complexity of the host dense tissue cannot be effectively replicated in vitro. High variance observed between Bb motility patterns in commonly used in vitro media and skin tissues in vivo, only bolsters this argument.[146]. Research also indicates that the spirochete utilizes different chemotactic homologs depending on the host it occupies, for a variety of functions[50, 51] (unpublished data, Moledina et al); 2) Results obtained from studying the host immune cell-pathogen interactions also display high levels of variability between an in vitro and in vivo environment. While MØs and neutrophils efficiently clear the pathogen in vitro, Bb can effectively escape the immune cell clearance in vivo resulting in a persistent disseminated infection.[29, 67, 69, 147] Hence an accurate assessment of the immune evasion mechanisms utilized by Bb can only be visualized in an in vivo setup. 36

In advent of this knowledge, several research groups have utilized IVM to investigate Bb within its natural host environment. Fluorescent Bb strains have been generated either by introducing shuttle vectors containing the gfp gene in the bacteria or by insertion of a gfp expression cassette into the cp26 plasmid of the bacteria, to facilitate in vivo imaging. Using this GFP-expressing Bb, Dunham-Ems et al identified the characteristic biphasic mode of dissemination of Bb within ticks while investigating Bb translocation in ticks.[148] Investigation of Bb translocation in the microvasculature of live mice using IVM allowed Moriarty et al to characterize this multi-stage process which involves short-term interactions and stationary adhesions.[149] Bockenstedt et al investigated Bb persistence and demonstrated that months after infection with GFP-Bb, motile bacteria along with GFP+ deposits (Bb antigen and cellular debris) were observed in both ears of infected mice. After administration of antibiotics, motile GFP-Bb were no longer observed, suggesting that antibiotic treatment cleared the bacteria. However, GFP+ debris persisted in the ear of these mice which could still be immunogenic; providing an explanation for development of antibiotic-refractory Lyme disease.[88] Several groups including ours have investigated Bb persistence and motility patterns observed within the mouse dermis. We along with others have determined the velocity of the bacteria in vivo and confirmed that the spirochete traverses ≥ 40 times faster than any observed host immune cell (Lavik et al, unpublished data).[31, 32] Based on these results, it is a safe assessment to say that IVM of fluorescent Bb has great potential to provide accurate and detailed information for Bb, motility, dissemination, and persistence providing a snapshot of the interactions occurring between the pathogen and host immune cells in vivo.

1.3.1.1 Preliminary assessment of Bb in ear tissue

Figure 1 – 5: B. burgdorferi persistence at site of infection in mice. Bb numbers in each ear (1 mm2 area of the ear consisting of 10 different viewing fields) at different times post-injection. Each blue circle represents average number of bacteria per 1mm2 of ear. n ≥ 12; Error bars represent standard error of mean (SEM).

As emphasized in the previous sub-section, accurate assessment of Bb virulence and immune evasion mechanisms necessitates an in vivo model. While in the Wooten Lab,

Dr. J. P. Lavik and Dr. Padmapriya Sekar initiated and standardized a protocol which utilizes IVM for visualizing Bb and its interaction with various immune cells within intact murine skin tissue and in real-time. Using these techniques, Dr. Sekar established a Bb persistence curve within intact murine skin tissue of a WT mouse. She further went on to 38

display a potential negative-correlation between Bb numbers and appearance of Bb- specific antibodies. Also, she described the initial neutrophil infiltration observed at the site of infection after Bb infection. In this sub-section we will discuss these findings, as they will determine the rationale for our current study.

1.3.1.1.1 Bb dissemination and persistence at the injection site

In order to directly visualize bacterial persistence at the injection site using IVM,

WT-GFP (wild-type bacteria expressing green fluorescent protein) Bb (explained in section 2.1) were injected intradermally (i.d.) in the center of the ear. At the indicated times post-infection, a confocal microscope was utilized to visualize the spirochetal number present within the ear tissues of anesthetized mice and the number of Bb in each viewing field was manually counted. The spirochetes were confined to the initial circular injection site of 4mm in diameter for the first 48 hours of the infection. By day 4 post-infection, Bb was observed to have disseminated throughout the ear (at least a circular area of 8mm in diameter). Shortly after injection, Bb rapidly proliferated and reached peak levels by 5-8 days post-injection. At this stage the innate immune response should ideally control bacterial proliferation. However, it appears unable to do so and Bb numbers increase by almost 100-fold. After achieving a maximum level, Bb numbers begin to plummet thereafter, such that by day 14 the bacteria numbers reach a basal level. The mice used for the above experiment were continuously imaged at regular intervals for more than 2 years post-injection. Dr. Sekar demonstrated that, once Bb numbers reach a basal level at day 14

Figure 1 – 6: Correlation between B. burgdorferi specific antibody production and bacterial numbers. B. burgdorferi-specific IgG serum antibody levels in mice infected with 5*104 WT-GFP Bb were quantified using ELISA. Each green triangle represents the average Bb-specific antibody level from 4-6 mice (right axis). Each blue circle represents average number of bacteria per ear (right axis; taken from Figure 1 – 5); Error bars represent standard error of mean (SEM).

post-infection, the bacteria continue to maintain at that level with no significant changes and persist in the ear for more than 2 years. Overall, these findings using IVM demonstrated a very distinct persistence curve which is characterized by 1) rapid increase in Bb numbers right after infection. This is the phase where the host innate immunity should be responding to the infection and establish control, however it appears unable to do so; 2) after achieving peak bacterial numbers, Bb numbers plummet reaching a basal level and lastly 3) once a basal level is achieved, Bb continue to persist at the injection site for more than 2 years

(Figure 1 – 5). Intriguingly, even at 2 years post-infection, Bb maintain their characteristic back-and-forth motion and continue to traverse at extremely high velocity. This data also supports the fact that Bb velocity is essential not only for its infectivity but also for its persistence.[28, 29]

1.3.1.1.2 Correlation between Bb numbers and Bb-specific antibodies

Previous work from our lab showed that Bb-specific antibodies generated in mice by day 7 post-injection have potent bactericidal activity.[72] Since the Bb persistence curve displays a significant decrease in Bb number at the site of infection around the same time,

(Figure 1 – 5 ), kinetics associated with Bb-specific antibody production was assessed to determine if appearance of Bb-specific antibody corresponds with the decrease in Bb levels. Serum isolated via retro-orbital bleeds from long-term WT-GFP Bb infected mice was analyzed at various times post-injection, and Bb-specific antibody production was assessed by ELISA. As shown in Figure 1 – 6, Dr. Sekar identified that Bb-specific IgG antibody levels appeared at significant levels by Day 5 and the levels continued to increase over time. This suggested that Bb-specific antibodies may be involved in the early decrease in bacterial numbers since a potential negative correlation exists between appearance of

Bb-specific antibodies and decrease in Bb numbers starting at Day 7. Intriguingly, even if antibody-mediated killing is responsible for this decrease, they are unable to completely clear Bb, since Bb reach a basal level by day 14 and persist at that level indefinitely.

Overall, these results suggest a role of Bb-specific antibodies in early control of Bb numbers, but the activity is short-lived as the antibodies lose their potency once the spirochete have host-adapted.

1.3.1.1.3 Early neutrophil response to Bb infection in skin

Figure 1 – 7: Early neutrophil response to B. burgdorferi at the injection site. Neutrophils present at the injection site were manually counted at the indicated times post-injection. Each red triangle represents the average number of neutrophils per ear (right axis). Each blue circle represents average number of bacteria per ear (right axis; taken from Figure 1 – 5). Error bars represent standard error of mean (SEM).

During any extracellular infection, a successful primary immune response includes neutrophil recruitment and infiltration of infected tissues. Being the fastest immune cell within mammalian tissues, neutrophils feasibly provide the best possibility of capturing and killing the highly motile spirochete compared to the slower-moving tissue-resident immune cells. To characterize neutrophil infiltration in response to Bb infection, LysM- eGFP mice were infected with WT-GFP Bb. At various times post-infection, these mice were imaged to observe neutrophil infiltration (Figure 1 – 7). Dr. Sekar observed neutrophil

infiltration at the site of infection starting within an hour of infection and achieved maximum numbers by 6hours post-injection. Neutrophil numbers started to decline by 48 hours and neutrophil numbers are seen reaching a basal level by day 10 post-injection.

Overall, this study indicated that while injection of the spirochete is able to initiate neutrophil infiltration at the site of infection, there exist some immune dysregulation which prevents continuation of neutrophil influx in spite of an exponential increase in Bb numbers during the first 7 days of infection.

1.4 Goals for dissertation

Previous studies have showed that APCs and neutrophils are capable of killing Bb effectively in vitro. Macrophages and dendritic cells can not only recognize Bb but also internalize the spirochete resulting in immune cell activation and release of pro- inflammatory cytokines/chemokines. However, in vivo this pro-inflammatory response appears to be dysregulated since Bb does not get effectively cleared from host tissues consequently leading to a persistent infection. Hence, the primary goal for this study is to visualize the immune dysregulation occurring in vivo which allows Bb to cause a persistent infection. In vitro assessment from our lab identified that Bb can induce a potent IL-10 response. IL-10 is an anti-inflammatory cytokine and primarily acts as an inhibitory cytokine which diminishes the pro-inflammatory cascade after an infection. We hypothesize that this early IL-10 response renders the innate immune response ineffective in controlling Bb numbers resulting in the aforementioned immune dysregulation. To achieve this goal, the following objectives were proposed:

1. Identify cell types involved in IL-10 production in response to Bb at the site of

infection.

2. Visualize the effect of Bb-induced IL-10 on innate cell function at the site of infection.

Both B cells and T cells have been implicated in asserting control on Bb persistence.

Previous studies performed by Dr. Sekar suggested Bb-specific antibodies might play a role in the significant decrease in Bb numbers observed between Day 8 and Day 14 post- infection. While direct contribution of T cells on Bb persistence still remains largely unanswered, they are essential for providing co-stimulatory activation to B cells and elicit a Bb-specific antibody response. On the other hand, T cell-independent B cell activity also have been found to be important in controlling Bb numbers. Our second goal is to evaluate the role of the adaptive immune response in the significant decrease in Bb number observed between Day 8 and Day 14 post-infection. To achieve this goal, Bb persistence was evaluated in B-cell and T-cell deficient mouse models.

Bb motility has been identified as an important virulent factor for this spirochete.

Non-motile Bb are unable to cause a persistent infection. The spirochete can travel at extremely high velocities in dense host tissues, achieving velocities of almost 230 µm/min, and that these spirochetes maintain this high velocity for the entire lifespan of the infected host. Hence, this high bacterial velocity has been proposed as an immune evasion strategy employed by Bb to simply outrun the relatively slow-moving host immune response. Our last goal for this study is to determine the importance of this high Bb velocity in spirochetal evasion of innate immune responses. To achieve this goal, we utilized mutant Bb species

(which were evaluated to move at different velocities in vivo) to visualize the effect of differential Bb velocity of spirochetal persistence at the site of infection.

Chapter 2

Material and Methods

2.1 Bacterial strains and growth conditions

A low-passage, virulent strain of B. burgdorferi, B31-A3-K10, was used as the wild- type (WT) clone [150] (a kind gift from R. Rego and P. Rosa, Rocky Mountain Laboratories,

NIH). The K10 strain is a derivative of parent B31-A3 strain with an inactivated bbe002 gene (restriction modification system) located in lp25. This strain was utilized by Dr.

Motaleb for construction of WT-GFP, ΔcheY3-GFP and ΔflaA-GFP Bb strains.[51] For the

ΔcheY3 strain, the cheY3 gene on the flaA flagellar operon was inactivated by replacing the coding sequence of cheY3 with the kan sequence for kanamycin resistance. For the ΔflaA strain, the flaA gene on the flaA flagellar operon was inactivated by replacing the coding sequence of flaA with the strep sequence for streptomycin resistance. All 3 strains (WT

Bb, ΔcheY3 Bb and ΔflaA Bb) were transformed by Dr. Motaleb with a gentamicin- resistant (PflgB-aac1) suicide plasmid containing constitutively-expressed gfp (PflaB-gfp; produce Green Fluorescence Protein [GFP]) flanked by the endogenous cp26 DNA. [29, 148]

These GFP-expressing strains (WT-GFP, ΔcheY3-GFP and ΔflaA-GFP Bb) were utilized for IVM analysis. The genome of all utilized strains possesses 12 linear and 9 circular

plasmids, for a total of 21 plasmids, in addition to a 960-kbp linear chromosome. [11, 12]

They lack circular plasmid 9 (cp9) but retain infectivity.[150] Bb cultures were grown at

33°C in Barbour-Stoenner-Kelly II (BSK-II) media with 6% rabbit serum (Pel-freez

Biologicals) and appropriate antibiotics. All cultures were temperature shifted to 37°C 24- hours prior to infection to allow Bb to host-adapt for mammalian infection. When required, culture media was supplemented with appropriate antibiotics at the following concentrations: 200 μg/ml kanamycin, 100μg/ml streptomycin and 40μg/ml gentamycin.

Bb were enumerated using a Petroff-Hausser counting chamber (Hausser Scientific) and diluted as needed to get a concentration of 105 Bb in 10µl bolus.

2.2 Mice

C57BL/6 (BL/6) mice at 5-6 weeks were purchased from Charles River Laboratories

(NCI-Frederick) River and housed in a conventional mouse room. These mice served as a

WT mouse model for the entire study. LysM-eGFP-ki (LysM+) and Iaβ-eGFP-ki (Iaβ+) mice were a kind gift of Dr. Akira Takashima, University of Toledo College of Medicine and Life Sciences. In LysM+ mice, enhanced GFP (EGFP) gene was inserted downstream in the murine lysozyme M (lys) locus, causing any immune cell that expressed the enzyme lysozyme-M (mainly neutrophils and to a lesser extent macrophages) to fluoresce green.

[151] Similarly, in Iaβ mice, the I–A β gene of the MHC II molecule was replaced with a construct encoding a protein composed of an intact class II β-chain fused to an E-GFP moiety at its carboxy terminus. Hence, every cell which expressed MHC II on their surface

(primarily APCs) fluoresce green in color. [152] IL-10-deficient (IL-10-/-) mice (B6.129P2-

Il10tm1Cgn/J; stock 002251) and TLR2-deficient (TLR2-/-) mice (B6.129-Tlr2tm1Kir/J; stock

004650) were purchased from the Jackson Laboratory. In the IL-10-/- mouse, the Il-10 gene was disrupted by introducing a targeting vector designed to replace codons 5-55 of exon 1 of the targeted gene with a 24 bp linker and a neo expression cassette. [153] A termination codon was also introduced into codon 3. For generating TLR2-/- mice, the sequence encoding for the C-terminus of the extracellular and part of the transmembrane domains on tlr2 gene was disrupted by targeting vector containing neomycin resistance and herpes simplex virus thymidine kinase genes. [154] For visualization of neutrophils and APCs in an

IL-10- and TLR2-deficient background, LysM+ and Iaβ+ mice were backcrossed with IL-

10-/- and TLR2-/- mice to generate IL-10-/- LysM+, IL-10-/- Iaβ+, TLR2-/- LysM+ and TLR2-

/- Iaβ+ mice. Activation-Induced Cytidine Deaminase-deficient (AID-/-) mice were a gift from Dr. Patricia J. Gearhart at the laboratory of Molecular Biology and Immunology,

National Institute on Aging, National Institutes of Health. To generate these mice, a target vector was used which replaces exon 2 and the 5’ end of exon 3 (encoding the cytidine deaminase motif) of the AID gene with a neomycin resistance (NeoR) gene, hence disrupting the AID gene.[155] B cell-deficient (µMt-/-) mice (B6.129S2-Ighmtm1Cgn/J; stock 002288) were purchased from the Jackson Laboratory. These mice are devoid of mature B cells since a neomycin resistance cassette is inserted to disrupt one of the membrane exons of the gene encoding immunoglobulin heavy chain of the class µ.[156]

Athymic/nude (Foxn1-/-) mice (NU/J, stock 002288) were purchased from the Jackson

Laboratory. These mice are homozygous for the nude spontaneous mutation (Foxn1nu, formerly Hfh11nu) and demonstrate abnormal hair growth and defective development of the thymic epithelium. [157] These athymic mice are devoid of mature T cells.

T cell receptor-deficient (TCR-/-) mice (B6.129P2-Tcrbtm1Mom Tcrdtm1Mom/J; stock

002122) were purchased from the Jackson Laboratory. These mice are deficient in alpha- beta and gamma-delta T cell receptors and are completely devoid of T cells.[158] Tiger mice

(B6.129S6-Il10tm1Flv/J; stock 008379) were purchased from the Jackson Laboratory. This reporter strain has GFP inserted downstream of the IL-10 promoter, which allows monitoring and detection of cells committed to IL-10 production.[159] All mice, except

C57BL/6 mice, were bred in specific pathogen-free housing conditions. All mouse strains were housed in the Department of Laboratory Animal Medicine at the University of

Toledo, College of Medicine and Life Sciences, Health Sciences Campus. All usage protocols were reviewed and approved by the Institutional Animal Care and Use

Committee. All mice used for this study have been listed in table 2.1.

Table 2.1: List of mice used in this study

Innate NK cells Helper T cells Cytotoxic T B cells cells cells WT +++ +++ +++ +++ +++

Nude +++ - - - +++ TCR-/- +++ +++ - - +++ µMT ++ +++ +++ +++ - AID-/- ++ +++ +++ +++ +/- TLR2-/- - +++ +++ +++ +++ IL-10-/- +++ +++ +++ +++ +++ (+++) represents normal immune response; (+++) represents increased immune response whereas (++) represents reduced immune response; (+/-) represents partial deficiency whereas (–) represents complete deficiency

2.3 Intravital microscopy

Intradermal injection of the ear: Two days prior to injection, the dorsal ear surface was depilated (Nair) and rinsed 3 times with H2O after 30 seconds of application. Low-passage cultures of WT-GFP, ΔflaA-GFP or ΔcheY3-GFP were directly counted using a Petroff-

Hausser chamber and dark-field microscopy to contain the standard 105 B. burgdorferi inoculum in 10μl BSK-II medium; other concentrations were generated to perform dose- response studies. B6 mice were anesthetized via intraperitoneal (i.p.) injection of a standard restraint cocktail containing ketamine (100 mg/ml, Hospira, Inc.), 65 to 75 mg/kg of body weight; xylazine (20 mg/ml, Ben Venue Laboratories), 6 to 8 mg/kg; and acepromazine

(10 mg/ml, Boehringer Ingelheim), 1 to 2 mg/kg using a 28G½ insulin syringe (Becton-

Dickinson). Anesthetized mice were injected with 105 of appropriate eGFP expressing Bb intradermally in a 10μl bolus using a 31G insulin syringe (Becton-Dickinson) in the dorsal surface of the mouse ear. Mouse ears were divided into 5 quadrants using a template developed by the Wooten lab (Figure 2 – 1); C: center, P: proximal, D: Distal, S: Superior and I: Inferior. The inner circle (C) is 4mm in diameter while the outer circle is 8mm. Bb were always injected in the center quadrant. Infected mice were allowed to rest for 6 hours post-infection and then prepared for imaging. Infected mice were re-anesthetized at the indicated time-points and imaged for bacterial or immune cell quantification.

