Time-dependent alterations in memory CD8 T cell function after infection

Transcription

1 University of Iowa Iowa Research Online Theses and Dissertations Spring 2016 Time-dependent alterations in memory CD8 T cell function after infection Matthew David Martin University of Iowa Copyright 2016 Matthew David Martin This dissertation is available at Iowa Research Online: Recommended Citation Martin, Matthew David. "Time-dependent alterations in memory CD8 T cell function after infection." PhD (Doctor of Philosophy) thesis, University of Iowa, Follow this and additional works at: Part of the Immunology of Infectious Disease Commons

2 TIME-DEPENDENT ALTERATIONS IN MEMORY CD8 T CELL FUNCTION AFTER INFECTION by Matthew David Martin A thesis submitted in partial fulfillment of the requirements for the Doctor of Philosophy degree in Immunology in the Graduate College of The University of Iowa May 2016 Thesis Supervisor: Associate Professor Vladimir P. Badovinac

3 Copyright by Matthew David Martin 2016 All Rights Reserved

4 Graduate College The University of Iowa Iowa City, Iowa CERTIFICATE OF APPROVAL This is to certify that the Ph.D. thesis of PH.D. THESIS Matthew David Martin has been approved by the Examining Committee for the thesis requirement for the Doctor of Philosophy degree in Immunology at the May 2016 graduation. Thesis Committee: Vladimir P. Badovinac, Thesis Supervisor Hai-Hui Xue John T. Harty Kevin L. Legge Steven M. Varga

5 To my friends, family, and mentors, thank you for all of your support along the way. ii

6 ACKNOWLEDGEMENTS I would like to begin by thanking my Ph.D. advisor, Dr. Vladimir Badovinac for his support and mentorship through the years. He showed patience with me as I learned what was involved in advanced scientific research while I was a Master s student, and encouraged me to think independently and to further develop my scientific skills during my Ph.D. studies. I would like to thank the Interdisciplinary Graduate Program in Immunology for financial support during portions of my graduate student tenure as well as the tremendous efforts to ensure student success by administrative personnel including current director Dr. Steven Varga, former director Dr. Gail Bishop, Paulette Villhauer, and Josh Lobb. I would also like to thank the members of my thesis dissertation committee, Dr. Hai-Hui Xue, Dr. John Harty, Dr. Kevin Legge, and Dr. Steven Varga. They encouraged me to think critically about my project and provided pointed advice that was very helpful for constructing a Thesis dissertation that I can be proud of. I would like to thank the past and present members of the Badovinac laboratory, Deepa Rai, Shaniya Khan, Stephanie Condotta, Sean Duong, Robby Markwart, Derek Danahy, Robert Strother, and Stacey Hartwig. Their support and scientific discussion has been invaluable through the years. I would like to thank members of the Harty, Varga, Legge, Xue, and Waldschmidt laboratories for their helpful discussions during weekly joint-lab meetings. These meetings spurred me to think critically about my projects and how to advance my ideas, iii

7 and provided a supportive platform as I developed the skill of effective scientific communication. I would like to thank several people for their assistance with experiments and data analysis included in this Thesis. Hai-Hui Xue and Ramakrishna Sompallae for assistance in microarray data analysis. Deepa Rai for assistance in determining viral titers. Marie Kim for assistance with preparing samples and analyzing data for metabolic studies. Qiang Shan for assistance with mrna collection and RT-PCR analysis. Finally, I would like to thank the Pathology Department, especially Carla Hartl and Dr. Tom Waldschmidt. Through the Pathology Master s program I was given the opportunity to test the waters of advanced scientific research. Without that chance I would never have had the opportunity to pursue and complete my Ph.D. iv