Intravital microscopy of Bb-infected mice: Imaging was performed using an Olympus

FV1000 laser confocal inverted microscope system. For imaging, anesthetized mice were placed on a 37°C heated imaging stage and the ear was temporarily mounted on a coverslip, with the dorsal surface facing downwards. For bacterial enumeration 3D Z-stack images

were collected at each time point using a 20× dry objective with a 2× optical zoom. The Z- stack volume is comprised of an 80µm depth in the dermal layer of the skin, which begins below the epidermal layer. To identify the position of the dermal layer of the skin, the hexagonal epidermal skin cells are visualized. The 80µm of assessed tissue begins below the epidermal layer (below the layer of hexagonal skin cells). The epidermal layer was identified to be 20µm deep starting from the surface of the dorsal layer of ear skin. The dermal layer was identified to be 80µm deep starting at 20µm below the dorsal surface of the ear skin and extending up to 100µm below the dorsal surface of the ear skin. For bacterial cell enumeration, 2 images were captured for each of the 5 quadrants (10 images per ear) and Bb numbers were manually enumerated. Each point represents the sum of total

Bb present in all 10 compressed Z-stack images. Bacterial velocity was determined using a time-lapse video compiled of 2D images collected at 20× dry objective with a 2× optical zoom at 1 frame/1100 milliseconds for 66 seconds (60 images) at different times post- injection. Immune cell numbers and motility assessment were visualized by procuring a time-lapse video comprising of 3D Z-stack images at 10× dry objective with a 1× optical zoom (or 2x optical zoom for tiger mice) at 1 frame/minute for 60 minutes at different times post-injection. Velocity and motility assessments were conducted by tracking immune cells moving in a single Z-plane over a period of time. Cell enumeration was assessed by compressing the entire Z-stack volume (80µm) and each cell present in present in the 1st compressed image was manually enumerated.

Image Analysis: For bacterial and immune cell enumeration, the entire Z-stack volume was compressed, and bacteria or immune cells present in that image were manually enumerated using ImageJ software (National Institutes of Health). Bacterial and immune 51

Figure 2 - 1: Ear template used for mouse infection. The template divides the ear into 5 quadrants. Bb is always injected in the central quadrant.

cell velocities were assessed by manually tracking bacterial or immune cell present at one particular Z-plane for 60 frames in a time-lapse video. All 60 frames of the dataset were assessed using MetaMorph software (Molecular Devices), and every bacterium and immune cell noted on the first image was tracked through 60 images. For tracking, one end of the spirochete was chosen, and it was followed until that end is not visible for more than

1 frame. Observations for bacteria and immune cells which stayed in the field of view for less than 5 frames were discarded. Similar methods were used for tracking Iaβ+ cells where one end of the cell process was used to track movement of cells. The center of the cell body was used for tracking LysM+ cells and velocity determination. Tracking of bacteria and immune cell using MetaMorph software (Molecular Devices) provided, 1) total distance travelled, 2) displacement from initial site and 3) average velocity achieved by the

spirochete and immune cells. Total distance was denoted as the distance covered by the cell during the course of tracking event. Displacement was calculated by measuring the linear distance between the initial and final position of the cell. Average velocity was determined by taking a ratio of the total distance travelled by the cell and the time required by the cell to cover that distance. Confinement ratio was defined as the ratio of the displacement of a cell to the total distance it travelled. The mean diameter of Iaβ+ cells was calculated using ImageJ and represented as an average of the diameter measured twice from perpendicularly opposite poles of the cell. The images in ImageJ were calibrated to the same scale as obtained from the Flowview software (Olympus microscope). Cell area was determined by using the “Analyze Particle” selection tool on Image J. Briefly, a 2-D

XY image at a particular depth was converted to an 8-bit image. An automated binary process (in-built in Image J) was utilized to set up threshold and adjusted to get clear cell margins. Using this binary image, which segregates cell area and background, cell area was determined by utilizing the “Analyze Particle” selection tool to conduct area measurement of complex objects (such as macrophages which have ameboid-like undefined shapes). A

Pearson coefficient was calculated by plotting the area of a cell with the corresponding diameter of that particular cell. Pearson coefficient allows to assess the relation between the 2 parameters; namely cell area and cell diameter.

Immune cell morphology in tiger mouse: Time-lapse videos comprising of 3D Z-stack images at 10× dry objective with a 2× optical zoom at 1 frame/minute for 60 minutes at different times post-injection were obtained to visualize IL-10 producing cells in response to Bb infection in a tiger mouse. Cellular morphology of GFP-producing cells and depth in skin were utilized to identify cell types. Cells with long branched cell processes present 53

in the epidermal layer of the skin (within 20µm from the surface of the dorsal ear skin) were identified as Langerhans cells. Ameboid shaped cells present in the dermal layer of the skin (ranging from 20µm to 100µm below the dorsal surface of the ear skin) were classified as macrophages/ dermal dendritic cells. These cells have smaller, fewer and shorter cell processes than Langerhans cells, they are present in the dermal layer of the skin as opposed to the epidermis and are ~ 10-fold faster than the Langerhans cells. Small, round fast-moving cells in the dermal layer of the skin which did not possess any cell processes were identified as potential neutrophils. Neutrophils are more spherical with an average diameter of around 10-12 µm and can move achieve velocities of ~ 6-8 µm/minute.

Macrophages are ameboid in shape and have a diameter range of 20-25 µm. They move much slower than neutrophils at around 1.5-2 µm/minute.

2.4 Quantitative measurement of B. burgdorferi in murine ear skin

Bb persistence in skin was evaluated for mice infected with ΔcheY3-GFP Bb to determine spirochetal numbers in ear skin at different times post-infection. For certain experiments, the bacterial numbers in murine skin were quantified by qPCR (quantitative real-time PCR) as previously described.[51] Briefly, mice infected with ΔcheY3-GFP Bb were euthanized at indicated times post-infection and both ears were isolated. Isolated ear tissues were treated with 1mg/ml Collagenase A and 0.1mg/ml Proteinase K. DNA was then isolated from each tissue by phenol-chloroform (1:1) extraction (twice) followed by a chloroform extraction, and DNA was precipitated by adding ethanol. RNA digestion was carried out between the 2 sets of phenol-chloroform (1:1) extraction using 10μg/ml

DNAse-free RNAse. Isolated DNA was resuspended in Tris-EDTA (TE) buffer, and the 54

DNA quantity and quality was estimated by measuring the absorbance at 230, 260 and

280nm. Stock DNA was then diluted to 50ng/μl (O.D.260=1 samples) for quantitative PCR

(qPCR) analyses. qPCR was performed using a Light Cycler 96 (Roche Diagnostics).

Mouse DNA levels were determined by amplifying the nidogen gene whereas the flaB gene was utilized to quantify Bb DNA. Mouse and B. burgdorferi genomic copy numbers were determined by extrapolation to standard curves using LightCycler software (Roche

Diagnostics). The primers used to detect mouse (nidogen) and Bb genome (flaB) are outlined in Table 2.2.

Table 2.2: List of primers used in the study

Gene Primer code Sequence

Nidogen Nido-F CCA GCC ACA GAA TAC CAT CC

(mouse gene) Nido-R GGA CAT ACT CTG CTG CCA TC

Nido-Probe CCT TTC CTG GCT GAC TTG GAC ACA flab (Bb gene) FlaB-F TTG CTG ATC AAG CTC AAT ATA ACC A

FlaB-R TTG AGA CCC TGA AAG TGA TGC

FlaB-Probe CAG CTG AAG AGC TTG GAA TGC AGC

Both probes are dual labelled with 5’ 6-FAM and 3’ Iowa Black

Quencher

2.5 B. burgdorferi-specific antibody detection by ELISA

Mice were retro-orbitally bled at the indicated times for serum isolation and total- immunoglobulin (Ig), Immunoglobulin G (IgG) and Immunoglobulin M (IgM) content

were assessed by ELISA using previously described methods.[51] For detection of Bb- specific antibodies, 96-well high-binding ELISA plates (Costar) were coated with 5 μg/ml sonicated B. burgdorferi grown at 33°C and then shifted to 37°C. For sonication, the temperature-shifted culture was resuspended in PBS and sonicated using a bath sonicator

(Sonifier® Cell Disruptor). Standard lanes were coated with goat anti-mouse total immunoglobulin (IgG+ IgM+ IgA; Southern Biotech), Immunoglobulin G (IgG; Southern

Biotech) or Immunoglobulin M (IgM; Southern Biotech) in 0.1M Carbonate-Bicarbonate

Buffer (pH 9.5) to provide a quantifiable purified standard. After overnight coating, plates were washed with ELISA wash buffer (0.05% Tween 20 in PBS) and filled with ELISA blocking solution (10% FBS in PBS) before being stored at -20ºC. Serial dilutions of individual sera from uninfected or infected mice were added to the Bb-coated plates overnight before removing unbound antibodies by washing. Bound murine Bb-specific total Ig, IgG and IgM were detected using goat anti-mouse Ig, goat anti-mouse IgG or goat anti-mouse IgM conjugated to biotin (Southern Biotech), respectively. After 2–3 h incubation at room temperature, plates were washed and avidin-conjugated Horseradish

Peroxidase (avidin-HRP; Vector Labs) was added for 30 min. Plates were again washed to remove unbound avidin-HRP before adding the chemiluminescent solution (0.4mg/ml of

O-phenylenediamine and 0.01% H2O2 in citrate buffer, pH 5) and subsequently inhibiting the reaction with 1 N hydrochloride. Antibody titers were quantified by comparison with standard curves constructed using purified mouse IgG (Southern Biotech) for total Ig and

IgG quantification, and purified mouse IgM (Southern Biotech) for IgM quantification.

Chapter 3

Role of adaptive immunity in controlling Bb persistence at the site of infection

Evidence for contribution of the adaptive immune response in controlling Lyme disease symptoms dates back to 1992. Barthold et al performed Bb infections in severe combined immune deficient (SCID) mice along with their immune-competent counterparts as controls (WT mice).[160] All mice developed spirochetemia which persisted up to day

30 and resolved by day 45 post-infection when assessed using old “quantification” methodologies (i.e. Bb outgrowth from mouse tissues in vitro). While all genotypes developed arthritis and carditis by day 14 post-infection, only the immune-competent mice were able to resolve these symptoms. As expected, only the WT mice seroconverted in response to Bb infection, with high IgG titers, while the SCID-mice did not seroconvert.

This evidence demonstrated the role of adaptive immunity in resolving Lyme symptoms, however the role of adaptive immunity in controlling Bb persistence still remains elusive.

Bb is an extracellular pathogen and elicits a very potent antibody response, due to the presence of highly agonistic and immunogenic lipoproteins on its outer surface. [72, 161]

While slightly delayed, both B cells and T cells respond to Bb infection and we observe significant levels of Bb-specific antibodies by day 7 post-infection.[70] These Bb-specific antibodies are capable of bactericidal properties and have been shown to mediate direct 57

killing of Bb in vitro and in vivo.[162, 163] Passive transfer of antisera isolated from infected mice confers protection to naïve mice, and even prevents infection if administered within

≤7 days of infection.[160, 164] However, this antiserum is unable to mediate complete Bb clearance if administered at later times post-infection. This is indicative that these antibodies are effective only during early phase of the infection and unable to clear “host- adapted” Bb.

Dr. Sekar also derived similar conclusions from her study using IVM to evaluate

Bb persistence at the site of infection.[165] After achieving peak spirochetal loads by day 5-

8 post-infection, Bb numbers plummet and reach a basal level by day 14 post-infection, which they maintain for the entire lifespan on the host. The significant decrease in Bb numbers observed between day 7 – 14 corresponds to the appearance of Bb-specific antibodies, suggesting that the naturally developing antibodies may be responsible for this

Bb clearance. However, this antibody response is unable to completely clear the infection as Bb persist in skin for ≥2 years post-infection, confirming the ineffectiveness of Bb- specific antibodies on “host-adapted” Bb.

While T cells have not directly been attributed in controlling Bb infection, they do play a major role in precipitating Lyme disease manifestations. A predominated Th1 response results in exacerbated Lyme arthritis. [137-139] CD4+ and CD8+ T cells have also been implicated in production of high levels of arthritis-promoting IFN-γ, creating an inflammatory environment seen in the synovial tissue of patients with post-treatment Lyme disease.[143] As far as their role in Bb clearance is concerned, follicular T cells rapidly differentiate B cells to antibody secreting plasma cells resulting in strongly-induced short-

lived antibodies instead of protective high-affinity long-lived antibodies. [70] However, their direct contribution in controlling Bb infection remains unclear.

In this study, we utilize several T cell and B cell deficient mouse models to answer two major questions:

1. What is the contribution of adaptive immunity in controlling Bb numbers at the site

of infection?

2. Are Bb-specific antibodies responsible for the significant decrease in Bb numbers

at the site of infection observed between day 8-14 post-infection?

3.1 Bb persistence in T cell-deficient mouse models

To evaluate the contribution of T cells in controlling Bb numbers at the site of infection we utilized 2 different T cell deficient mouse models. As discussed previously, a nude mouse is an athymic mouse which is devoid of T cells and NK cells. [157] We also used a TCR-/- mouse which is deficient in α/β and γ/δ T cells.[158] Groups of WT, nude and

TCR-/- mice were injected i.d with 105 WT-GFP Bb in the dorsal surface of the ear. Bb persistence was visualized at the indicated times post-injection using a confocal microscope and bacterial persistence was manually evaluated (as explained in section 2.3).

Sera was isolated from these mice via retro-orbital bleed at the indicated times post- infection and evaluated for quantitative assessment of Bb-specific antibody production by

ELISA.[72]

3.1.1 Bb persistence in nude mice

WT mice elicit a strong and potent Bb-specific antibody response.[51, 72] Antibody analysis of sera obtained from nude mice revealed a significant reduction in Bb-specific total immunoglobulin (Ig) and Immunoglobulin G (IgG) levels, as would be expected in the absence of T cell help (Figure 3 – 1 b, c). Basal levels of Bb-specific immunoglobulin

M (IgM) levels were observed in both WT and nude mice prior to infection. IgM levels increase until day 14 post-infection and appear to plateau thereafter. As expected, no significant difference in IgM levels between WT and nude mice was observed at all times assessed post-injection (Figure 3 – 1 d). Hence, T cell deficiency significantly reduces Bb- specific total Ig and IgG levels, but does not alter Bb-specific IgM levels.

Both WT and nude mice had similar Bb numbers at 6 hours post-infection. This confirms that the intradermal infection and bacterial dose enumeration (prior to infection) were executed correctly and involved no manual-error. As previously observed, Bb numbers in WT mice display an increase during the early phase of the infection reaching a peak around day 8 post-infection.[165] The numbers plummet thereafter and reach a basal level at day 14 post-infection. This basal level was maintained for the entire duration of the study. Bb persistence in a nude mouse followed a very similar trend as in WT mice with no significant difference until day 28 post-infection. Bb numbers in nude mice were significantly higher at day 28 post-infection but the numbers decrease to WT levels by day

60 (Figure 3 – 1 a).

Figure 3 – 1: Bb persistence and Bb-specific antibody production in WT and nude mice. (a) 1*105 WT-GFP Bb were injected i.d. into dorsal surface of mouse ear skin of both ears. 3D Z-stack images were collected at each timepoint. For bacterial cell enumeration, 2 images were captured in each of the 5 quadrants (10 images per ear) and Bb number is represented as the sum of total Bb present in all 10 images (0.045 mm3). Each point represents the average number of bacteria per viewing field. n ≥ 6 ears. At different times post-injection, Bb-infected WT and nude mice were bled, serum separated and (b) Bb-specific total Ig, (c) Bb-specific IgG and (d) Bb-specific IgM antibodies were measured using ELISA. Significance was determined by ANOVA Test followed by Tukey-Kramer Multiple Comparisons Test * p < 0.05, ** p < 0.01, *** p < 0.001 with respect to WT mice. n ≥ 3 mice. Error bars represent standard error of mean (SEM).

3.1.2 Bb persistence in TCR-/- mice

The antibody-production profiles in a TCR-/- mice was similar to the nude mouse.

In comparison to WT mice, TCR-/- mice displayed no significant difference in IgM levels

(Figure 3 – 2 d). However, total Ig and IgG levels in TCR-/- mice was significantly lower, as seen in a nude mouse (Figure 3 – 2 b, c).

Bb numbers in TCR-/- mice at 6 hours post-injection were also similar to WT mice confirming the accuracy of the intradermal injection and dose-enumeration. In comparison to WT mice, Bb numbers in a TCR-/- mouse were assessed to be significantly higher from day 1 post-injection. They remained significantly higher until they reach a similar peak levels at day 8 post-infection. Even after reaching the peak, Bb numbers in TCR-/- mice remain significantly higher until day 90 post-infection, demonstrating a significant delay in reaching a basal level in Bb numbers. However, the Bb numbers eventually reach WT basal levels after 90 days post-infection (Figure 3 – 2 a).

Over all, this data suggests that T cells appear to play a role in the kinetics of controlling Bb numbers. Bb numbers in both, nude and TCR-/- mice, persist significantly better and take longer to reach WT basal levels. However, it appears that T cells play a minimal role in controlling Bb numbers long-term, since the decrease in Bb numbers observed after day 8 post-infection in WT mice also occurs in nude and TCR-/- mice, albeit this drop is delayed. T cell deficiency did not alter the Bb-specific IgM response in mice.

While both T cell-deficient mouse models used in this study display almost a 2-log decrease in Bb-specific IgG levels, this defect does not appear to be essential for long-term

Bb control since Bb numbers in both nude and TCR-/- mice eventually reach WT basal level. 62

Figure 3 – 2: Bb persistence and Bb-specific antibody production in WT and TCR- /- mice. (a) 1*105 WT-GFP Bb were injected i.d. into dorsal surface of mouse ear skin of both ears. 3D Z-stack images were collected at each timepoint. For bacterial cell enumeration, 2 images were captured in each of the 5 quadrants (10 images per ear) and Bb number is represented as the sum of total Bb present in all 10 images (0.045 mm3). Each point represents the average number of bacteria per viewing field. n ≥ 6 ears. At different times post-injection, Bb-infected WT and TCR-/- mice were bled, serum separated and (b) Bb-specific total Ig, (c) Bb-specific IgG and (d) Bb-specific IgM antibodies were measured using ELISA. Significance was determined by ANOVA Test followed by Tukey-Kramer Multiple Comparisons Test * p < 0.05, ** p < 0.01, *** p < 0.001 with respect to WT mice. n ≥ 3 mice. Error bars represent standard error of mean (SEM).

3.2 Bb persistence in B cell-deficient mouse models

To assess the importance of B cells and the elicited Bb-specific antibody response in controlling Bb in skin tissues, we utilized two different B cell-deficient mouse models.

1) µMT mice are completely devoid of mature B cells, including plasma cells; hence, this mouse model should not elicit an antibody response. [156] Thus if Bb numbers decrease in this mouse model compared to WT mice, it would confirm that Bb-specific antibodies are not responsible for the significant decrease in spirochetal numbers. 2) An AID-/- mouse lacks the enzyme activation-induced cytidine deaminase (AID) protein. While this mouse possesses functional B cells, they are unable to undergo isotype switching since AID is essential for B cells to change the constant region of the antibody heavy chain, allowing it to produce different classes of highly pathogen-specific antibodies such as IgG. [155] Hence, this mouse model should be able to produce a Bb-specific IgM response, but would be unable to elicit a Bb-specific IgG response. Similar to the T cell deficient mouse study, groups of WT, µMT, and AID-/- mice were injected i.d with 105 WT-GFP Bb in the dorsal surface of the ear. Bb persistence was visualized at indicated times post-injection using a confocal microscope and bacterial persistence was manually evaluated (as explained in section 2.3). To confirm the antibody profile matched the expected outcome, sera isolated from these mice via retro-orbital bleed at indicated times post-infection was assessed for

Bb-specific antibody levels by ELISA

3.2.1 Bb persistence in µMT mice

Antibody analysis by ELISA confirmed that µMT mice were completely devoid of

Bb-specific antibodies (Figure 3 – 3 b, c, d). Bb persistence was observed to be significantly 64

Figure 3 – 3: Bb persistence and Bb-specific antibody production in WT and µMT mice. (a) 1*105 WT-GFP Bb were injected i.d. into dorsal surface of mouse ear skin of both ears. 3D Z-stack images were collected at each timepoint. For bacterial cell enumeration, 2 images were captured in each of the 5 quadrants (10 images per ear) and Bb number is represented as the sum of total Bb present in all 10 images (0.045 mm3). Each point represents the average number of bacteria per viewing field. n ≥ 6 ears. At different times post-injection, Bb-infected WT and µMT mice were bled, serum separated and (b) Bb-specific total Ig, (c) Bb-specific IgG and (d) Bb-specific IgM antibodies were measured using ELISA. Significance was determined by ANOVA Test followed by Tukey-Kramer Multiple Comparisons Test * p < 0.05, ** p < 0.01, *** p < 0.001 with respect to WT mice. n ≥ 3 mice. Error bars represent standard error of mean (SEM).

better in µMT mice in comparison to WT mice (Figure 3 – 3 a). µMT mice harbor significantly higher spirochetal loads right from day 4 post-infection and sustained higher levels even at day 150 post-infection. Between days 28 and 150 post-infection, Bb persists 65

at a basal level in µMT mice which is significantly higher than the basal level the spirochete maintains in WT mice. This clearly indicates that B cells directly or indirectly (through

Bb-specific antibodies) do play an important role in controlling Bb numbers at the site of infection. However, even in the complete absence of Bb specific antibodies, a decrease in

Bb numbers occurred after day 8 post-infection (peak Bb load), albeit the decrease is significantly delayed in µMT mice. While Bb numbers achieve basal level by day 14 post- infection in WT mice, Bb loads decrease to those levels only after day 150 post-infection in µMT mice. To summarize, while lack of B cells does allow enhanced Bb persistence at the site of infection, Bb-specific antibody responses do not appear to be responsible for the significant decrease in Bb numbers observed after day 8 post-infection.