8 ABSTRACT CD8 T cells play a critical role in the clearance of pathogenic bacteria, viruses, and protozoan parasites. Upon encountering their cognate antigen through either infection or vaccination, naïve CD8 T cells undergo robust proliferative expansion, which is followed by contraction and the formation of a memory population. Memory CD8 T cells are long-lived, and because they persist in increased numbers and possess enhanced functional abilities compared to naïve CD8 T cells, they are able to provide the host with increased protection following re-infection. Because of these properties, vaccines designed to elicit memory CD8 T cells have the potential to reduce health care burdens related to infection with pathogens including human immuno deficiency virus (HIV), malaria, influenza, and hepatitis virus. However, stimulating protective CD8 T cell responses against these pathogens through vaccination has proven challenging. Therefore, a better understanding of the properties of memory CD8 T cells generated following vaccination, and the characteristics of memory CD8 T cells best suited for providing protection against diverse pathogens is needed. While memory CD8 T cells can be maintained for as long as the life of the host, evidence suggests that their properties change with time after infection. Because CD8 T cell-mediated protection is based upon both the numbers and quality or functional abilities of memory cells present at the time of re-infection, changes in memory CD8 T cell function over time could impact their ability to provide protection upon re-infection. Therefore, a better understanding of how memory CD8 T cells change with time after infection is needed. As part of the studies presented in this thesis, I found that the phenotype and function of memory CD8 T cells including localization, interleukin (IL)-2 v

9 cytokine production, responsiveness to homeostatic cytokines, metabolic capabilities, and proliferation and secondary memory generation potential change with time after infection. Interestingly functional changes could not be completely explained by changes in subset composition that occur with time, as changes over time were also seen in defined CD62L hi subsets. Importantly, functional changes of memory CD8 T cells that occurred with time led to an increased ability to provide protection against a chronic viral infection. These data improve our knowledge of the capabilities of memory CD8 T cells generated following infection, and suggests that the outcome of vaccination strategies designed to elicit protective memory CD8 T cells using single or prime-boost immunizations will depend upon the timing between antigen encounters. Following re-infection, memory CD8 T cells become activated and produce effector cytokines and cytolytic molecules that aid the host in clearing invading microbes. Activation can be triggered not only through cognate antigen recognition, but also by antigen-independent cytokine driven signals. However, our knowledge of how antigendependent and independent signals contribute to CD8 T cell activation and protection following infection is incomplete. In the second part of my thesis, I show that the ability of memory CD8 T cells to become activated in response to inflammation decreases with time after infection, that antigen and inflammation act synergistically to induce activation of memory CD8 T cells, that the presence of cognate antigen enhances activation of memory CD8 T cells that contribute to clearance of infection, and that bystander memory CD8 T cell responses following unrelated bacterial infection do not provide the host with a protective benefit. vi

10 Together, the data in this thesis further our understanding of memory CD8 T cells generated following infection and/or vaccination, and the properties of memory CD8 T cells important for providing protection upon re-infection with invading pathogens. vii

11 PUBLIC ABSTRACT Generating protective memory CD8 T cells through vaccination holds the potential for reducing public health threats from pathogens including HIV, malaria, influenza, and hepatitis virus. However, the pursuit of this goal has remained challenging, and a better understanding of the properties of memory CD8 T cells generated following infection and/or vaccination, and the characteristics of protective memory CD8 T cells is needed. Memory CD8 T cells can be maintained for long-periods of time, but some of their functional properties have been shown to change with time after infection. I found that many functional abilities of memory CD8 T cells including effector cytokine production, metabolic capabilities, and proliferation following reinfection increase with time. Importantly, these changes led to increased memory CD8 T cell-mediated protection against chronic viral infection. Furthermore, I examined how antigen and inflammatory cytokine-induced activation of memory CD8 T cells is regulated, and the contributions of antigen-dependent and -independent memory CD8 T cell responses to protection following infection. I found that the ability of memory CD8 T cells to sense inflammation decreases with time after infection, that antigen promotes robust CD8 T cell responses following infection, and that bystander responses during non-related infection do not provide the host with a protective benefit. This data furthers our understanding of the properties of memory CD8 T cells generated following infection and/or vaccination and the characteristics of protective memory CD8 T cell responses. It suggests that the outcome of single and prime/boost vaccination will depend on the timing of re-infection. viii