3.2.2 Bb persistence in AID-/- mice

As predicted, antibody analysis of serum obtained from Bb-infected AID-/- mice confirmed the presence of Bb-specific IgM at levels similar to Bb-infected WT mice

(Figure 3 – 4 d). However, AID-/- mice were completely devoid of Bb-specific IgG (Figure

3 – 4 c Lack of IgG also resulted in significantly lower levels of Bb-specific total Ig levels in AID-/- mice (Figure 3 – 4 b).

Bb persistence in AID-/- mice overall appears similar to Bb persistence in WT mice.

AID-/- mice achieve similar spirochetal peak load as observed in WT mice at day 8 post- infection. While significantly higher Bb numbers were observed at 6 hours and day 28 post-infection, Bb numbers decline even in AID-/- mice, eventually reaching basal bacterial levels as seen in WT mice (Figure 3-4 a). Thus, Bb-specific IgG response does not contribute to the overall decline in Bb numbers observed after day 8 post-infection. 66

Figure 3 – 4: Bb persistence and Bb-specific antibody production in WT and AID- /- mice. (a) 1*105 WT-GFP Bb were injected i.d. into dorsal surface of mouse ear skin of both ears. 3D Z-stack images were collected at each timepoint. For bacterial cell enumeration, 2 images were captured in each of the 5 quadrants (10 images per ear) and Bb number is represented as the sum of total Bb present in all 10 images (0.045 mm3). Each point represents the average number of bacteria per viewing field. n ≥ 6 ears. At different times post-injection, Bb-infected WT and AID-/- mice were bled, serum separated and (b) Bb-specific total Ig, (c) Bb-specific IgG and (d) Bb-specific IgM antibodies were measured using ELISA. Significance was determined by ANOVA Test followed by Tukey-Kramer Multiple Comparisons Test * p < 0.05, ** p < 0.01, *** p < 0.001 with respect to WT mice. n ≥ 3 mice. Error bars represent standard error of mean (SEM).

Comparison of Bb persistence between µMT mice and AID-/- mice also identified the important of an early Bb-specific IgM response. While µMT mice are completely devoid of B cells and Bb-specific antibodies, AID-/- mice still possess B cells and are able to generate an Bb-specific IgM response. Bb persist at higher levels in µMT mice, with Bb numbers significantly higher in these mice even at day 150 post-infection. In contrast, Bb levels in AID-/- mice reach levels similar to WT mice after day 28 post-infection. Hence, the presence of B cells and/or early IgM responses appear to play a significant role in controlling Bb numbers at the site of infection. Contrarily, a Bb-specific IgG appears to play a minimal role in controlling Bb numbers at the site of infection.

Chapter 4

Role of Bb-elicited IL-10 in suppression of innate immunity

In chapter 3, we examined the contribution of adaptive immunity in controlling Bb persistence at the site of infection. In this section, we will shift our focus to assess the interaction the Lyme disease spirochete has with the innate immune responses. After tick- mediated transfer, Bb is initially located in the dermis, where it replicates and host-adapts by changing its outer surface protein repertoire, a mechanism it uses to evade the developing adaptive immune response. The innate immunity plays a major role at this stage in controlling Bb numbers in the skin.[98] The resident immune cells and infiltrating granulocytes utilize various pathogen recognition receptors (PRRs) on their surface to recognize different pathogen associated molecular patterns (PAMPs), thus eliciting an inflammatory response. However, in spite of eliciting a robust inflammatory response, the innate immune responses are unable to control this extracellular pathogen and Bb disseminates to multiple disparate tissues. Hence, there is some dysregulation occurring in vivo which allows the pathogen to evade this immune response. In this section, we will examine various aspects which contribute to this dysregulation and allow this spirochete to cause a persistent infection.

Previous studies from our lab has investigated the innate immune response against

Bb. Unlike most pathogens, Bb does not express LPS or lipoteichoic acid moieties on its outer surface. Instead Bb codes for ≥127 different putative lipoproteins, which all possess similar triacyl modifications on their N-terminal cysteine residues. [11, 99] These modifications can be directly recognized by various innate immune cells via Toll-like

Receptor 2 (TLR2). [100, 101] Injection of these purified lipoproteins into joints or skin elicits an acute inflammatory response predominated by neutrophils, confirming that lipoproteins play a major role in inflammatory induction in vivo.[166] Macrophages derived from TLR2-

/- mice demonstrated significantly lower production of proinflammatory products, such as

IL-6, NO, and TNF-α in response to B. burgdorferi sonicate and purified Bb outer surface protein A (OspA).[66] This study identified TLR2 as a key receptor used by resident immune cells for Bb recognition. The impact of TLR2 on host defense to Bb is evident in vivo as well, as TLR2-/- mice harbor almost 100-fold higher Bb in all target tissues up to 4 weeks post-infection; this was the longest time point assessed.[66] In this section, we aim to visualize the immune processes affected by loss of TLR2 in vivo, which allow for this enhanced Bb persistence. We hypothesize that loss of TLR2 in vivo prevents Bb recognition by host immune cells, severely attenuating multiple resident immune cell activities. The diminished activation of macrophages/ DCs results in creation of a hypoactive immune environment which allows for enhanced Bb persistence.

While TLR2 is essential for Bb recognition, it still remains unclear what constitutes an effective immune response against Bb. The host generates a Bb-specific antibody response, but these antibodies are unable to prevent a persistent infection. While passive transfer of these antibodies can protect a naïve mouse, it cannot confer protection against 70

host-adapted spirochetes.[135] T cells are important for development of a Bb-specific antibody response, however, passive transfer of purified T cells derived from immunized mice to naive mice were also unable to confer protection.[132, 167] Based on this data and the evidence provided in Chapter 3 of this study, we are confident to say that the adaptive immunity plays a minimal role in controlling Bb persistence. Transgenic mice with modified host cytokine profiles, such as IL-4, IFN-γ, and IL-12, which normally display dramatic effect on bacterial clearance, had no effect on Bb clearance.[140, 142, 168] Similar results were obtained by using mice deficient in NO or ROS generation.[169] Thus, normally vital components of phagocyte-mediated defenses appear to have minimal contribution in controlling Bb infection in vivo.

A previous study comparing differences in Lyme pathology within C3H and

C57BL/6 (BL/6) mice identified that macrophages derived from naive C3H mice produced significantly higher levels of the inflammatory mediators, TNF-α, IL-6, and NO compared to B6 macrophages. The diminished levels of these proinflammatory cytokine production by B6 macrophages was inversely correlated with the production of high levels of the anti- inflammatory cytokine IL-10.[129] This was the first indication that IL-10 might be associated in attenuation of a pro-inflammatory response surmounted against Bb infection.

Since then, several studies have investigated the role of IL-10 in Bb infection. Bb elicits a rapid and robust IL-10 response in vitro, within 4-6h after co-culture with macrophages.[69]

A significant increase in IL-10 transcript levels is observed in skin tissues within 24h post- infection as well and IL-10-/- mice display ~10-fold decrease in Bb persistence.[68] This makes IL-10 the only known cytokine to induce such a significant effect on Bb clearance.

In vitro studies from our lab revealed that Bb-elicited IL-10 can diminish Bb 71

uptake/trafficking by MØs and suppress ROS production. This was accompanied by decreased production of several Lyme-associated pro-inflammatory cytokines, such as IL-

6, IL-12, and TNFα and chemokines like CXCL1, CXCL2, CCL3 and CCL2.[69]

We hypothesized that Bb-elicited IL-10 creates a dysregulated innate immune response in vivo, which allows for better persistence of the spirochete as the site of infection. In this study, we aim to visualize the effects of IL-10 on the innate immune responses and delineate how it alters innate immune functions in vivo. Based on this hypothesis, loss of IL-10 should enhance immune cell activity, creating a hyperactive immune environment, allowing for better clearance of the pathogen.

4.1 Visualization of Bb-elicited IL-10 using the tiger mouse model

Previous studies from our lab identified that Bb-elicits a rapid IL-10 response in vitro, when co-cultured with macrophages or dendritic cells. Similarly, mouse infection with low-passage Bb identified a significant increase in IL-10 transcript levels in skin tissues within 24h post-infection. To visualize and identify the cell types involved in this premature IL-10 response and develop a kinetic profile associated with IL-10 production in response to Bb infection, we utilized an IL-10 reporter mouse (tiger mouse). [159] To visualize the Bb-elicited IL-10 response, groups of tiger mice were injected with 105 low passage WT Bb intradermally into the dorsal surface of the ear skin. At indicated times post-infection, these mice were anesthetized and IVM was used to quantify GFP- expressing cells. Since the gfp gene is fused to the IL-10 promoter, a green fluorescing cell represents a cell responding to Bb infection by producing IL-10. Z-stack images using a

confocal microscope allowed for quantification of IL-10-producing cells, whereas magnified images were captured to assess IL-10-producing cell morphology.

4.1.1 Identification of cell types associated with IL-10 production in response to Bb

infection within the skin

Intravital assessment of tiger mice identified several cell types associated with IL-

10 production in response to Bb infection. Based on morphological assessments,

Langerhans cells were visualized to be a primary GFP-expressing (i.e. IL-10-producing) cells at 6 hours post-infection (Figure 4 – 1 a). At 12 hours post-infection, Langerhans cell were still the dominant IL-10-producing cell type, however some macrophages/ dermal

DCs are first seen fluorescing green (Figure 4 – 1 b), suggesting that macrophages/ dermal

DCs also produce IL-10. While Langerhans cells were observed fluorescing green at day

2 post-infection, their numbers appear to be dramatically reduced compared to earlier times

(Figure 4 – 1 c). Alternatively, macrophages/ dermal DCs appear to be the primary IL-10 producers in skin tissue at this time. Also, GFP-expressing cells which are smaller in size and spherical in shape were first observed at this time, which appear to be neutrophils responding to Bb infection. The small cells with spherical morphology persist even at day

4 post-infection, however, macrophages/ dermal DCs appear to be the predominant IL-10 producing cell types (Figure 4 – 1 d). At day 8 post-infection (Figure 4 – 1 e), we could only identify macrophages/ dermal DCs as the sole IL-10-producing cell type and similar observation were made even on day 21 post-infection (Figure 4 – 1 f). To summarize, while

Langerhans cells appear to the primary cells associated with IL-10 production at the earliest

times post-infection, macrophages/ dermal DCs were identified as the major source of Bb- elicited IL-10 within skin tissues at all time points ≥ 24h post-infection.

Figure 4 – 1: Cell types associated with IL-10 production in response to Bb infection. 1*105 WT Bb were injected i.d. into both ears of 3 tiger mice. At indicated time post-infection, magnified 3-D Z-stack images were captured for morphology assessment. Red boxes highlight Langerhans Cells; Blue boxes highlight neutrophils; Yellow boxes highlight MØs / dermal DCs. (a) 6 hours post-infection (b) 12 hours post- infection (c) Day 2 post-infection (d) Day 4 post-infection (e) Day 8 post-infection (f) Day 21 post-infection.

4.1.2 Kinetics associated with IL-10 production in response to Bb infection within

the skin

IVM of Bb-infected tiger mice was used to quantitatively assess the appearance of

IL-10-producing cells in Bb-infected skin. Bb-infected mice were anesthetized at indicated times post-infection and 3-D Z-stack images were procured for manual quantification of

GFP+ cells. Using this technique, a temporal profile describing the kinetics of an IL-10 response to Bb infection was evaluated. A gradual increase in GFP+ cells was observed during the first 24h post-infection. However, a decline in GFP+ cell numbers was observed after 24 hours, reaching almost basal level by day 4 post-infection (Figure 4 – 2 a). Notably, the vast majority of GFP+ cells were observed in the central quadrant (injection site) (Figure

4 – 2 b). Previous studies indicate that Bb are confined to the central quadrant for 2 – 4 days post-infection, suggesting that only the cells in close proximity of the resident spirochetes are secreting IL-10. [165] At day 7 post-infection, a gradual increase in GFP+ cells is again observed, with GFP+ cell number peaking at day 14 post-infection. Post day

14, GFP+ cell numbers undergo a gradual decline. Although some GFP+ cells are still present in the central quadrant, the majority of GFP+ cells observed between 7 – 14 days post-infection are present in the surrounding quadrants. This corresponds with the previously observed dissemination of Bb from the central quadrant of the ear to the surrounding quadrants, which begins after day 4 post-infection complete spread to all quadrants is observed by day 7 post-infection. To summarize, a rapid IL-10 response is observed within 24 hours of Bb infection at the injection site (central quadrant). An IL-10 response in the surrounding quadrant is elicited only after Bb has disseminated to those

regions. Bb- elicited IL-10 responses in ear tissues peaks at day 14 and is still notable at day 28 post-infection.

Figure 4 – 2: Kinetics associated with IL-10 production in response to Bb infection. 1*105 WT Bb or BSK was injected i.d. into both ears of 3 tiger mice. At indicated time post-infection, 3-D Z-stack images were captured for quantification of GFP+ cells. Cells were quantified by manual counting. 10 images were captured for each ear and the number of GFP+ cells in each ear is the summation of all GFP+ cells in 10 images combined. (n ≥ 6 ears) (a) GFP+ cells present in tiger mice infected with WT Bb or BSK (b) Segregation of GFP+ cells present in central quadrant and surrounding quadrants (superior, inferior, proximal and distal quadrant; depicted in insert image of ear). Error bars represent standard error of mean (SEM).

4.2 WT-GFP Bb persistence in WT, TLR2-/- and IL-10-/- mice

Previous studies from our lab have reported that IL-10-/- mice display a 10-fold greater clearance of Bb up to 4 weeks post-infection compared to WT mice [68], this is vastly different from TLR2-/- mice which are defective in Bb clearance and harbor significantly higher Bb burdens in its tissues.[66] To more definitively compare Bb persistence in these mice at the site of infection, WT, TLR2-/- and IL-10-/- mice were intradermally injected with 105 WT-GFP Bb into the dorsal surface of the ear. At indicated times post-infection, these mice were anesthetized, and Bb persistence at the site of infection was visualized using IVM. The bacterial numbers were enumerated by manually counting the bacteria in each image. WT-GFP Bb infection of WT mice generated a similar

Bb persistence curve as reported by Dr. Sekar during her studies. [51, 165] WT-GFP Bb were confined to the initial injection site for the first 48 hours of the infection. At day 4 post- infection, WT-GFP Bb are first observed in all quadrants of the ear. Rapid proliferation of

WT-GFP Bb is first observed by day 2 post-infection and peak spirochetal levels are achieved at day 8 post-infection (Figure 4 – 3). Subsequently, Bb levels steadily decline until day 14-28 post-infection, after which WT-GFP Bb attain a basal level. Spirochetal persistence in the ear was assessed for 6 months post-infection and no significant change

in Bb numbers was observed after the basal level was achieved at day 14-28 post-infection.

This is consistent with previous observations made by Dr. Sekar [165], where Bb was observed to persist at basal levels for more than 2 years post-infection.

IVM assessment of WT-GFP Bb persistence in IL-10-/- mice revealed a few interesting facts. At 6 hours post-infection, similar Bb loads were observed in WT and IL-

10-/- mice. However, by day 2 IL-10-/- display significantly better clearance of WT-GFP Bb and harbor approximately 10-fold lower Bb burden than WT mice (Figure 4 – 3). While peak spirochetal load in IL-10-/- mice is achieved at the same time as WT mice (day 8 post- infection), peak Bb numbers in IL-10-/- mice were 1-log (10-fold) lower compared to WT mice. Bb loads remained significantly lower through day 10 post-infection, but then reached a similar basal level as observed in WT mice by day 14. Bb persistence in IL-10-/- mice could only be evaluated until day 60 post-infection, since IL-10-/- mice are extremely vulnerable to rectal prolapse and gastrointestinal inflammation, making accurate long-term assessment and survival of IL-10-/- mice extremely difficult. To summarize, IL-10-/- mice display significantly higher Bb clearance up to 2 weeks post-infection. This bolsters our previous findings of enhanced Bb clearance in IL-10-/- mice. [68] However, this enhanced clearance of Bb observed in IL-10-/- mice is short-lived, since WT-GFP Bb achieve a similar basal level in IL-10-/- mice as observed in WT mice. Since the enhanced clearance is only observed during the initial acute phase of the infection (2 weeks post-infection), it is suggestive of the fact that the innate immune response could be responsible for the observed lower Bb persistence in IL-10-/- mice. This is also supported by previous publications from the lab which have attributed innate immunity for enhanced Bb clearance in IL-10 deficient conditions. [67, 68] 78

TLR2-/- mice have been reported to harbor significantly higher spirochetal loads than WT mice even at 8 weeks post-infection.[66] IVM assessment of WT-GFP Bb persistence at the site of infection in TLR2-/- mice was performed to better understand the infection kinetics and persistence in this model. TLR2-/- mice display significantly higher

Bb numbers by 6 hours post-infection (Figure 4 – 3). Bb levels remain significantly higher at day 1 post-infection continue to rise and attain peak Bb levels at the same time as WT-

GFP Bb (day 8 post-infection). Interestingly, while Bb numbers plummet in WT mice after peak spirochetal infection on day 8, such a fall is not observed in TLR2-/- mice. TLR2-/- mice maintain significantly higher Bb loads through day 8 post-infection and continue to harbor significantly higher Bb loads even at 6 months post-infection, with almost 10-fold higher Bb numbers than in WT mice. Notably, this is the only reported mouse line that does not demonstrate a significant decrease in Bb numbers after day 8 post-infection and confirms that TLR2-/- mice are severely defective in Bb clearance, possibly due to defective

Bb recognition.

To rule out the contribution of adaptive immunity in enhanced clearance of Bb (IL-

10-/- mice) or defective Bb clearance (TLR2-/- mice), serum isolated from WT-GFP Bb infected WT, IL-10-/- and TLR2-/- mice was evaluated for Bb-specific antibody levels by

ELISA. Both IL-10-/- and TLR2-/- mice display similar levels of Bb-specific Total Ig

(Figure 4 – 4 a), Bb-specific IgG (Figure 4 – 4 b) Bb-specific IgM (Figure 4 – 4 c) levels.

This data supports that, while the adaptive immune response in both IL-10-/- and TLR2-/- mice remains intact, the variation in Bb persistence is introduced due to a defect in innate immune function.

The remainder of this chapter will investigate the effect of loss of IL-10 and TLR2 on innate immune cell function during Bb infection.

Figure 4 – 3: WT Bb persistence in WT, IL-10-/- and TLR2-/- mice. 1*105 WT-GFP Bb was injected i.d. into both ears of 3 WT, IL-10-/- and TLR2-/- mice. At indicated times post-infection, mice were anesthetized, and 3D Z-stack images were captured for spirochetal enumeration. 10 images were captured for each ear (total volume 0.045mm3) and the number of Bb present in each ear is the summation of all Bb present in 10 images combined. (n ≥ 6 ears) * p < 0.05; ** p < 0.01; *** p < 0.001 as compared to Bb number in WT mice. Man-Whittney U Test. Error bars represent standard error of mean (SEM).

Figure 4 – 4: Bb-specific antibody production in WT, IL-10-/- and TLR2-/- mice. 1*105 WT-GFP Bb were injected i.d. into dorsal surface of mouse ear skin of both ears. At different times post-injection, Bb-infected WT IL-10-/- and TLR2-/- mice were bled, serum separated and (a) Bb-specific total Ig, (b) Bb-specific IgG and (c) Bb-specific IgM antibodies were measured using ELISA. Significance was determined by ANOVA Test followed by Tukey-Kramer Multiple Comparisons Test * p < 0.05, ** p < 0.01, *** p < 0.001 with respect to WT mice. n ≥ 3 mice. Error bars represent standard error of mean (SEM). .

4.3 Neutrophil response to Bb infection in WT, TLR2-/- and IL-10-/- mice

Findings in Figure 4 – 3 confirm that loss of IL-10 enhances Bb clearance whereas loss of TLR2 results in enhanced Bb persistence. To better understand the mechanisms responsible for these infection phenotypes, LysM+ and Iaβ+ mice were backcrossed onto both an IL-10-/- and TLR2-/- backgrounds. To assess effects on neutrophil infiltration, groups of WT LysM+, IL-10-/- LysM+ and TLR2-/- LysM+ mice were infected with 105 WT-

GFP Bb intradermally into the dorsal surface of the ear. At indicated times post-infection, these mice were anesthetized and LysM+ cells present at the site of infection was visualized using confocal intravital microscopy. The total number of LysM+ cells present at the site of infection were assessed using the 3-D time-lapse images. Since the vast majority of

LysM+ cells observed in this mouse model are neutrophils[151], LysM+ cells will be denoted as neutrophils for the remainder of this chapter.

Similar to preliminary studies in our lab, a large neutrophil influx was observed by

6 hours post-infection and numbers continue to increase up to 12 hours post-infection, which was the peak for neutrophil numbers. Interestingly, a drastic decrease in neutrophil numbers are observed 12 hours post-infection and the numbers are significantly reduced by day 4. By day 14 post-infection, neutrophil numbers reach normal basal levels and very few neutrophils are observed thereafter through day 60 post-infection (Figure 4 – 5).

Notably, this decrease in neutrophil numbers (starting at 24 hours post-infection) occurs at the same time when Bb numbers are continuing to increase exponentially (Figure 4 – 3).

This data highlights the grave Bb-elicited immune dysregulation occurring in vivo which would feasibly allow increased Bb persistence and dissemination in host tissues.

To assess if the observed dysregulation in neutrophil infiltration was due to Bb- elicited IL-10, Bb infection was carried out in IL-10-/- LysM+ mice and neutrophil numbers at various times post-infection were evaluated. In an IL-10 deficient environment, the neutrophil influx is similar to WT LysM+ mice at 6 hours post-infection, with peak numbers again achieved at 12 hours post-infection. However, loss of IL-10 lessens the significant decrease in neutrophil numbers observed in WT mice, and neutrophil numbers persist at significantly higher levels compared to WT mice even at 24 hours post-infection. Post 24 hours, a gradual decrease in neutrophil levels is observed which is accompanied by significantly higher neutrophil numbers compared to WT mice through day 7 post- infection. Interestingly, neutrophil numbers in IL-10-/- LysM+ mice drop to WT LysM+ levels (Figure 4 – 5) by day 14 post-infection. Notably, day 14 was the same time at which 82

Bb numbers in IL-10-/- mice reach levels similar to WT mice (Figure 4 – 3). To summarize, loss of IL-10 allows for longer persistence of neutrophils at the site of infection. This enhanced neutrophil persistence could potentially be the reason for enhanced Bb clearance in IL-10-/- mice. However, this effect appears to have only acute activity, since neutrophil levels reach basal levels by day 14 post-infection.

Lastly, we investigated if defective Bb clearance in a TLR2-/- mouse was associated with any corresponding defect in neutrophil infiltration compared to WT mice. While similar neutrophil numbers were observed at 6 hours post-infection in TLR2-/- and WT mice, loss of TLR2 completely abolished the peak neutrophil influx observed at 12 hours post-infection. Neutrophil levels remain significantly lower in TLR2-/- LysM+ compared to

WT LysM+ mice at 12 hours post-infection and similar to WT LysM+ mice, neutrophil numbers achieve basal levels by day 14 post-infection (Figure 4 – 5). To summarize, while neutrophil infiltration is dysregulated in WT LysM+ mice, this dysregulation is further exaggerated in TLR2-/- LysM+ mice, most notably in the decreased peak neutrophil influx at 12 hours post-infection. This could potentially allow better Bb persistence during the acute phase of the infection.

Since neutrophils have to be recruited to the site of infection by resident antigen presenting cells (APCs), we next investigated the effect of IL-10 and TLR2 deficiency on

APCs in response to Bb at the site of infection.

Figure 4 – 5: LysM+ cell infiltration in WT, IL-10-/- and TLR2-/- mice. 1*105 WT- GFP Bb was injected i.d. into both ears of 3 WT LysM+, IL-10-/- LysM+ and TLR2-/- LysM+ mice. At indicated times post-infection, mice were anesthetized, and 3D Z-stack images were captured for LysM+ cell enumeration. (n ≥ 6 ears) * p < 0.05; ** p < 0.01; *** p < 0.001 as compared to LysM+ cell number in WT mice. Man-Whittney U Test. Error bars represent standard error of mean (SEM).

4.4 Resident immune cell activity in WT, TLR2-/- and IL-10-/- mice during Bb

infection

Neutrophils have been implicated to play a major role in Bb clearance at the initial

site of infection.[117] Assessment of neutrophil infiltration in the previous section

identified a major dysregulation in neutrophil infiltration at the site of infection, which

was at least partially induced by Bb-elicited IL-10 (Figure 4 – 5). However, neutrophils

are recruited to the site of infection by active resident immune cells. To identify any

defect in resident immune cell activity caused by Bb-elicited IL-10, WT Iaβ+, IL-10-/-

Iaβ+ and TLR2-/- Iaβ+ mice were intradermally infected with 105 WT-GFP Bb in the dorsal surface of the ear. At indicated times post-infection, these mice were anesthetized and Iaβ+ cells present at the site of infection was visualized using confocal intravital microscopy. 3-D images and XYZ time-lapse videos of the infection site allowed us to determine Iaβ+ cell number, cell diameter, cell area, and cell velocity at each time point.

Figure 4 – 6: Iaβ+ cell numbers in WT, IL-10-/- and TLR2-/- mice. 1*105 WT-GFP Bb was injected i.d. into both ears of 3 WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice. At indicated times post-infection, mice were anesthetized, and 3D Z-stack images were captured for Iaβ+ cell enumeration. (n ≥ 6 ears) * p < 0.05; ** p < 0.01; *** p < 0.001 as compared to Iaβ+ cell number in WT mice. Man-Whittney U Test. Error bars represent standard error of mean (SEM).

4.4.1 Iaβ+ cell numbers in response to Bb infection

Uninfected WT Iaβ+, TLR2-/- Iaβ+ and IL-10-/- Iaβ+ mice display similar basal numbers of Iaβ+ cells within ear skin tissue. Basal level of Iaβ+ cells is present at the site of infection in WT Iaβ+ mice infected with WT-GFP Bb at 6 hours post-infection. Iaβ+ cell numbers then gradually increase and achieve peak level by day 14 post-infection; these levels are maintained through at least day 60 post-infection (Figure 4 – 6). In contrast, while basal levels of Iaβ+ cells are present at the site of infection in TLR2-/- Iaβ+ mice at 6 hours post-infection, no increase in Iaβ+ cell number is observed in these mice. Iaβ+ cells are maintained at their basal level throughout day 60 post-infection. While interesting, this result was not surprising since TLR2 is vital for detection of Bb via its outer surface lipoproteins and loss of TLR2 suppresses innate immune responses in Bb recognition.

Thus, TLR2 deficiency prevents an increase in Iaβ+ cell numbers at the site of infection, suggestive of decreased resident immune cell activity and potentially contributes to the defective neutrophil infiltration observed in TLR2-/- LysM+ mice, since active resident immune cells are vital for neutrophil recruitment to the site of infection. Surprisingly, Iaβ+ cell number in IL-10-/- Iaβ+ mice were similar to WT Iaβ+ mice for most time points assessed. While significantly higher Iaβ+ cells were observed in IL-10-/- Iaβ+ at day 10 and

60 post-infection, day 14 post-infection harbored significantly lower number of Iaβ+ cells.

Due to these fluctuations, no clear trend could be established, and we were compelled to conclude that we could not observe any clear differences between Iaβ+ cell numbers present in WT-/- Iaβ+ and IL-10-/- Iaβ+ mice. Based on these findings, we hypothesized that IL-10 could potentially be altering Iaβ+ cell activity rather than the actual numbers. Hence, we

assessed various parameters to compare Iaβ+ cell activity in WT Iaβ+, IL-10-/- Iaβ+ and

TLR2-/- Iaβ+ mice, which will be discussed next.

4.4.2 Iaβ+ cell size in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice during Bb infection

To determine the activation state of Iaβ+ cells, various published morphological characteristics were analyzed. Previous research has used macrophage and dendritic cell diameter as a physical measure of cell activity; the larger the cell diameter, the higher the cell activity.[170] As explained in Section 2.3, 2-D XY images at a particular depth using

IVM were captured at 0 hours, 6 hours, day 6 and day 14 post-infection. Using Image J software, Iaβ+ cell diameter in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice was determined at the indicated time points. Uninfected WT Iaβ+, TLR2-/- Iaβ+ and IL-10-/- Iaβ+ mice possess

Iaβ+ cells of similar diameter (15.54 µm, 13.46 µm and 12.18 µm respectively). At 6 hours post-infection, Iaβ+ cells in both WT Iaβ+ and TLR2-/- Iaβ+ mice have similar cell diameter

(16.34 µm and 13.17 µm respectively), whereas Iaβ+ cells in IL-10-/- Iaβ+ mice have significantly larger cell diameter (27.15 µm); i.e. ~ 1.5-fold larger (Figure 4 – 7 a). At day

6 post-infection, Iaβ+ cells in IL-10-/- Iaβ+ mice maintain their larger diameter (31.02 µm) while Iaβ+ cells in WT Iaβ+ mice undergo ~ 1.5-fold increment in diameter length (29.51

µm), achieving similar diameter length as Iaβ+ cells in IL-10-/- Iaβ+ mice. As expected, Iaβ+ cells in TLR2-/- Iaβ+ mice do not undergo any change in cell diameter (10.08 µm) and are significantly smaller than Iaβ+ cells in both WT Iaβ+ and IL-10-/- Iaβ+ mice, even at day 6 post-infection. This trend in cell activity continues through day 14 post-infection, where both WT Iaβ+ and IL-10-/- Iaβ+ cells maintain larger cell diameter (21.89 µm and 16.14 µm respectively) while TLR2-/- Iaβ+ cells remain unaltered from 6 hours post-infection (13.30 87

µm) and significantly smaller than both WT Iaβ+ and IL-10-/- Iaβ+ cells. This data suggests that, while WT Iaβ+ cells take about 6 days post-infection to activate in response to Bb infection, loss of Bb-elicited IL-10 allows these cells to activate much faster, starting as early as 6 hours post-infection. In contrast, TLR2-/- Iaβ+ cells are unable to activate in response to Bb infection, presumably due to defective Bb recognition in the absence of

TLR2.

To verify that cell diameter accurately represented a measure of cell size, we also evaluated cell area as a measure of cell size. Individual Iaβ+ cells present in 2-D XY images captured using IVM at 0 hours, 6 hours, 6 day and day 14 post-infection were measured for their area using Image J. No difference in cell area was determined between Iaβ+ cells in uninfected WT Iaβ+, TLR2-/- Iaβ+ and IL-10-/- Iaβ+ mice (Figure 4 – 7 b). Iaβ+ cell area analysis revealed a similar trend to the cell diameter study where IL-10-/- Iaβ+ cells were significantly larger than WT Iaβ+ and TLR2-/- Iaβ+ cells at 6 hours post-infection. By day 6 post-infection, WT Iaβ+ undergo an increase in cell area reaching similar levels to IL-10-/-

Iaβ+ cells and these levels are maintained even at day 14 post-infection. On the other hand,

TLR2-/- Iaβ+ cells do not undergo any change in cell area (similar to cell diameter) even at day 14 post-infection; an indication of their inactive state.

To prove that the increase in cell diameter correlates to increased cell area, individual Iaβ+ cell areas were graphed against corresponding Iaβ+ cell diameter of each cell. Using this graph, a Pearson Correlation Coefficient was determined and the Pearson

R-value of 0.9502 was calculated, indicating a strong positive correlation between cell area and cell diameter (Figure 4 – 7 c).

Figure 4 – 7: Iaβ+ cell size in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice during Bb infection. 1*105 WT-GFP Bb was injected i.d. into both ears of 3 WT Iaβ+, IL-10- /- Iaβ+ and TLR2-/- Iaβ+ mice. At indicated times post-infection, mice were anesthetized, and 2D XY images were captured for Iaβ+ cell size assessment. (n ≥ 6 ears; each point represents individual cell assessed; n ≥ 10 cells; at least one cell from each ear) * p < 0.05; ** p < 0.01; *** p < 0.001 Kruskal–Wallis test followed by Dunn’s Multiple Comparison Test as a post-hoc procedure. (a) Iaβ+ cell diameter in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice (b) Iaβ+ cell area in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice (c) scatter plot for correlation analysis between Iaβ+ cell diameter and Iaβ+ cell area. Error bars represent standard error of mean (SEM).

4.4.3 Dynamic behavior of Iaβ+ cells in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice

during Bb infection

Iaβ+ cell activity was further assessed by evaluating the dynamic behavior of these cells in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice. Iaβ+ cell motility was analyzed by calculating 2 dynamic parameters; mean cell velocity and cell confinement ratio. The mean velocity represented the moving speed of Iaβ+ cells whereas the confinement ratio was defined as a ratio of the displacement of an Iaβ+ cells to the corresponding total length traveled by that cell in an X-Y plane. No significant difference in mean cell velocity and cell confinement ratio was observed between uninfected Iaβ+ cells in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice (Figure 4 – 8 a). The mean velocity of Iaβ+ cells revealed that IL-10-

/- Iaβ+ cells move at 13.7 µm/min at 6 hours post-infection, which is significantly faster than

WT Iaβ+ (5 µm/min) and TLR2-/- Iaβ+ cells (7.5 µm/min). However, these IL-10-/- Iaβ+ cells do not maintain this high velocity, with velocity dropping to 7.4 µm/min at day 6 post- infection and 6.3 µm/min at day 14 post-infection. Alternatively, we do not observe such drastic fluctuations in WT Iaβ+ and TLR2-/- Iaβ+ cell mean velocity, and the mean velocity for these cells ranges from 5 µm/min – 7.5 µm/min at every assessed time-point.

2D XY time-lapse videos of the infection site allowed us to calculate displacement and total travel length for each Iaβ+ cell observed. The confinement ratio for each Iaβ+ cell was assessed using the displacement and length values. A high confinement ratio means the cells are more actively probing their surroundings, which represent higher cell activity.

IL-10-/- Iaβ+ cells exhibited a higher confinement ratio at each assessed time point compared to WT and TLR2-/- cells (Figure 4 – 8 b). Confinement ratio assessment of WT Iaβ+ cells

again revealed a major delay in activation of these cells (similar to delay in cell diameter;

Figure 4 – 7 a). WT Iaβ+ cells exhibit significantly lower confinement ratio at 6 hours and day post infection as compared to IL-10-/- Iaβ+ cells, and reach similar levels to IL-10-/- Iaβ+ cells only by day 14 post-infection. Alternatively, TLR2-/- Iaβ+ cells do not exhibit any change in confinement ratio at any assessed time point and remain significantly lower than

IL-10-/- Iaβ+ cells at all assessed time points. To summarize, loss of IL-10 enhances Iaβ+ cell activity, allowing for increased surveillance of the surrounding tissue by these resident immune cells. WT Iaβ+ cells display a significant lag in activation and achieve confinement ratio similar to IL-10-/- Iaβ+ cells only at day 14 post-infection. In contrast, TLR2 deficient

Iaβ+ cells never exhibit any increase in confinement ratio which indicates a severe defect in activation suggesting that these cells are unable to probe for the spirochetal pathogen in its surrounding. The visual evidence provided in this chapter strongly suggests that Bb- elicited IL-10 causes delayed activation of WT immune cells at the site of infection, diminishing innate cell activity early during infection while Bb undergoes host adaption at the site of infection. Loss of Bb-elicited IL-10 appears to create a “hyperactive immune environment” which enables enhanced Bb clearance. In contrast, loss of TLR2 cripples innate immune responses, disrupting Bb recognition by resident immune cells. Defective recognition hinders immune cell activation, creating a “hypoactive immune environment” which allows for enhanced Bb persistence in TLR2-/- mice.

Figure 4 – 8: Dynamic behavior of Iaβ+ cell in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice during Bb infection. 1*105 WT-GFP Bb was injected i.d. into both ears of 3 WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice. At indicated times post-infection, mice were anesthetized, and time-lapsed 3D XYZ images were captured to assess Iaβ+ cell velocity and confinement ratio. (n ≥ 6 ears; each point represents individual cell assessed; n ≥ 10 cells; at least one cell from each ear) * p < 0.05; ** p < 0.01; *** p < 0.001 Kruskal–Wallis test followed by Dunn’s Multiple Comparison Test as a post-hoc procedure. (a) Iaβ+ cell velocity in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice (b) Iaβ+ cell confinement ratio in WT Iaβ+, IL-10-/- Iaβ+ and TLR2-/- Iaβ+ mice. Error bars represent standard error of mean (SEM).

Chapter 5

Role of Bb velocity in evasion of innate immune response

Bb does not encode for any known virulence factors, like toxins and secretion systems. However, this spirochete invests 6% of its genome on its motility and chemotaxis apparatus, making it the largest group of genes essential for Bb infectivity in ticks and mice. [11] Bb utilizes its complex motor machinery not only to disseminate within vertebrate hosts, but also to translocate to and from an arthropod vector which is necessary for it to complete its enzootic lifecycle.

Dr. Lavik and Dr. Sekar from our lab began investigating Bb motility characteristics in vivo within murine skin using IVM. These studies showed that B. burgdorferi average velocity in host skin tissues is ≥ 200 μm/min [149] (Lavik et al, unpublished). This velocity of the bacteria is almost 40 times faster than any immune cell assessed in the skin.[31, 32] (Lavik et al unpublished). Thus, Bb velocity is considered as a major virulence factor for this pathogen which allows the spirochete to outrun the host immune response. Currently, any mutation which hampers Bb velocity has been shown to be detrimental for Bb survival in vivo. A ΔflaB Bb mutant that lacks flagella was non- motile and was found to be non-infectious, [28, 36] supporting the role of spirochetal motility for infectivity and virulence. Similarly, ΔmotB Bb which are devoid of the stator protein 94

MotB, are non-motile as well. ΔmotB Bb get cleared from the site of infection within 24 hours of infection and are unable to cause a persistent infection. [29] Importance of Bb velocity for infectivity was also observed while assessing Bb chemotactic mutants. ΔcheY3

Bb lacks the chemotactic response regulator CheY3. Loss of CheY3 prevents the spirochete from reversing its direction, therefore, ΔcheY3 Bb are unable to perform the characteristic back-and-forth motion in vivo. While ΔcheY3 Bb are capable of traversing at a similar velocity to WT Bb; in dense host tissue ΔcheY3 Bb supposedly get stuck since they cannot reverse their direction, resulting in loss of motility.[51] Hence, ΔcheY3 Bb while being a chemotactic mutant, also serves as a ‘pseudo-motility’ mutant despite having an intact motor-flagella assembly. These studies provide further support to the claim that motility is necessary for Bb persistence in vivo.

While it is clear that motility is absolutely necessary for Bb virulence in vivo, the significance of this high Bb velocity still remains uncertain. Why does Bb expend vast resources on maintaining such high velocity (≥ 200 μm/min)? Is Bb utilizing its velocity for immune evasion? If Bb is utilizing its velocity to outrun the host immune response, does it have to maintain a velocity 40x faster than the immune cells? This study will address these questions by utilizing Bb chemotactic and motility mutants as tools to study Bb persistence in vivo at variable spirochetal velocity.