12 TABLE OF CONTENTS LIST OF TABLES... xii LIST OF FIGURES... xiii LIST OF ABBREVIATIONS... xvi CHAPTER I: INTRODUCTION...1 The CD8 T Cell Response to Infection and/or Vaccination...1 Characteristics of Memory CD8 T Cells...2 Memory CD8 T Cell Activation and Effector Functions...3 Memory CD8 T Cell Heterogeneity...6 The Process of Memory CD8 T Cell Differentiation...8 Maintenance and Longevity of Primary Memory CD8 T Cells...10 Changes in Primary Memory CD8 T Cell Properties with Time after Infection...13 Effects of Additional Ag Encounters on Memory CD8 T Cell Properties...16 LM and LCMV as Model Pathogens Used to Study the CD8 T Cell Response to Infection...20 Thesis Rationale and Objectives...24 CHAPTER II: PHENOTYPIC AND FUNCTIONAL ALTERATIONS IN CIRCULATING MEMORY CD8 T CELLS WITH TIME AFTER PRIMARY INFECTION...32 Abstract...32 Introduction...33 Materials and Methods...36 Mice, Bacteria, and Viruses...36 Adoptive Transfers and Generation of earlym and latem P14 and Endogenous Memory Cells...37 Lymphocyte Isolation, Quantification of CD8 T Cell Responses and Surface Marker Expression, ICS and Degranulation Analysis, Transcription Factor Staining, and KI67 Staining...38 Detection of BrdU Uptake, Responsiveness to IL-15, and Homeostatic Proliferation in Rag-/- Mice...40 Measure of Bacterial and Viral Clearance...42 Microarray Data Acquisition and Analysis...43 Quantitative RT-PCR...44 Metabolism Assays...46 Results...47 Changes Occur in Memory CD8 T Cell Location, Phenotype, Function, and Maintenance with Time after Infection...47 Antigen-driven Proliferation and Memory Generation Potential Increase in Memory CD8 T Cells over Time...49 ix

13 Phenotypic Heterogeneity of CD62L hi Memory CD8 T Cells Decreases with Time...50 Gene Expression Patterns among CD62L hi Memory CD8 T Cells Change with Time...53 Cytokine Production, Degranulation, and Functional Avidity of CD62L hi Memory CD8 T Cells is not Affected by Time...55 Ability to Undergo Homeostatic and Ag-driven Proliferation Increases with Time among CD62L hi Memory CD8 T Cells...57 Protective Capacity of Memory CD8 T Cells Changes with Time in a Pathogen Dependent Manner...59 Mitochondrial Function of Memory CD8 T Cells Improves with Time after Infection...62 Discussion...64 CHAPTER III: ANTIGEN-DEPENDENT AND INDEPENDENT CONTRIBUTIONS TO PRIMARY MEMORY CD8 T CELL ACTIVATION AND PROTECTION FOLLOWING INFECTION Abstract Introduction Materials and Methods Mice Bacterial and Viral Infections, Memory Generation, Adoptive Transfers, and Ampicillin Treatment Detection of Memory Cells and ex vivo P14 Cell Analysis In vitro Activation of P14 Cells Detection of Serum IL Microarray Data Acquisition and Analysis Statistical Analysis Results Ability of Memory CD8 T Cells to Undergo Bystander Activation Decreases with Time after Memory Generation Ability of Memory CD8 T Cells to Sense Inflammation Decreases with Time after Infection Ag and Inflammation Act Synergistically in vitro to Induce Memory CD8 T Cell Activation Activation of latem CD8 T Cells is more Tightly Regulated, but they Remain Fully Capable of Responding to Infection Expressing Recognized Ags Early Activation of Memory CD8 T Cells that do not Significantly Contribute to Clearance of Infection is not Influenced by Cognate Ag..125 Cognate Ag and Continued Infection Drive Early Activation and Sustained Responses of Memory CD8 T Cells that Contribute to Pathogen Clearance Bystander Memory CD8 T Cell Responses are Influenced by Dose of Infection and Amount of Inflammation Elicited upon Infection x

14 Bystander Memory CD8 T Cell Responses Provide Protection in IFN-γ Deficient, but not Immuno-competent Hosts Discussion CHAPTER IV: DISCUSSION AND FUTURE PERSPECTIVES Changes in Memory CD8 T Cells with Time after Infection Genes and Proteins Regulating Memory CD8 T Cell Function Changes in Phenotype and Function of Multiply Stimulated Memory CD8 T Cells with Time after Infection Regulation of Antigen-dependent and independent Memory CD8 T Cell Activation Expression of Cytokine Receptors and Regulation of Cytokine Signaling Protection Provided by Bystander CD8 T Cell Responses Regulation of Antigen-independent Functions of Memory CD8 T Cells that have Encountered Antigen Multiple Times Outbred Mice as a Model to Study CD8 T Cell Responses REFERENCES xi

15 LIST OF TABLES Table 1: Functional Annotation of Genes Differentially Expressed between CD62L hi earlym and latem CD8 T Cells...85 xii