To test this, ΔcheY3 Bb were used as a pseudo non-motile Bb mutant. As mentioned earlier (section 1.1.3.3), ΔcheY3 Bb lose their motility in vivo and get cleared by day 3-4 post-infection. We chose to use ΔcheY3 Bb instead of a true motility mutant (e.g. ΔmotB

Bb) because most motility mutants get cleared from the system by day 1 post-infection.

Since ΔcheY3 Bb persist at the site of infection relatively longer (3-4 days post-infection) 95

and still function as a non-motile control, ΔcheY3 Bb serves as a better tool to assess the role of Bb velocity on immune evasion.

Figure 5 – 1. Comparison of WT, ΔflaA, ΔcheY3 Bb velocity. 1*105 WT-GFP, ΔflaA-GFP or ΔcheY3-GFP Bb were injected i.d. into both ears of 3 WT mice. At day 8 post-infection, time lapsed 2-D images were captured for velocity assessment. The velocity of bacteria present in both ears of three mice was calculated using MetaMorph software. Each circle represents average velocity of individual bacteria analyzed. (n ≥ 25 cells) *** p < 0.001 as compared to WT Bb velocity. Kruskal-Wallis Test followed by Dunn's Multiple Comparisons Test. Error bars represent standard error of mean (SEM).

We also used a flagellar mutant of Bb. This mutant is devoid of the flagellar sheath protein FlaA, which has been speculated to provide flagellar rigidity around the hook, which is necessary for optimal flagellar rotation and motility. Studies performed with ΔflaA mutants of Brachyspira hyodysenteriae, which have unsheathed periplasmic flagella, demonstrated that these spirochetes are still motile, but move slower than wild-type.[37]

Based on this study, we assessed ΔflaA Bb motility patterns within murine ear tissues.

While WT Bb moved with an average velocity of 231.5 µm/ min, ΔflaA was significantly slower with an average velocity of 122.9 µm/ min, approximately half of WT Bb (Figure

5 – 1). The reduced velocity was not accompanied by any other noticeable motility defect.

Thus, ΔflaA Bb serves as a motility mutant that moves at a velocity half of WT Bb, but still retains its normal motility pattern regarding its characteristic back-and-forth motion in ear

푛표푟푚푎푙 푠푝푒푒푑 tissues. Hence, by using WT Bb (normal speed), ΔflaA Bb ( ) and ΔcheY3 Bb 2

(lose motile in vivo) we have 3 strains of Bb that traverse at 3 different velocities, providing the perfect tool to delineate the importance and relevance of not only Bb motility but also the significance of why Bb maintains such high speeds while in a mammalian host.

To more critically assess the importance of Bb velocity in evading innate immunity, WT Bb, ΔflaA Bb and ΔcheY3 Bb persistence was observed in WT, IL-10-/- and

TLR2-/- mice. As discussed in chapter 4, IL-10-/- mice display enhanced WT Bb clearance and serve as a hyperactive immune mouse model. TLR2-/- mice are defective in Bb recognition and display a significant defect in WT Bb clearance, thus acting as a hypoactive immune mouse model. Our hypothesis was, decreased Bb velocity would result in decreased Bb persistence in WT mice. Analysis of ΔcheY3 and ΔflaA Bb in IL-10-/- and

TLR2-/- mice would reveal how Bb velocity contributes to Bb persistence in hyperactive

(IL-10-/-) and hypoactive (TLR2-/-) immune conditions.

5.1 ΔcheY3-GFP Bb persistence in WT, TLR2-/- and IL-10-/- mice

ΔcheY3-GFP Bb persistence in WT mice has been investigated previously by Dr.

Sekar.[51] ΔcheY3-GFP Bb could perform translational motion but were unable to reverse their direction. These spirochetes were postulated to eventually get stuck in dense host tissues, resulting in loss of motility in vivo. Hence, while ΔcheY3-GFP Bb is a chemotactic mutant, its phenotype mimics a motility-deficient spirochete in vivo. As expected, ΔcheY3

Bb were unable to establish a persistent infection, however it took 4 days post-infection for the host immune system to completely clear this mutant spirochete.

In our study, we took advantage of the fact that ΔcheY3-GFP Bb eventually becomes non-motile in vivo and used it as a tool to assess the role of spirochetal motility in evading host immunity. To visualize the effect of cheY3 on Bb persistence, 105 ΔcheY3-

GFP Bb were intradermally injected into the dorsal surface of the ear and bacterial numbers were visualized using confocal microscopy. The bacterial numbers were enumerated manually by counting the bacteria in each image. The study revealed similar results as previously reported. While WT-GFP Bb number continue to increase exponentially between day 1 and day 8 post-infection in WT mice (Figure 5 – 2 a), ΔcheY3-GFP Bb numbers demonstrate a steady decline and no ΔcheY3-GFP spirochete was visible at day 4 post-infection (Figure 5 – 2 b). ΔcheY3-GFP Bb clearance was also confirmed by qPCR.

This indicates that the mutant spirochete gets cleared from the system by day 4 post-

infection. Thus, loss of cheY3 results in early clearance of Bb from mice, potentially because the spirochete losses its motility in vivo.

ΔcheY3-GFP Bb persistence was also visualized in IL-10-/- and TLR2-/- mice. As stated earlier, loss of IL-10 promotes Bb clearance and IL-10-/- mice harbor significantly lower WT Bb loads. Increased Bb clearance in IL-10-/- mice has been attributed to enhanced innate immune responses due to absence of IL-10 (chapter 4). [68] Hence this mouse model represents an in vivo environment with enhanced (hyperactive) immune activity. To visualize the effect of this enhanced immune activity on ΔcheY3-GFP Bb persistence, groups of IL-10-/- mice were intradermally injected with 105 ΔcheY3-GFP Bb into the dorsal surface of the ear. As expected, ΔcheY3-GFP Bb were cleared from an IL-10-/- mice more rapidly than WT mice, and no ΔcheY3-GFP Bb were observed by day 2 post-infection in an IL-10-/- mouse (Figure 5 – 2 b). ΔcheY3-GFP Bb clearance was also confirmed by qPCR. Thus, the enhanced immune activity (due to loss of IL-10) results in more rapid clearance ΔcheY3-GFP Bb compared to WT mice.

TLR2-/- mice are defective in Bb recognition and eventually harbor significantly higher WT Bb loads through at least 6 months post-infection, making them severely defective at Bb clearance (Figure 4 – 3). [65, 66, 106] Due to these defects, a TLR2-/- mouse serves as a in vivo model with diminished (hypoactive) immune activity. To assess if spirochetal velocity affects Bb persistence in this hypoactive immune environment of a

TLR2-/- mouse, groups of TLR2-/- mice were intradermally injected with 105 ΔcheY3-GFP

Bb into the dorsal surface of the ear and visualized using intravital confocal microscopy.

ΔcheY3-GFP Bb still persist in TLR2-/- mice until day 5 post-infection, which is significantly longer than in WT mice (Figure 5-2 b). ΔcheY3-GFP Bb clearance was also 99

confirmed by qPCR. These findings indicate that TLR2-/- immune cells can detect and clear

Bb if they lose their motility, although in a slower manner than WT cells.

Figure 5 – 2. WT and ΔcheY3 Bb persistence in WT, IL-10-/- and TLR2-/- mice. 1*105 WT-GFP, or ΔcheY3-GFP Bb were injected i.d. into both ears of WT, IL-10-/- and TLR2-/- mice. At indicated times post-infection, mice were anesthetized, and 3D Z-stack images were captured for spirochetal enumeration. 10 images were captured for each ear (total volume 0.045mm3) and the number of Bb present in each ear is the summation of all Bb present in 10 images combined. (n ≥ 6 ears) * p < 0.05; ** p < 0.01; *** p < 0.001 as compared to Bb number in WT mice. Man-Whittney U Test. (a) WT-GFP Bb persistence in WT, IL-10-/- and TLR2-/- mice (Taken from Figure 4 – 3 for comparison). (b) ΔcheY3-GFP Bb persistence in WT, IL-10-/- and TLR2-/- mice. Error bars represent standard error of mean (SEM).

5.2 ΔflaA-GFP Bb persistence in WT, TLR2-/- and IL-10-/- mice

In Figure 5 – 2, the importance of continuous spirochetal motility in establishing persistent infection was confirmed, even in a TLR2-/- mouse (which is defective in Bb recognition). To investigate the importance the velocities observed in WT Bb for establishing persistent infection, ΔflaA-GFP Bb which are significantly slower than WT

Bb (but are still 20 times faster than any immune cells in skin) were assessed for persistence

100

Figure 5 – 3. WT and ΔflaA Bb persistence in WT, IL-10-/- and TLR2-/- mice. 1*105 WT-GFP, or ΔflaA-GFP Bb were injected i.d. into both ears of WT, IL-10-/- and TLR2- /- mice. At indicated times post-infection, mice were anesthetized, and 3D Z-stack images were captured for spirochetal enumeration. 10 images were captured for each ear (total volume 0.045mm3) and the number of Bb present in each ear is the summation of all Bb present in 10 images combined. (n ≥ 6 ears) * p < 0.05; ** p < 0.01; *** p < 0.001 as compared to Bb number in WT mice. Man-Whittney U Test. (a) WT-GFP Bb persistence in WT, IL-10-/- and TLR2-/- mice (Taken from Figure 4 – 3 for comparison). (b) ΔflaA-GFP Bb persistence in WT, IL-10-/- and TLR2-/- mice. Error bars represent standard error of mean (SEM).

in mice. Assessment of ΔflaA-GFP Bb persistence revealed that (Figure 5 – 3 b), 1) ∆flaA

Bb display a longer lag period before Bb numbers reach a peak. While WT Bb achieve peak spirochetal loads by day 8 post-infection, ∆flaA Bb take 2-6 days longer and achieve peak spirochetal load by day 10-14 post-infection; 2) Peak numbers achieved by ∆flaA Bb are 5-fold lower than WT Bb, which indicate that despite being significantly faster than all host immune cells, ∆flaA Bb still gets cleared more efficiently than WT Bb; and 3) While

∆flaA Bb can cause a persistent infection, it maintains a basal level 2-fold lower than WT

Bb. Thus, even a two-fold reduction in Bb velocity results in increased clearance of the spirochete along with a reduction in spirochetal basal levels.

101

ΔflaA-GFP Bb persistence was also visualized in IL-10-/- and TLR2-/- mice. As stated earlier, IL-10-/- mice represent a hyperactive immune environment while TLR2-/- mice display a diminished immune response to Bb. ΔflaA-GFP Bb were cleared significantly better in IL-10-/- mice compared to WT mice. ΔflaA-GFP Bb persist at significantly lower levels beginning at day 6 post-infection and remain significantly lower even through 6 months post-infection. This result was expected since IL-10 deficiency does enhance macrophage activity and neutrophil infiltration at the sight of infection (Chapter

4). [68] Surprisingly, ΔflaA-GFP Bb persistence in a TLR2-/- mice was similar to their persistence in WT mice. Our initial hypothesis was ΔflaA-GFP Bb would persist significantly better in TLR2-/- mice as compared to WT mice since TLR2-/- mice are defective in Bb recognition. This hypothesis was based on the fact that WT-GFP Bb persist at significantly higher levels in TLR2-/- mice as compared to WT mice even at 6 months post-infection (Figure 4 – 3). However, this is not the case with ΔflaA-GFP Bb, as it observes a similar persistence curve in both TLR2-/- and WT mice. This suggests that a decrease in Bb velocity allows better recognition and clearance of Bb, and it is essential for Bb to maintain its high velocity (~ 230 µm/min) to persist at its optimal basal level.

102

Chapter 6

Discussion

6.1 Overview

Many microbial pathogens cause illness by secreting a toxin or directly inflicting cellular and tissue damage to the host. Unlike those diseases, Lyme disease pathology is not precipitated by its causative agent; instead it’s the host immune response elicited against the persistent pathogen which causes Lyme sequelae. Lyme disease (LD) is the most prevalent vector-borne disease in the United States. Caused by the spirochetal bacteria, Borrelia burgdorferi (Bb), the disease spreads between hosts by infected tick bites. On entering the skin, Bb elicits a strong innate response and a subsequent Bb-specific adaptive response. While these host immune responses are critical for controlling Bb, they are ineffective at eradicating the pathogen from the site of infection allowing Bb to cause a persistent and disseminated infection. In this dissertation project, we performed studies to visualize this immune response elicited against Bb by the host. The aim of this study was to assess the dysregulation in the host-elicited response which permits this extracellular pathogen to cause a persistent infection.

103

Previous in vitro studies revealed that both macrophages and dendritic cells co- incubated with Bb were capable of killing the spirochete.[67-69] Similarly, immune sera isolated from Bb-infected mice can confer protection against Bb to a naïve mouse, however, the serum loses its efficacy if passively transferred after day 5 post-infection, suggesting that it cannot confer protection against host-adapted spirochetes.[135] Thus, in vitro models to investigate Bb are unable to recapitulate the complex and dynamic in vivo host environment, precluding accurate assessment of the immune response generated against Bb. To counteract this problem, we developed an in vivo intravital microscopy

(IVM) technique which allows for assessment of the host – pathogen interactions within intact murine skin of a live mouse, in real-time. Not only can we visualize Bb persistence, dissemination and motility within intact murine skin, with the help of transgenic mice, we can quantify immune cell infiltration, persistence and assess physical parameters for immune cell activity. Hence, this technique allows us to obtain live in vivo snapshots of the immune response generated against Bb at various times post-infection.

Initial studies from our lab utilized IVM to assess Bb persistence at the site of infection. This study determined that Bb undergoes a massive proliferation between days

2 – 7 post-infection and peak bacterial numbers are achieved in skin tissues around day 8 post-infection. Attainment of peak bacterial numbers is followed by a significant decrease in Bb numbers until day 14 post-infection; subsequently, this low level is maintained for over 2 years.[165] Based on these infection kinetics in skin, the three primary objectives of this study were; 1) to identify the dysregulation in innate immunity which allows Bb to cause a persistent and disseminated infection, and determine whether Bb-elicited IL-10 plays a role in this dysregulation[68] ; 2) to assess whether adaptive immunity is responsible 104

for the significant decrease in Bb numbers after day 8 post-infection, since the decrease in

Bb numbers coincides with the appearance of Bb-specific antibodies; and 3) to determine the importance of WT Bb motility parameters/velocity in evading host immunity.

6.2 Role of Bb-elicited IL-10 on innate cell function

An IL-10 reporter (tiger) mouse was used to visualize and construct a kinetic profile of the IL-10 response to Bb infection. These studies identified Langerhans cells as the chief producers of IL-10 up to 12 hours post-infection. Langerhans cells are specialized dendritic cells present in skin epidermis.[171] This was a peculiar observation because Bb is not known to persist in the epidermis. Previous studies conducted by Dr. Lavik (unpublished data) provided extensive evidence that Bb is quickly confined within the dermal layer of skin ranging between 20 µm – 60 µm below the dorsal epidermal layer of ear skin.

Experiments in our current study involving IVM based assessment of Bb persistence confirms this data. There are at least two plausible explanations for this observation, 1) the tissue injury associated with the artificial system of needle inoculation. Since the needle breaks the epithelial barrier in order to introduce the inoculum to the infection site, it might be activating the epidermal Langerhans cells in doing so. Basal levels of IL-10 producing

Langerhans cells were observed in BSK-infected (control group) tiger mice as well, however at a much lower frequency; 2) the injection introduces the spirochete right at the dermal – epidermal interface. This would provide a transient interaction between the spirochete and Langerhans cells resulting in an IL-10 response. While it remains controversial whether Bb enters the dermis directly or traverses through the epidermis after tick bite, it is certain that Bb does not persist within the epidermal layer. This is also 105

supported by the fact that Langerhans cell display a very transient IL-10 response.

Subsequently, macrophages and dendritic cells predominate as IL-10 producing cells from day 2 through day 28 post-infection. This is consistent with previous in vitro studies performed in our lab which identified a rapid IL-10 response from macrophages and dendritic cells cocultured with Bb.[69] Sonderegger et al utilized the tiger mouse model to identify IL-10 producing cells in ankles during Bb infection and also identified macrophages as primary source of IL-10 at day 14 post-infection within ankle joints using flow cytometric analysis.[128]. These findings indicate that resident immune cells as primary source of IL-10 near the skin inoculation site.

IVM analysis of Bb-infected tiger mice allowed us to assess the kinetics associated with the appearance of IL-10 producing cells. The IL-10 response is localized to the central quadrant of the ear (injection site) and proliferates across the ear skin tissue as Bb disseminates to peripheral sections of the ear after day 4 post-infection. This indicates that only the cells in close proximity to Bb are responding to the infection by producing IL-10.

This is also consistent with previous work documented from our lab. Dr. Zhang assessed the signaling mechanisms associated with Bb-elicited IL-10 responses, and concluded that

IL-10 production required Bb interaction with the immune cell surface, but was phagocytosis-independent (unpublished).[172] These findings suggest that, as Bb disseminates through skin tissues, it elicits an IL-10 response from the resident immune cells within its proximity, extending an immune-suppressive environment which should allow for better evasion of the immune response.

To better understand innate immune cell dysregulation associated with Lyme disease, we utilized transgenic mice which allow visual quantification of the cellular 106

immune responses generated during Bb infection using IVM. Iaβ+ mice have an e-GFP moiety fused to an intact MHC class II β-chain of the the I–A β gene. Hence, every cell expressing MHCII molecule (primarily antigen presenting cells like macrophages and dendritic cells) will fluoresce green. The Iaβ+ mice were backcrossed on an IL-10-/- background (IL-10-/- Iaβ+ mice) to assess the effect of loss of IL-10 on resident immune cell function. These mice were also backcrossed on a TLR2-/- background. While most bacteria express lipopolysaccharides on their surface, Bb is unique and primarily expresses lipoproteins on its outer surface. These lipoproteins are highly immunogenic and TLR2 on immune cell surfaces are vital for Bb recognition. Hence, TLR2-/- Iaβ+ mice not only serve as a good negative control for these experiments, but also allows investigation of TLR2 effects on immune cell function which results in enhanced Bb persistence.

Iaβ+ cells display a gradual increase in skin-resident immune cell number in response to Bb infection, where they increased 5-fold during the first 8 weeks of Bb infection. Interestingly, IL-10-/- Iaβ+ cells were observed at similar numbers as WT Iaβ+ cells up to day 8 post-infection. Eventually, IL-10-/- mice possess significantly higher Iaβ+ cells at day 10 and 60 post-infection compared to WT mice; notably, the one outlier to this trend was at day 14 post-infection, where these numbers were lower than WT mice. The reason for these fluctuations observed in IL-10-/- Iaβ+ cells is currently not known. Previous work demonstrated that patients with antibiotic-responsive Lyme arthritis display higher levels of IL-4, which is reported to suppress macrophage activity.[173] Thus, a transient IL-

4 response could be inducing the inconsistent results in IL-10-/- mice. Regarding TLR2-/-

Iaβ+ cells, no increased in resident cell numbers were observed through 8 weeks post- infection and remained at very low levels. These data indicate that TLR2-mediated 107

signaling is required for resident immune cell activation in Bb-infected skin. Previous investigations displaying enhanced Bb persistence observed in TLR2-/- mouse tissues also support this theory. [66]

To accurately assess the effect of IL-10 and TLR2-deficiency on resident immune cell function during Bb infection, we evaluated cell diameter, cell area, average velocity and confinement ratios for Iaβ+ cells. Qiao et al in their study utilized these motility and morphological parameters to investigate the activation state of macrophages.[170] Using confocal microscopy, they determined that injection of melanoma cells resulted in an increase in mean macrophage diameter. The increase in diameter was accompanied with a

2.36-fold increase in velocity and significantly higher confinement ratio. Based on these assessments they were able to conclude that macrophages displayed persistent activation which enhanced brain metastasis. Increased macrophage diameter has also been observed in smokers and COPD patients as compared to non-smoking healthy controls. [174]

Confinement ratio has been used as a cell activation parameter for neutrophils, macrophages, and T cells in various pathological conditions.[175, 176] Based on these reports, these motility and morphological parameters were used to investigate the activation state of Iaβ+ cell during Bb infection.

IL-10-/- Iaβ+ cells display a significant increase in cell area and diameter at 6 hours post-infection whereas WT Iaβ+ cells only attain these levels at day 14 post-infection, suggesting a significant delay in Bb-activation of WT Iaβ+ cells. Confinement ratio is defined as the ratio of the displacement of a cell from its origin to the total length it has traveled. It is a direct measure of the probing activity of a cell where a higher confinement ratio represents a highly active cell. IL-10-/- Iaβ+ cells attain a significantly higher 108

confinement ratio by 6 hours post-infection and maintain this level even at day 14 post- infection. Alternatively, WT Iaβ+ cells again display a lag in activation and achieve confinement ratios similar to IL-10-/- Iaβ+ cells at day 14 post-infection. This is consistent with in vitro studies conducted by Chung et al.[69] APCs derived from IL-10-/- mice displayed a faster and more potent pro-inflammatory response than WT APCs when cocultured with Bb. The enhanced immune cell activity in IL-10-/- mice corresponds with significantly better Bb clearance. While Bb clearance was evaluated by IVM in our study,

Lazarus et al reported similar results by using qPCR assessment.[68] Together, these findings indicates that Bb-elicited IL-10 dampens the APC immune function resulting in a severe delay in cell activation which could allow Bb to more efficiently evade host cellular immunity at the site of infection. The enhanced activation of resident immune cells also provides an explanation why IL-10-/- mice display heightened Lyme pathology. Shin et al reported that patients experiencing antibiotic-refractory Lyme arthritis exhibit higher levels of proinflammatory cytokines synovial fluids and diminished IL-10 levels.[173] A consistent trend is observed even in C3H/HeN mice which are susceptible to Lyme disease and produce significantly lower levels of IL-10 in response to Bb as compared to Lyme- resistant C57BL/6 mice. Lower IL-10 levels are accompanied with increased ankle swelling during Lyme infection and higher levels of inflammatory mediators like, TNFα,

NO, and IL-6.[129] Together, these data suggest the heightened confinement ratio and activity of resident immune cells in absence of IL-10 could contribute to increased tissue damage, leading to elevated Lyme pathology.

TLR2-/- Iaβ+ cells did not demonstrate any parameters associated with activation during Bb infection; no change in cell diameter, area or confinement ratio were observed 109

at any times post-infection. This data suggests that TLR2-/- Iaβ+ cells are significantly deficient to recognize or respond to the presence of the resident spirochetes. Dr. Zhang reported that macrophages generate a pro-inflammatory response against Bb in a TLR2- dependent manner, consistent with our observations. Previous studies also indicated that the Bb-elicited IL-10 response was only partially dependent on TLR2. Though loss of

TLR2 resulted in a decrease in IL-10 levels, it did not eliminate the anti-inflammatory response.[172] This suggests that TLR2-/- mice are defective in Bb recognition which prevents APC activation, however a Bb-elicited IL-10 response is still generated and could be potentially dampening APC activities even further. This could be a reason why TLR2 displays such a drastic defect in Bb clearance.

One of the key functions of resident immune cells is to recruit neutrophils to the site of infection or injury. Previous studies have identified that neutrophils can efficiently kill Bb in vitro,[177] and that treatments which enhance neutrophil recruitment to the site of infection demonstrate increased Bb clearance.[117] Based on these results, we used LysM+ mice to visualize neutrophil recruitment to the site of infection during Bb infection. Rapid neutrophil influx was observed at the injection site in WT mice, however these numbers rapidly decreased after 12h post-injection, even though bacterial numbers continued to increase. Bb-elicited IL-10 appears to contribute to this premature suppression of neutrophil infiltration, as IL-10-/- mice display significantly better neutrophil infiltration up to day 7 post-infection. The enhanced neutrophil infiltration in IL-10-/- possibly contribute to the enhanced Bb clearance observed in these mice. However, loss of IL-10 only partially rectifies the dysregulated neutrophil infiltration, since neutrophil levels reach basal levels by day 14 post-infection; this suggests additional IL-10 – independent factors contribute to 110

this suppression of neutrophil recruitment. Currently, no additional contributing factors have been identified. TLR2-/- LysM+ mice were used as controls for this experiment for several reasons. Loss of TLR2 results in decreased production of chemokines (KC, MIP-

2α, MIP-1α, MIP-1β) from Bb-macrophage co-cultures in vitro.[69, 172] The effect of this decreased chemokine levels was observed in vivo during the current study, where TLR2-/-

LysM+ mice display a significant decrease in neutrophil infiltration. Thus, diminished activity of resident immune cells confounded by decreased neutrophil infiltration in TLR2-

/- mice seem to allow enhanced Bb persistence in these mice.

6.3 Role of adaptive immunity in Bb clearance

A major goal for this study was to evaluate the role of the adaptive immune response in the significant decrease in Bb numbers observed between day 8 and day 14 post- infection. Several studies have demonstrated the importance of adaptive immune responses in controlling Lyme disease manifestations, however, the role of adaptive immunity in controlling Bb persistence remains controversial. Barthold et al evaluated arthritis severity in Bb-infected SCID and RAG-/- mice lacking both B and T cells. Both of these mice developed severe arthritis which persisted throughout the course of infection without any signs of spontaneous remission. Further investigation identified B cells as vital for resolution of disease phenotype, as observed in WT BL/6 mice. On the other hand, T cells appear to be responsible for mediating Lyme pathology. While T cell-deficiency had minimal impact on arthritis resolution and other inflammatory Lyme manifestations in

BL/6 mice, adoptive transfer of CD4+ T cells into RAG-/- mice significantly enhanced

111

arthritis and carditis severity during Bb infection. Similar experiments identified CD8+ T cells as prime inducers of Lyme arthritis. [132, 160, 178]

Since B cells were identified to be primarily responsible for control of Lyme disease progression and resolution, several groups have explored the dynamics of B cell responses to Bb infection. Dr. Baumgarth’s group has been seminal in unraveling the reasons for the ineffectiveness of B cell responses in clearing Bb infection.[70, 134, 179] Hastey et al identified distinct stages of B cell responses during active Bb infection using in vivo murine models.

The first stage is dominated by strong T-independent accumulation of B cells in lymph nodes, which is accompanied by induction of Bb-specific Abs in the absence of germinal centers. This is the same stage where a significant decrease in Bb numbers was observed the current IVM studies. This phase lasts for 2 – 2.5 weeks post-infection and is followed by onset of the second stage, which is characterized by the development of germinal centers in lymph nodes. However, these germinal centers are transient and lack clear T and B cell zone demarcation. The final stage exhibits a slow accumulation of long-lived Ab-secreting plasma cells in bone marrow which generates a strong Bb-specific antibody response, but is ultimately ineffective in clearing the pathogen.[70]

To better understand the role of adaptive immunity in controlling Bb infection in the current study, groups of B cell- and T cell-deficient mice were infected with Bb and the effect of these deficiencies on Bb persistence was quantified with the help of IVM. Several aspects of these assays are unique for Lyme disease studies: 1) this is the first study which directly investigates the impact of B and T cell responses on Bb persistence in vivo, since previous studies assessing adaptive immunity focused on resolution of Lyme pathology; and 2) this is the first study which evaluates the contribution of B and T cells at the skin 112

infection site. While most studies focused on effects to the germinal centers and lymph node to understand the dynamics of adaptive immune response, we utilized IVM to visualize Bb persistence in these immune-deficient mice. Since IVM allows us to assess

Bb persistence without the need to euthanize the infected mice, the current studies are the first to continuously and comprehensively monitor Bb persistence in the same mouse in vivo, within intact skin tissues, in real time, over a period of at least 6 months post-infection.

Bb persistence in both nude and TCR-/- mice underwent a significant decrease in Bb numbers after attaining peak Bb load at day 8 post-infection, suggesting that T cell mediated responses do not appear to play a significant role in mediating this control in Bb persistence. However, this decrease was delayed in both these mouse models and variations in Bb persistence were observed between both of these T cell-deficient mouse models. Bb persistence in nude mice closely resembled Bb persistence in WT mice, with significantly higher Bb numbers observed only at day 28 post-infection; the TCR-/- mice demonstrates significantly better Bb persistence until day 60 post-infection. While Bb persistence in both mouse strains display the significant decrease post-day 8 and eventually reach WT levels, the inherent differences in Bb persistence between nude and TCR-/- mice could be justified by one plausible explanation. While the TCR-/- mice are completely devoid of T cells, nude mice (i.e. athymic) still possess relatively small numbers of functional CD4+ and CD8+ T cells, which increase with age. McKisic et al demonstrated that these small number of functional T cells in nude mice might be insufficient to elicit T cell-dependent responses, but may be adequate to mediate and help in a T cell-independent-like response. [132]

µMT mice are completely devoid of mature B cells and cannot generate a Bb-specific antibody response. However, these mice still demonstrate a significant decrease in Bb 113

numbers after attaining peak Bb numbers at day 8 post-infection. This provides direct evidence that Bb-specific antibodies are not responsible for the significant decrease in Bb numbers post-day 8. Interestingly, higher Bb numbers are observed in this mouse line through day 150 post-infection compared to WT mice, highlighting that B cell-mediated responses do have some effect on controlling Bb infection. Bb persistence was observed to be significantly better in µMT mice as compared to AID-/- mice even at day 150 post- infection. While µMT mice are completely devoid of B cells and antibodies, AID-/- mice still possess B cells and can elicit an IgM response. However, AID-/- mice are incapable of undergoing isotype switching and cannot elicit an IgG response. This suggests that the presence of B cells and/or early IgM levels (i.e. natural antibodies) appear to play a significant role in controlling Bb numbers at the site of infection. Conversely, Bb-specific

IgG appears to play minimal role in controlling Bb numbers at the site of infection. Similar conclusions were drawn by Barthold et al while evaluating the efficacy of passive immunizing activity of sera isolated from Bb-infected mice. Studies evaluating the passive immunizing activity of sera against original and autologous late isolates of Bb indicated that sera from chronically infected mice did not discriminate between antigenic differences in the initial and late Bb isolates. The studies also suggested that protective antibody is produced early in the course of Bb infection, which is indicative of a Bb-specific IgM response playing a vital role.[135] Elsner et al also reported that non-switched Bb-specific

IgM antibodies are produced continuously during Bb infection even during chronic disease.

Their data suggests that Bb induces rapid differentiation of cells into antibody-secreting plasma cells instead of developing long-lived T cell-dependent antibody responses.[179]

Hence, Bb infection alters the humoral response away from protective, high-affinity, and 114

long-lived antibody responses, which could explain why the developing IgG response is not effective at controlling Bb persistence. To summarize, while Bb-specific antibodies are not responsible for inducing the significant decrease in Bb numbers after day 8 post- infection, a Bb-specific IgM response is required for better Bb clearance.

6.4 Role of Bb motility in evading host immunity

Studies reported in section 6.2 allows for comparison of the current in vivo assessment of innate responses during Bb infection with the in vitro analysis of these responses conducted by Dr. Zhang during his dissertation research. In general, the effects of IL-10- and TLR2-deficiencies on the innate immune responses were consistent between the in vitro and in vivo studies. To broadly summarize, Bb-elicited IL-10 caused a delay in resident immune cell activation which is accompanied with a significant defect in neutrophil infiltration. Eliminating this Bb-elicited IL-10 results in resident immune cell activation by 6 hours post-infection. This allows for enhanced immune cell activity, as well as better enables neutrophil infiltration to the site of infection. Overall, this produces significantly better Bb clearance, suggesting that IL-10-/- creates a “hyperactive immune environment” which allows for better Bb clearance. On the other hand, TLR2 is vital for

Bb recognition. Absence of TLR2 completely abolishes resident immune cell activation and displays significantly diminished neutrophil infiltration to the infection site. Hence,

TLR2-/- represents a “hypoactive immune environment” which allows significantly better

Bb persistence even at 6 months post-infection. However, the resident immune cells are capable of completely eliminating the spirochete in vitro whereas in vivo these immune responses fail to do so, resulting in a persistent infection; this is a major inconsistency. A 115

major factor contributing to this inconsistency is that the in vitro system is unable to recapitulate spirochetal motility. Hence, the current studies investigated the role of spirochetal motility in evading host immune responses in vivo utilizing IVM.

Bb maintains a velocity of ≥200µm/min which is almost 40 times faster than any immune cell observed in the skin (Lavik et al, unpublished data). To achieve this degree of express motility, Bb dedicates 6% of its genome to its motility and chemotaxis apparatus. [11] Needless to say, both chemotaxis and motility are vital for Bb to maintain its enzootic life cycle. Almost 70% of the genes associated with chemotaxis and flagella assembly are essential for Bb infectivity. [26] Previous investigations assessing infectivity of non-motile Bb (ΔflaB and ΔmotB) also confirmed that motility is essential for Bb infectivity. .[28, 36] ΔcheY3 Bb, though motile, are unable to reverse their direction.[51] Due to this defect they get stuck in dense host tissue which renders them non-motile in vivo and unable to cause infection, further establishing that Bb motility is essential for infectivity.

However, the degree of motility essential for infectivity still remains undetermined. Why does Bb expend such tremendous resources to maintain such high velocity? Does it have to move 40 times faster than host immune cells? The design of the current study allowed these questions to be addressed.

푛표푟푚푎푙 푠푝푒푒푑 WT Bb (normal speed), ΔflaA Bb ( ), and ΔcheY3 Bb (lose motile in 2 vivo) were used to decipher the degree of motility essential for Bb infection, which is a novel methodology. Persistence of these Bb strains was assessed in three different host environments; WT mice (normal immune environment), IL-10-/- mice (hyperactive immune environment) and TLR2-/- mice (hypoactive immune environment). The major highlights from this study were, 1) ΔflaA Bb displayed a significant defect in persistence 116

in WT mice, and achieved a basal level two-fold lower than WT Bb, and 2) ΔflaA Bb display similar persistence curve in both TLR2-/- and WT mice. This was a major finding, since this data suggests that if Bb loses 50% of its velocity, even a hypoactive immune environment is capable of clearing Bb similar to WT mice. TLR2 is essential for Bb recognition[106] and the current studies provide further in vivo evidence that TLR2 provides critical signals for activating resident immune cells during Bb infection. Hence, two major conclusions from studies on ΔflaA Bb persistence are: 1) Reduction of spirochetal average velocity by 50%, which is still 20-fold faster than all immune cells in the skin, significantly hampers Bb persistence in vivo, and 2) the enhanced clearance appears to be TLR2- independent, since enhanced Bb clearance is also observed in TLR2-/- mice.

The true significance of the current motility studies may be best appreciated in relation to the natural enzootic cycle for Bb. 1) Within the mammalian host, Bb must maintain the ability to migrate back into a feeding tick, which can occur anywhere on the host surface and at any time. We propose that the natural high velocities demonstrated by infectious Bb is essential for Bb to be successfully acquired by a tick during its limited feeding time on a susceptible host. Notably, the reduced motility velocities and diminished tissue levels demonstrated by ΔflaA Bb could potentially hamper the chances of Bb acquisition. A xeno-diagnostic study to evaluate ΔflaA Bb acquisition by a naïve tick from an infected mouse is warranted to make any confirmatory remarks. 2) ΔflaA Bb also demonstrate significantly lower peak load and display a longer lag period in skin tissues.

This is a clear indication of a more efficient innate immune response that promotes enhanced Bb clearance. While the reduced velocity of ΔflaA Bb evidently results in better clearance, ΔflaA Bb do persist at significant levels in skin tissues for at least 6 months post- 117

infection. There are several reasons contributing to chronic ΔflaA Bb persistence1) while the reduced velocities do allow enhanced clearance, the velocity could be sufficient to sustain Bb persistence, albeit at a lower basal level. 2) the inoculum used for this study was

105 Bb/ear. While this high inoculum allowed for easier visualization of ΔflaA Bb

[68] persistence, it is almost 3-logs higher than the ID50 for WT Bb. The high inoculum used could potentially overwhelming the normal innate immunity bottleneck, allowing for chronic ΔflaA Bb persistence. A final possible confounder is that, during Bb transmission to a vertebrate host, the spirochete has to migrateopposite the directionality of blood flow into the feed tick. An experiment using artificially-infected nymphal ticks for infecting mice would allow us to determine the effects of reduced ΔflaA Bb velocity on Bb transmission. Similarly, ID50 values for ΔflaA Bb would be useful to determine if the reduced ΔflaA velocity alters the inoculum needed for Bb infection.

A final important finding from these studies is that the enhanced clearance of ΔflaA

Bb appears to be independent of TLR2. It is difficult to speculate on other potential receptors which allow for ΔflaA Bb recognition. Both TLR1 and TLR6 form heterodimers with TLR2, and loss of TLR2 should prevent recognition via these two receptors as well.

CD14 can serve as a potential recognition receptor. CD14 normally acts as a co-receptor for TLR2 and TLR4 and has the ability to recognize both LPS as well as lipoproteins. [113-

115] However, CD14 does not possess intracellular signaling motifs, making it unlikely to provide such activation. ΔflaA Bb persistence in TLR2 and CD14 double KO mice would provide evidence to test this hypothesis.

118

6.5 Concluding remarks

Intravital microscopy (IVM) has been a revelation for studying the immune response generated against the Lyme disease pathogen. Bb is an obligate parasite and the complex interactions between the host immune factors and the spirochete virulence mechanisms that are only functional in host tissues necessitates an in vivo system. Our study is the first to provide a direct snapshot of the innate immune response generated against Bb within intact skin tissues of live mice, and in real-time. While in vitro and ex vivo analysis measure surface expression of receptors and cytokine/ chemokine production in response to Bb infection, our study directly visualizes the immune cells infiltrating the site of infection.

With the help of IVM, we analyzed morphological and dynamic parameters of cellular activity. Thus, our study identifies physical alternation within the host immune response which contribute to enhanced or diminished bacterial clearance.

While a strong technique, a few caveats associated with IVM and our study need to be mentioned: 1) IVM allows us to assess Bb persistence and the immune response generated against Bb within the same mouse, over a period of time, but it restricts the investigation only to the ear skin. While this was not a restriction for our study, since we wanted to characterize the immune response at the site of infection, IVM assessment of other tissue sites usually is a terminal investigation; 2) We adopted an artificial method of needle inoculation for our study. While we acknowledge the fact that tick saliva has several immuno-modulatory effects, previous comparison of Bb persistence in needle-infected and tick-infected mice did not exhibit any significant differences. On the other hand, needle- injections allow us the advantage of controlling the infectious dose, time of infection as well as precise control over the injection site. These advantages permit us to conduct a 119

comprehensive kinetic analysis of the host immune responses. With these limitations in mind, our experimental design has provided a first accurate “live” in vivo view of the host immune response generated against Bb. This experimental system will allow for subsequent visualization of the effects of future therapeutic interventions, vaccines or pharmacological moieties in controlling Bb persistence and/or modulating the host immune response against Bb.

120

References

1. Brock T, Madigan M, Martinko J, Parker J. Biology of microorganisms 7th ed. Prentice Hall Upper Saddle River, New Jersey 1994; 648:650.

2. Burgdorfer W, Barbour AG, Hayes SF, Benach JL, Grunwaldt E, Davis JP. Lyme disease-a tick-borne spirochetosis? Science 1982; 216(4552):1317-1319.

3. Radolf JD, Caimano MJ, Stevenson B, Hu LT. Of ticks, mice and men: understanding the dual-host lifestyle of Lyme disease spirochaetes. Nature reviews microbiology 2012; 10(2):87.

4. Steere AC, Coburn J, Glickstein L. The emergence of Lyme disease. The Journal of clinical investigation 2004; 113(8):1093-1101.

5. Stanek G, Wormser GP, Gray J, Strle F. Lyme borreliosis. Lancet 2012; 379(9814):461- 473.

6. Kurtenbach K, Hanincova K, Tsao JI, Margos G, Fish D, Ogden NH. Fundamental processes in the evolutionary ecology of Lyme borreliosis. Nat Rev Microbiol 2006; 4(9):660-669.

7. McNabb SJ, Jajosky RA, Hall-Baker PA, Adams DA, Sharp P, Worshams C, et al. Summary of notifiable diseases--United States, 2006. MMWR Morb Mortal Wkly Rep 2008; 55(53):1-92.

8. Radolf JD, Samuels DS. Borrelia: molecular biology, host interaction and pathogenesis. Horizon Scientific Press; 2010.

9. Baum SG. Zoonoses-with friends like this, who needs enemies? Transactions of the American Clinical and Climatological Association 2008; 119:39.

10. Murray TS, Shapiro ED. Lyme disease. Clin Lab Med 2010; 30(1):311-328.

11. Fraser CM, Casjens S, Huang WM, Sutton GG, Clayton R, Lathigra R, et al. Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi. Nature 1997; 390(6660):580-586.

12. Casjens S, Palmer N, van Vugt R, Huang WM, Stevenson B, Rosa P, et al. A bacterial genome in flux: the twelve linear and nine circular extrachromosomal DNAs in an infectious isolate of the Lyme disease spirochete Borrelia burgdorferi. Mol Microbiol 2000; 35(3):490-516.

121

13. Palmer N, Fraser C, Casjens S. Distribution of twelve linear extrachromosomal DNAs in natural isolates of Lyme disease spirochetes. J Bacteriol 2000; 182(9):2476- 2480.

14. Hyde FW, Johnson RC. Characterization of a circular plasmid from Borrelia burgdorferi, etiologic agent of Lyme disease. J Clin Microbiol 1988; 26(10):2203-2205.

15. Schwan TG, Burgdorfer W, Garon CF. Changes in infectivity and plasmid profile of the Lyme disease spirochete, Borrelia burgdorferi, as a result of in vitro cultivation. Infect Immun 1988; 56(8):1831-1836.

16. Barbour AG. Plasmid analysis of Borrelia burgdorferi, the Lyme disease agent. J Clin Microbiol 1988; 26(3):475-478.

17. Purser JE, Lawrenz MB, Caimano MJ, Howell JK, Radolf JD, Norris SJ. A plasmid- encoded nicotinamidase (PncA) is essential for infectivity of Borrelia burgdorferi in a mammalian host. Mol Microbiol 2003; 48(3):753-764.

18. Yang XF, Pal U, Alani SM, Fikrig E, Norgard MV. Essential role for OspA/B in the life cycle of the Lyme disease spirochete. J Exp Med 2004; 199(5):641-648.

19. Fischer JR, Parveen N, Magoun L, Leong JM. Decorin-binding proteins A and B confer distinct mammalian cell type-specific attachment by Borrelia burgdorferi, the Lyme disease spirochete. Proc Natl Acad Sci U S A 2003; 100(12):7307-7312.

20. Guo BP, Norris SJ, Rosenberg LC, Hook M. Adherence of Borrelia burgdorferi to the proteoglycan decorin. Infect Immun 1995; 63(9):3467-3472.

21. Guo BP, Brown EL, Dorward DW, Rosenberg LC, Hook M. Decorin-binding adhesins from Borrelia burgdorferi. Mol Microbiol 1998; 30(4):711-723.

22. Zhang JR, Hardham JM, Barbour AG, Norris SJ. Antigenic variation in Lyme disease Borreliae by promiscuous recombination of VMP-like sequence cassettes. Cell 1997; 89(2):275-285.

23. Zhang JR, Norris SJ. Genetic variation of the Borrelia burgdorferi gene vlsE involves cassette-specific, segmental gene conversion. Infect Immun 1998; 66(8):3698- 3704.

24. Strother KO, de Silva A. Role of Borrelia burgdorferi linear plasmid 25 in infection of Ixodes scapularis ticks. J Bacteriol 2005; 187(16):5776-5781.

25. Stewart PE, Byram R, Grimm D, Tilly K, Rosa PA. The plasmids of Borrelia burgdorferi: essential genetic elements of a pathogen. Plasmid 2005; 53(1):1-13.

122

26. Lin T, Gao L, Zhang C, Odeh E, Jacobs MB, Coutte L, et al. Analysis of an ordered, comprehensive STM mutant library in infectious Borrelia burgdorferi: insights into the genes required for mouse infectivity. PLoS One 2012; 7(10):e47532.

27. Li C, Xu H, Zhang K, Liang FT. Inactivation of a putative flagellar motor switch protein FliG1 prevents Borrelia burgdorferi from swimming in highly viscous media and blocks its infectivity. Mol Microbiol 2010; 75(6):1563-1576.

28. Sultan SZ, Manne A, Stewart PE, Bestor A, Rosa PA, Charon NW, et al. Motility is crucial for the infectious life cycle of Borrelia burgdorferi. Infect Immun 2013; 81(6):2012-2021.

29. Sultan SZ, Sekar P, Zhao X, Manne A, Liu J, Wooten RM, et al. Motor rotation is essential for the formation of the periplasmic flagellar ribbon, cellular morphology, and Borrelia burgdorferi persistence within Ixodes scapularis tick and murine hosts. Infect Immun 2015; 83(5):1765-1777.

30. Sze CW, Zhang K, Kariu T, Pal U, Li C. Borrelia burgdorferi needs chemotaxis to establish infection in mammals and to accomplish its enzootic cycle. Infect Immun 2012; 80(7):2485-2492.

31. Peters NC, Egen JG, Secundino N, Debrabant A, Kimblin N, Kamhawi S, et al. In vivo imaging reveals an essential role for neutrophils in leishmaniasis transmitted by sand flies. Science 2008; 321(5891):970-974.

32. Ng LG, Qin JS, Roediger B, Wang Y, Jain R, Cavanagh LL, et al. Visualizing the neutrophil response to sterile tissue injury in mouse dermis reveals a three-phase cascade of events. J Invest Dermatol 2011; 131(10):2058-2068.

33. Hovind-Hougen K. Ultrastructure of spirochetes isolated from Ixodes ricinus and Ixodes dammini. Yale J Biol Med 1984; 57(4):543-548.

34. Charon NW, Goldstein SF, Marko M, Hsieh C, Gebhardt LL, Motaleb MA, et al. The flat-ribbon configuration of the periplasmic flagella of Borrelia burgdorferi and its relationship to motility and morphology. J Bacteriol 2009; 191(2):600-607.

35. Xu H, Raddi G, Liu J, Charon NW, Li C. Chemoreceptors and flagellar motors are subterminally located in close proximity at the two cell poles in spirochetes. J Bacteriol 2011; 193(10):2652-2656.

36. Ge Y, Charon NW. An unexpected flaA homolog is present and expressed in Borrelia burgdorferi. J Bacteriol 1997; 179(2):552-556.

37. Li C, Wolgemuth CW, Marko M, Morgan DG, Charon NW. Genetic analysis of spirochete flagellin proteins and their involvement in motility, filament assembly, and flagellar morphology. J Bacteriol 2008; 190(16):5607-5615.

123

38. Ge Y, Li C, Corum L, Slaughter CA, Charon NW. Structure and expression of the FlaA periplasmic flagellar protein of Borrelia burgdorferi. J Bacteriol 1998; 180(9):2418-2425.

39. Samatey FA, Matsunami H, Imada K, Nagashima S, Shaikh TR, Thomas DR, et al. Structure of the bacterial flagellar hook and implication for the molecular universal joint mechanism. Nature 2004; 431(7012):1062-1068.

40. Sal MS, Li C, Motalab MA, Shibata S, Aizawa S, Charon NW. Borrelia burgdorferi uniquely regulates its motility genes and has an intricate flagellar hook-basal body structure. J Bacteriol 2008; 190(6):1912-1921.

41. Limberger RJ, Slivienski LL, Samsonoff WA. Genetic and biochemical analysis of the flagellar hook of Treponema phagedenis. J Bacteriol 1994; 176(12):3631-3637.

42. Chen S, Beeby M, Murphy GE, Leadbetter JR, Hendrixson DR, Briegel A, et al. Structural diversity of bacterial flagellar motors. EMBO J 2011; 30(14):2972-2981.

43. Liu J, Lin T, Botkin DJ, McCrum E, Winkler H, Norris SJ. Intact flagellar motor of Borrelia burgdorferi revealed by cryo-electron tomography: evidence for stator ring curvature and rotor/C-ring assembly flexion. J Bacteriol 2009; 191(16):5026-5036.

44. Bren A, Eisenbach M. How signals are heard during bacterial chemotaxis: protein- protein interactions in sensory signal propagation. J Bacteriol 2000; 182(24):6865- 6873.

45. Rao CV, Glekas GD, Ordal GW. The three adaptation systems of Bacillus subtilis chemotaxis. Trends Microbiol 2008; 16(10):480-487.

46. Parkinson JS, Ames P, Studdert CA. Collaborative signaling by bacterial chemoreceptors. Curr Opin Microbiol 2005; 8(2):116-121.

47. Charon NW, Goldstein SF. Genetics of motility and chemotaxis of a fascinating group of bacteria: the spirochetes. Annu Rev Genet 2002; 36:47-73.

48. Park SY, Chao X, Gonzalez-Bonet G, Beel BD, Bilwes AM, Crane BR. Structure and function of an unusual family of protein phosphatases: the bacterial chemotaxis proteins CheC and CheX. Mol Cell 2004; 16(4):563-574.

49. Charon NW, Cockburn A, Li C, Liu J, Miller KA, Miller MR, et al. The unique paradigm of spirochete motility and chemotaxis. Annu Rev Microbiol 2012; 66:349- 370.

50. Xu H, Sultan S, Yerke A, Moon KH, Wooten RM, Motaleb MA. Borrelia burgdorferi CheY2 Is Dispensable for Chemotaxis or Motility but Crucial for the Infectious Life Cycle of the Spirochete. Infect Immun 2017; 85(1).

124

51. Novak EA, Sekar P, Xu H, Moon KH, Manne A, Wooten RM, et al. The Borrelia burgdorferi CheY3 response regulator is essential for chemotaxis and completion of its natural infection cycle. Cell Microbiol 2016; 18(12):1782-1799.

52. Steere AC, Malawista SE, Snydman DR, Shope RE, Andiman WA, Ross MR, et al. Lyme arthritis: an epidemic of oligoarticular arthritis in children and adults in three connecticut communities. Arthritis Rheum 1977; 20(1):7-17.

53. Steere AC, Malawista SE, Hardin JA, Ruddy S, Askenase W, Andiman WA. Erythema chronicum migrans and Lyme arthritis. The enlarging clinical spectrum. Ann Intern Med 1977; 86(6):685-698.

54. Steere AC, Malawista SE. Cases of Lyme disease in the United States: locations correlated with distribution of Ixodes dammini. Ann Intern Med 1979; 91(5):730-733.

55. Wallis RC, Brown SE, Kloter KO, Main AJ, Jr. Erythema chronicum migrans and lyme arthritis: field study of ticks. Am J Epidemiol 1978; 108(4):322-327.

56. Steere AC, Broderick TF, Malawista SE. Erythema chronicum migrans and Lyme arthritis: epidemiologic evidence for a tick vector. Am J Epidemiol 1978; 108(4):312- 321.

57. Bosler EM, Coleman JL, Benach JL, Massey DA, Hanrahan JP, Burgdorfer W, et al. Natural Distribution of the Ixodes dammini spirochete. Science 1983; 220(4594):321- 322.

58. Centers for Disease C. Lyme disease surveillance--United States, 1989-1990. MMWR Morb Mortal Wkly Rep 1991; 40(25):417-421.

59. Lyme disease : Data and Surveillance In. https://www.cdc.gov/lyme/: CDC; 2019.

60. Gerstenblith TA, Stern TA. Lyme disease: a review of its epidemiology, evaluation, and treatment. Psychosomatics 2014; 55(5):421-429.

61. Wang P, Glowacki MN, Hoet AE, Needham GR, Smith KA, Gary RE, et al. Emergence of Ixodes scapularis and Borrelia burgdorferi, the Lyme disease vector and agent, in Ohio. Front Cell Infect Microbiol 2014; 4:70.

62. Rydzewski J, Mateus-Pinilla N, Warner RE, Nelson JA, Velat TC. Ixodes scapularis (Acari: Ixodidae) distribution surveys in the Chicago metropolitan region. J Med Entomol 2012; 49(4):955-959.

63. Adams DA, Jajosky RA, Ajani U, Kriseman J, Sharp P, Onwen DH, et al. Summary of notifiable diseases--United States, 2012. MMWR Morb Mortal Wkly Rep 2014; 61(53):1-121.

125

64. Godar DA, Laniosz V, Wetter DA. Lyme disease update for the general dermatologist. Am J Clin Dermatol 2015; 16(1):5-18.

65. Wooten RM, Ma Y, Yoder RA, Brown JP, Weis JH, Zachary JF, et al. Toll-like receptor 2 plays a pivotal role in host defense and inflammatory response to Borrelia burgdorferi. Vector borne and zoonotic diseases 2002; 2(4):275-278.

66. Wooten RM, Ma Y, Yoder RA, Brown JP, Weis JH, Zachary JF, et al. Toll-like receptor 2 is required for innate, but not acquired, host defense to Borrelia burgdorferi. Journal of immunology 2002; 168(1):348-355.

67. Lazarus JJ, Kay MA, McCarter AL, Wooten RM. Viable Borrelia burgdorferi enhances interleukin-10 production and suppresses activation of murine macrophages. Infect Immun 2008; 76(3):1153-1162.

68. Lazarus JJ, Meadows MJ, Lintner RE, Wooten RM. IL-10 deficiency promotes increased Borrelia burgdorferi clearance predominantly through enhanced innate immune responses. Journal of immunology 2006; 177(10):7076-7085.

69. Chung Y, Zhang N, Wooten RM. Borrelia burgdorferi elicited-IL-10 suppresses the production of inflammatory mediators, phagocytosis, and expression of costimulatory receptors by murine macrophages and/or dendritic cells. PLoS One 2013; 8(12):e84980.

70. Hastey CJ, Elsner RA, Barthold SW, Baumgarth N. Delays and diversions mark the development of B cell responses to Borrelia burgdorferi infection. Journal of immunology 2012; 188(11):5612-5622.

71. Bockenstedt LK, Kang I, Chang C, Persing D, Hayday A, Barthold SW. CD4+ T helper 1 cells facilitate regression of murine Lyme carditis. Infect Immun 2001; 69(9):5264-5269.

72. Lazarus JJ, McCarter AL, Neifer-Sadhwani K, Wooten RM. ELISA-based measurement of antibody responses and PCR-based detection profiles can distinguish between active infection and early clearance of Borrelia burgdorferi. Clin Dev Immunol 2012; 2012:138069.

73. Vig DK, Wolgemuth CW. Spatiotemporal evolution of erythema migrans, the hallmark rash of Lyme disease. Biophys J 2014; 106(3):763-768.

74. Weedon D. Weedon's Skin Pathology E-Book: Expert Consult-Online and Print. Elsevier Health Sciences; 2009.

75. Salazar JC, Pope CD, Sellati TJ, Feder HM, Jr., Kiely TG, Dardick KR, et al. Coevolution of markers of innate and adaptive immunity in skin and peripheral blood of patients with erythema migrans. Journal of immunology 2003; 171(5):2660-2670.

126

76. Jacobsen M, Zhou D, Cepok S, Nessler S, Happel M, Stei S, et al. Clonal accumulation of activated CD8+ T cells in the central nervous system during the early phase of neuroborreliosis. The Journal of infectious diseases 2003; 187(6):963-973.

77. Lunemann JD, Gelderblom H, Sospedra M, Quandt JA, Pinilla C, Marques A, et al. Cerebrospinal fluid-infiltrating CD4+ T cells recognize Borrelia burgdorferi lysine- enriched protein domains and central nervous system autoantigens in early lyme encephalitis. Infect Immun 2007; 75(1):243-251.

78. Rupprecht TA, Koedel U, Muhlberger B, Wilske B, Fontana A, Pfister HW. CXCL11 is involved in leucocyte recruitment to the central nervous system in neuroborreliosis. Journal of neurology 2005; 252(7):820-823.

79. Rupprecht TA, Pfister HW, Angele B, Kastenbauer S, Wilske B, Koedel U. The chemokine CXCL13 (BLC): a putative diagnostic marker for neuroborreliosis. Neurology 2005; 65(3):448-450.

80. Steere AC. Lyme disease. The New England journal of medicine 2001; 345(2):115- 125.

81. Steere AC, Levin RE, Molloy PJ, Kalish RA, Abraham Iii JH, Liu NY, et al. Treatment of Lyme arthritis. Arthritis & Rheumatism: Official Journal of the American College of Rheumatology 1994; 37(6):878-888.

82. Dattwyler RJ, Wormser GP, Rush TJ, Finkel MF, Schoen RT, Grunwaldt E, et al. A comparison of two treatment regimens of ceftriaxone in late Lyme disease. Wiener Klinische Wochenschrift 2005; 117(11-12):393-397.

83. Steere AC, Klitz W, Drouin EE, Falk BA, Kwok WW, Nepom GT, et al. Antibiotic- refractory Lyme arthritis is associated with HLA-DR molecules that bind a Borrelia burgdorferi peptide. Journal of Experimental Medicine 2006; 203(4):961-971.

84. Drouin EE, Seward RJ, Strle K, McHugh G, Katchar K, Londono D, et al. A novel human autoantigen, endothelial cell growth factor, is a target of T and B cell responses in patients with Lyme disease. Arthritis Rheum 2013; 65(1):186-196.

85. Shen S, Shin JJ, Strle K, McHugh G, Li X, Glickstein LJ, et al. Treg cell numbers and function in patients with antibiotic-refractory or antibiotic-responsive Lyme arthritis. Arthritis Rheum 2010; 62(7):2127-2137.

86. Katchar K, Drouin EE, Steere AC. Natural killer cells and natural killer T cells in Lyme arthritis. Arthritis research & therapy 2013; 15(6):R183.

87. Bockenstedt LK, Wormser GP. Review: unraveling Lyme disease. Arthritis & rheumatology (Hoboken, NJ) 2014; 66(9):2313-2323.

127

88. Bockenstedt LK, Gonzalez DG, Haberman AM, Belperron AA. Spirochete antigens persist near cartilage after murine Lyme borreliosis therapy. J Clin Invest 2012; 122(7):2652-2660.

89. Zhang Y, Liang C. Innate recognition of microbial-derived signals in immunity and inflammation. Sci China Life Sci 2016; 59(12):1210-1217.

90. Hoffmann J, Akira S. Innate immunity. Curr Opin Immunol 2013; 25(1):1-3.

91. Chaplin DD. Overview of the immune response. J Allergy Clin Immunol 2010; 125(2 Suppl 2):S3-S23.

92. Janeway Jr CA, Travers P, Walport M, Shlomchik MJ. Principles of innate and adaptive immunity. In: Immunobiology: The Immune System in Health and Disease 5th edition: Garland Science; 2001.

93. Turvey SE, Broide DH. Innate immunity. Journal of Allergy and Clinical Immunology 2010; 125(2):S24-S32.

94. Borghesi L, Milcarek C. Innate versus adaptive immunity: a paradigm past its prime? Cancer research 2007; 67(9):3989-3993.

95. Bonilla FA, Oettgen HC. Adaptive immunity. Journal of Allergy and Clinical Immunology 2010; 125(2):S33-S40.

96. Paul WE. Bridging innate and adaptive immunity. Cell 2011; 147(6):1212-1215.

97. Forthal DN. Functions of Antibodies. Microbiol Spectr 2014; 2(4):1-17.

98. Troy EB, Lin T, Gao L, Lazinski DW, Camilli A, Norris SJ, et al. Understanding barriers to Borrelia burgdorferi dissemination during infection using massively parallel sequencing. Infect Immun 2013; 81(7):2347-2357.

99. Kenedy MR, Lenhart TR, Akins DR. The role of Borrelia burgdorferi outer surface proteins. FEMS immunology and medical microbiology 2012; 66(1):1-19.

100. Lien E, Sellati TJ, Yoshimura A, Flo TH, Rawadi G, Finberg RW, et al. Toll-like receptor 2 functions as a pattern recognition receptor for diverse bacterial products. The Journal of biological chemistry 1999; 274(47):33419-33425.

101. Takeuchi O, Sato S, Horiuchi T, Hoshino K, Takeda K, Dong Z, et al. Cutting edge: role of Toll-like receptor 1 in mediating immune response to microbial lipoproteins. Journal of immunology 2002; 169(1):10-14.

102. Shin OS, Isberg RR, Akira S, Uematsu S, Behera AK, Hu LT. Distinct roles for MyD88 and Toll-like receptors 2, 5, and 9 in phagocytosis of Borrelia burgdorferi and cytokine induction. Infect Immun 2008; 76(6):2341-2351. 128

103. Guerau-de-Arellano M, Huber BT. Chemokines and Toll-like receptors in Lyme disease pathogenesis. Trends in molecular medicine 2005; 11(3):114-120.

104. Singh SK, Girschick HJ. Toll-like receptors in Borrelia burgdorferi-induced inflammation. Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 2006; 12(8):705-717.

105. Cervantes JL, Dunham-Ems SM, La Vake CJ, Petzke MM, Sahay B, Sellati TJ, et al. Phagosomal signaling by Borrelia burgdorferi in human monocytes involves Toll-like receptor (TLR) 2 and TLR8 cooperativity and TLR8-mediated induction of IFN-beta. Proc Natl Acad Sci U S A 2011; 108(9):3683-3688.

106. Hirschfeld M, Kirschning CJ, Schwandner R, Wesche H, Weis JH, Wooten RM, et al. Cutting edge: inflammatory signaling by Borrelia burgdorferi lipoproteins is mediated by toll-like receptor 2. Journal of immunology 1999; 163(5):2382-2386.

107. Wang G, Ma Y, Buyuk A, McClain S, Weis JJ, Schwartz I. Impaired host defense to infection and Toll-like receptor 2-independent killing of Borrelia burgdorferi clinical isolates in TLR2-deficient C3H/HeJ mice. FEMS microbiology letters 2004; 231(2):219-225.

108. Brodsky I, Medzhitov R. Two modes of ligand recognition by TLRs. Cell 2007; 130(6):979-981.

109. Jin MS, Kim SE, Heo JY, Lee ME, Kim HM, Paik SG, et al. Crystal structure of the TLR1-TLR2 heterodimer induced by binding of a tri-acylated lipopeptide. Cell 2007; 130(6):1071-1082.

110. Fikrig E, Narasimhan S, Neelakanta G, Pal U, Chen M, Flavell R. Toll-like receptors 1 and 2 heterodimers alter Borrelia burgdorferi gene expression in mice and ticks. The Journal of infectious diseases 2009; 200(8):1331-1340.

111. Oosting M, Ter Hofstede H, Sturm P, Adema GJ, Kullberg BJ, van der Meer JW, et al. TLR1/TLR2 heterodimers play an important role in the recognition of Borrelia spirochetes. PLoS One 2011; 6(10):e25998.

112. Bulut Y, Faure E, Thomas L, Equils O, Arditi M. Cooperation of Toll-like receptor 2 and 6 for cellular activation by soluble tuberculosis factor and Borrelia burgdorferi outer surface protein A lipoprotein: role of Toll-interacting protein and IL-1 receptor signaling molecules in Toll-like receptor 2 signaling. Journal of immunology 2001; 167(2):987-994.

113. Giambartolomei GH, Dennis VA, Lasater BL, Philipp MT. Induction of pro- and anti-inflammatory cytokines by Borrelia burgdorferi lipoproteins in monocytes is mediated by CD14. Infection and immunity 1999; 67(1):140-147.

129

114. Wooten RM, Morrison TB, Weis JH, Wright SD, Thieringer R, Weis JJ. The role of CD14 in signaling mediated by outer membrane lipoproteins of Borrelia burgdorferi. J Immunol 1998; 160(11):5485-5492.

115. Sellati TJ, Bouis DA, Caimano MJ, Feulner JA, Ayers C, Lien E, et al. Activation of human monocytic cells by Borrelia burgdorferi and Treponema pallidum is facilitated by CD14 and correlates with surface exposure of spirochetal lipoproteins. Journal of immunology 1999; 163(4):2049-2056.

116. Sahay B, Singh A, Gnanamani A, Patsey RL, Blalock JE, Sellati TJ. CD14 signaling reciprocally controls collagen deposition and turnover to regulate the development of lyme arthritis. The American journal of pathology 2011; 178(2):724-734.

117. Xu Q, Seemanapalli SV, Reif KE, Brown CR, Liang FT. Increasing the recruitment of neutrophils to the site of infection dramatically attenuates Borrelia burgdorferi infectivity. Journal of immunology 2007; 178(8):5109-5115.

118. Ouyang W, O'Garra A. IL-10 Family Cytokines IL-10 and IL-22: from Basic Science to Clinical Translation. Immunity 2019; 50(4):871-891.

119. Couper KN, Blount DG, Riley EM. IL-10: the master regulator of immunity to infection. Journal of immunology 2008; 180(9):5771-5777.

120. Gabrysova L, Howes A, Saraiva M, O'Garra A. The regulation of IL-10 expression. Curr Top Microbiol Immunol 2014; 380:157-190.

121. Saraiva M, O'Garra A. The regulation of IL-10 production by immune cells. Nat Rev Immunol 2010; 10(3):170-181.

122. Redford PS, Murray PJ, O'Garra A. The role of IL-10 in immune regulation during M. tuberculosis infection. Mucosal Immunol 2011; 4(3):261-270.

123. Redford PS, Boonstra A, Read S, Pitt J, Graham C, Stavropoulos E, et al. Enhanced protection to Mycobacterium tuberculosis infection in IL-10-deficient mice is accompanied by early and enhanced Th1 responses in the lung. Eur J Immunol 2010; 40(8):2200-2210.

124. Newton SM, Smith RJ, Wilkinson KA, Nicol MP, Garton NJ, Staples KJ, et al. A deletion defining a common Asian lineage of Mycobacterium tuberculosis associates with immune subversion. Proc Natl Acad Sci U S A 2006; 103(42):15594-15598.

125. Giambartolomei GH, Dennis VA, Philipp MT. Borrelia burgdorferi stimulates the production of interleukin-10 in peripheral blood mononuclear cells from uninfected humans and rhesus monkeys. Infect Immun 1998; 66(6):2691-2697.

130

126. Pohl-Koppe A, Balashov KE, Steere AC, Logigian EL, Hafler DA. Identification of a T cell subset capable of both IFN-gamma and IL-10 secretion in patients with chronic Borrelia burgdorferi infection. J Immunol 1998; 160(4):1804-1810.

127. Ganapamo F, Dennis VA, Philipp MT. Early induction of gamma interferon and interleukin-10 production in draining lymph nodes from mice infected with Borrelia burgdorferi. Infect Immun 2000; 68(12):7162-7165.

128. Sonderegger FL, Ma Y, Maylor-Hagan H, Brewster J, Huang X, Spangrude GJ, et al. Localized production of IL-10 suppresses early inflammatory cell infiltration and subsequent development of IFN-gamma-mediated Lyme arthritis. Journal of immunology 2012; 188(3):1381-1393.

129. Brown JP, Zachary JF, Teuscher C, Weis JJ, Wooten RM. Dual role of interleukin- 10 in murine Lyme disease: regulation of arthritis severity and host defense. Infect Immun 1999; 67(10):5142-5150.

130. Gautam A, Dixit S, Philipp MT, Singh SR, Morici LA, Kaushal D, et al. Interleukin- 10 alters effector functions of multiple genes induced by Borrelia burgdorferi in macrophages to regulate Lyme disease inflammation. Infect Immun 2011; 79(12):4876- 4892.

131. Schaible UE, Kramer MD, Museteanu C, Zimmer G, Mossmann H, Simon MM. The severe combined immunodeficiency (scid) mouse. A laboratory model for the analysis of Lyme arthritis and carditis. J Exp Med 1989; 170(4):1427-1432.

132. McKisic MD, Barthold SW. T-cell-independent responses to Borrelia burgdorferi are critical for protective immunity and resolution of lyme disease. Infect Immun 2000; 68(9):5190-5197.

133. Connolly SE, Benach JL. The versatile roles of antibodies in Borrelia infections. Nat Rev Microbiol 2005; 3(5):411-420.

134. Tunev SS, Hastey CJ, Hodzic E, Feng S, Barthold SW, Baumgarth N. Lymphoadenopathy during lyme borreliosis is caused by spirochete migration- induced specific B cell activation. PLoS Pathog 2011; 7(5):e1002066.

135. Barthold SW, Bockenstedt LK. Passive immunizing activity of sera from mice infected with Borrelia burgdorferi. Infect Immun 1993; 61(11):4696-4702.

136. Belperron AA, Dailey CM, Booth CJ, Bockenstedt LK. Marginal zone B-cell depletion impairs murine host defense against Borrelia burgdorferi infection. Infection and immunity 2007; 75(7):3354-3360.

131

137. Iliopoulou BP, Alroy J, Huber BT. CD28 deficiency exacerbates joint inflammation upon Borrelia burgdorferi infection, resulting in the development of chronic Lyme arthritis. Journal of immunology 2007; 179(12):8076-8082.

138. Shanafelt MC, Kang I, Barthold SW, Bockenstedt LK. Modulation of murine Lyme borreliosis by interruption of the B7/CD28 T-cell costimulatory pathway. Infection and immunity 1998; 66(1):266-271.

139. Anguita J, Rincon M, Samanta S, Barthold SW, Flavell RA, Fikrig E. Borrelia burgdorferi-infected, interleukin-6-deficient mice have decreased Th2 responses and increased lyme arthritis. The Journal of infectious diseases 1998; 178(5):1512-1515.

140. Brown CR, Reiner SL. Experimental lyme arthritis in the absence of interleukin- 4 or gamma interferon. Infect Immun 1999; 67(7):3329-3333.

141. Keane-Myers A, Nickell SP. Role of IL-4 and IFN-gamma in modulation of immunity to Borrelia burgdorferi in mice. Journal of immunology 1995; 155(4):2020- 2028.

142. Anguita J, Persing DH, Rincon M, Barthold SW, Fikrig E. Effect of anti-interleukin 12 treatment on murine lyme borreliosis. J Clin Invest 1996; 97(4):1028-1034.

143. Whiteside SK, Snook JP, Ma Y, Sonderegger FL, Fisher C, Petersen C, et al. IL-10 Deficiency Reveals a Role for TLR2-Dependent Bystander Activation of T Cells in Lyme Arthritis. Journal of immunology 2018; 200(4):1457-1470.

144. de Silva AM, Telford SR, 3rd, Brunet LR, Barthold SW, Fikrig E. Borrelia burgdorferi OspA is an arthropod-specific transmission-blocking Lyme disease vaccine. J Exp Med 1996; 183(1):271-275.

145. Tilly K, Rosa PA, Stewart PE. Biology of infection with Borrelia burgdorferi. Infect Dis Clin North Am 2008; 22(2):217-234, v.

146. Harman MW, Dunham-Ems SM, Caimano MJ, Belperron AA, Bockenstedt LK, Fu HC, et al. The heterogeneous motility of the Lyme disease spirochete in gelatin mimics dissemination through tissue. Proc Natl Acad Sci U S A 2012; 109(8):3059-3064.

147. Cruz AR, Moore MW, La Vake CJ, Eggers CH, Salazar JC, Radolf JD. Phagocytosis of Borrelia burgdorferi, the Lyme disease spirochete, potentiates innate immune activation and induces apoptosis in human monocytes. Infect Immun 2008; 76(1):56- 70.

148. Dunham-Ems SM, Caimano MJ, Pal U, Wolgemuth CW, Eggers CH, Balic A, et al. Live imaging reveals a biphasic mode of dissemination of Borrelia burgdorferi within ticks. J Clin Invest 2009; 119(12):3652-3665.

132

149. Moriarty TJ, Norman MU, Colarusso P, Bankhead T, Kubes P, Chaconas G. Real- time high resolution 3D imaging of the lyme disease spirochete adhering to and escaping from the vasculature of a living host. PLoS Pathog 2008; 4(6):e1000090.

150. Elias AF, Stewart PE, Grimm D, Caimano MJ, Eggers CH, Tilly K, et al. Clonal polymorphism of Borrelia burgdorferi strain B31 MI: implications for mutagenesis in an infectious strain background. Infect Immun 2002; 70(4):2139-2150.

151. Faust N, Varas F, Kelly LM, Heck S, Graf T. Insertion of enhanced green fluorescent protein into the lysozyme gene creates mice with green fluorescent granulocytes and macrophages. Blood 2000; 96(2):719-726.

152. Boes M, Cerny J, Massol R, Op den Brouw M, Kirchhausen T, Chen J, et al. T-cell engagement of dendritic cells rapidly rearranges MHC class II transport. Nature 2002; 418(6901):983-988.

153. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. Interleukin-10-deficient mice develop chronic enterocolitis. Cell 1993; 75(2):263-274.

154. Werts C, Tapping RI, Mathison JC, Chuang TH, Kravchenko V, Saint Girons I, et al. Leptospiral lipopolysaccharide activates cells through a TLR2-dependent mechanism. Nat Immunol 2001; 2(4):346-352.

155. Muramatsu M, Kinoshita K, Fagarasan S, Yamada S, Shinkai Y, Honjo T. Class switch recombination and hypermutation require activation-induced cytidine deaminase (AID), a potential RNA editing enzyme. Cell 2000; 102(5):553-563.

156. Kitamura D, Roes J, Kuhn R, Rajewsky K. A B cell-deficient mouse by targeted disruption of the membrane exon of the immunoglobulin mu chain gene. Nature 1991; 350(6317):423-426.

157. Pantelouris EM. Absence of thymus in a mouse mutant. Nature 1968; 217(5126):370-371.

158. Mombaerts P, Clarke AR, Rudnicki MA, Iacomini J, Itohara S, Lafaille JJ, et al. Mutations in T-cell antigen receptor genes alpha and beta block thymocyte development at different stages. Nature 1992; 360(6401):225-231.

159. Kamanaka M, Kim ST, Wan YY, Sutterwala FS, Lara-Tejero M, Galan JE, et al. Expression of interleukin-10 in intestinal lymphocytes detected by an interleukin-10 reporter knockin tiger mouse. Immunity 2006; 25(6):941-952.

160. Barthold SW, Sidman CL, Smith AL. Lyme Borreliosis in Genetically Resistant and Susceptible Mice with Severe Combined Immunodeficiency. The American Journal of Tropical Medicine and Hygiene 1992; 47(5):605-613.

133

161. Erdile LF, Brandt MA, Warakomski DJ, Westrack GJ, Sadziene A, Barbour AG, et al. Role of attached lipid in immunogenicity of Borrelia burgdorferi OspA. Infect Immun 1993; 61(1):81-90.

162. LaRocca TJ, Holthausen DJ, Hsieh C, Renken C, Mannella CA, Benach JL. The bactericidal effect of a complement-independent antibody is osmolytic and specific to Borrelia. Proc Natl Acad Sci U S A 2009; 106(26):10752-10757.

163. LaRocca TJ, Benach JL. The important and diverse roles of antibodies in the host response to Borrelia infections. Curr Top Microbiol Immunol 2008; 319:63-103.

164. Wang X, Ma Y, Weis JH, Zachary JF, Kirschning CJ, Weis JJ. Relative contributions of innate and acquired host responses to bacterial control and arthritis development in Lyme disease. Infect Immun 2005; 73(1):657-660.

165. Sekar P. The effects of key motility and chemotaxis genes on Borrelia burgdorferi dissemination and evasion of immune clearance in murine tissues [Dissertation]: University of Toledo; 2015.

166. Gondolf KB, Mihatsch M, Curschellas E, Dunn JJ, Batsford SR. Induction of experimental allergic arthritis with outer surface proteins of Borrelia burgdorferi. Arthritis Rheum 1994; 37(7):1070-1077.

167. de Souza MS, Smith AL, Beck DS, Terwilliger GA, Fikrig E, Barthold SW. Long- term study of cell-mediated responses to Borrelia burgdorferi in the laboratory mouse. Infect Immun 1993; 61(5):1814-1822.

168. Potter MR, Noben-Trauth N, Weis JH, Teuscher C, Weis JJ. Interleukin-4 (IL-4) and IL-13 signaling pathways do not regulate Borrelia burgdorferi-induced arthritis in mice: IgG1 is not required for host control of tissue spirochetes. Infect Immun 2000; 68(10):5603-5609.

169. Seiler KP, Vavrin Z, Eichwald E, Hibbs JB, Jr., Weis JJ. Nitric oxide production during murine Lyme disease: lack of involvement in host resistance or pathology. Infect Immun 1995; 63(10):3886-3895.

170. Qiao S, Qian Y, Xu G, Luo Q, Zhang Z. Long-term characterization of activated microglia/macrophages facilitating the development of experimental brain metastasis through intravital microscopic imaging. J Neuroinflammation 2019; 16(1):4.

171. Burkitt HG, Young B, Heath JW, Wheater PR, Deakin PJ. Wheater's functional histology: a text and colour atlas. Churchill Livingstone; 1993.

172. Zhang N. Identification of Receptors and Signaling Pathways Involved in Borrelia burgdorferi-Elicited IL-10 and Potential Therapies for Lyme disease: University of Toledo; 2014.

134

173. Shin JJ, Glickstein LJ, Steere AC. High levels of inflammatory chemokines and cytokines in joint fluid and synovial tissue throughout the course of antibiotic- refractory lyme arthritis. Arthritis Rheum 2007; 56(4):1325-1335.

174. Lea S, Dungwa J, Ravi A, Singh D. Alveolar macrophage size is increased in COPD patients compared to controls. European Respiratory Journal 2017; 50(suppl 61):PA1011.

175. Egen JG, Rothfuchs AG, Feng CG, Winter N, Sher A, Germain RN. Macrophage and T cell dynamics during the development and disintegration of mycobacterial granulomas. Immunity 2008; 28(2):271-284.

176. Girgis NM, Gundra UM, Ward LN, Cabrera M, Frevert U, Loke P. Ly6C(high) monocytes become alternatively activated macrophages in schistosome granulomas with help from CD4+ cells. PLoS Pathog 2014; 10(6):e1004080.

177. Lusitani D, Malawista SE, Montgomery RR. Borrelia burgdorferi are susceptible to killing by a variety of human polymorphonuclear leukocyte components. The Journal of infectious diseases 2002; 185(6):797-804.

178. McKisic MD, Redmond WL, Barthold SW. Cutting edge: T cell-mediated pathology in murine Lyme borreliosis. Journal of immunology 2000; 164(12):6096- 6099.

179. Elsner RA, Hastey CJ, Baumgarth N. CD4+ T cells promote antibody production but not sustained affinity maturation during Borrelia burgdorferi infection. Infect Immun 2015; 83(1):48-56.

135

Appendix A

Permission to use Figure (I) in dissertation

Figure 1 – 1 was reproduced after obtaining permission from Springer Nature: Nat Rev

Microbiol 10(2): 87–99. doi:10.1038/nrmicro2714

136

Appendix B

Permission to use Figure (II) in dissertation

Figure 1 – 2 (panel a and b) was reproduced after obtaining permission from Elsevier:

Current Opinion in Microbiology 28 (2015): 106-113. 10.1016/j.m.09.006

137

Appendix C

IVM images of Bb numbers in WT, IL-10 KO and TLR2 KO mice

Representative IVM images of WT-GFP Bb (denoted by white arrows) persistence in ear skin of WT, IL-10-/- and TLR2-/- mice (quantified by manual enumeration in figure 4 – 3)

138

Appendix D

IVM images of LysM+ cells in WT, IL-10 KO and TLR2 KO mice

Representative IVM images of GFP-expressing cells (white arrows) in ear skin of LysM+ mice on a WT, IL-10-/- and TLR2-/- background (quantified manually in figure 4 – 5)

139

Appendix E

IVM images of Iaβ+ cells in WT, IL-10 KO and TLR2 KO mice

Representative IVM images of GFP-expressing cells (white arrows) in ear skin of Iaβ+ mice on a WT, IL-10-/- and TLR2-/- background (quantified manually in figure 4 – 6).

140