16 LIST OF FIGURES Figure 1: Model of the CD8 T Cell Response Following 1 o Infection and/or Vaccination...26 Figure 2: Model of Two Modes of Memory CD8 T Cell Activation...28 Figure 3: Effects of Multiple Ag Encounters on CD8 T Cell Properties...30 Figure 4: Localization, Phenotype, and Function of Memory CD8 T Cells Changes with Time after Infection...71 Figure 5: Proliferation and Memory Generation Potential in Response to Agreencounter Increases with Time after Infection...73 Figure 6: With Time, Memory CD8 T Cells Convert to CD62L hi and Become Enriched for T DIM S...75 Figure 7: Phenotype of CD62L hi Memory CD8 T Cells Changes with Time after Infection...77 Figure 8: Phenotypic Heterogeneity of Memory CD8 T Cells Decreases with Time after Infection...79 Figure 9: Phenotypic Heterogeneity of Endogenous and P14 Memory CD8 T Cells Decreases with Time after Infection...81 Figure 10: Gene Expression Patterns among CD62L hi Memory CD8 T Cells Change with Time after Infection...83 Figure 11: Cytokine Production, Degranulation, and Functional Avidity of CD62L hi Memory are not Influenced by Time...89 Figure 12: CD62L hi earlym and latem Cells Exhibit a Similarly Poised State for Effector Responses following Ag Re-encounter...91 Figure 13: CD62L hi earlym and latem Cells Regulate Expression of Surface/Activation Markers Similarly following Ag Re-encounter...93 Figure 14: CD62L hi earlym and latem Cells Regulate Expression of Transcription Factors Similarly following Ag Re-encounter...95 Figure 15: CD62L hi earlym and latem Cells Regulate Expression of Cell Cycle/Survival Genes Similarly following Ag Re-encounter...97 Figure 16: Cell Cycle Pathways are Differentially Regulated in CD62L hi Memory CD8 T Cells with Time...99 Figure 17: CD62L hi latem Cells Enter Cell Cycle Faster than earlym Cells following Ag Re-encounter xiii

17 Figure 18: Sensitivity to IL-15 and Ability to Undergo Homeostatic Proliferation for CD62L hi Memory Changes with Time Figure 19: Proliferation and Memory Generation Potential of CD62L hi Memory Cells Increases with Time after Infection Figure 20: Ability to Undergo Ag-driven Proliferation Increases in Endogenous Memory CD8 T Cells with Time after Infection Figure 21: latem Cells Provide Better Protection than earlym Cells following LCMV Clone-13 Infection Figure 22: Mitochondrial Function of Memory CD8 T Cells Increases with Time after Infection Figure 23: Expression of Components of the IL-12 and IL-18 Receptors Decreases with Time after Infection Figure 24: Ability to Undergo Bystander Activation Decreases with Time after Memory Generation Figure 25: Memory CD8 T Cell Ability to Respond to Inflammatory Cytokines Decreases with Time after Infection Figure 26: Ag and Inflammation Act Synergistically in vitro to Induce Memory CD8 T Cell Activation Figure 27: Synergistic Effects of Ag and Inflammation Upon Memory CD8 T Cell Activation Become more Pronounced with Time after Infection Figure 28: Cognate Ag Does Not influence Early Activation of Memory CD8 T Cells that Do Not Significantly Contribute to Clearance of Infection Figure 29: Ag and Continued Infection Promote Early Activation and Prolonged Responses by Memory CD8 T Cells that Contribute to Clearance of Infection Figure 30: Ag Influences Early Activation of Memory CD8 T Cells that Contribute to Clearance of Infection Regardless of LM Virulence Figure 31: Magnitude of Bystander Memory CD8 T Cell Responses Correlates with the Level of Infection and Inflammation Elicited upon Infection Figure 32: Bystander Responses by Memory CD8 T Cells Provide Protection Against LM in IFN-γ Deficient, but not IFN-γ Sufficient Hosts xiv

18 Figure 33: Bystander Memory CD8 T Cell Responses Fail to Provide Protection Against High-dose LM Infection Figure 34: Regulation of Memory CD8 T Cells following Multiple Ag Encounters and with Time after Infection- a Model Figure 35: Phenotype and Proliferative Abilities of Multiply Stimulated Memory CD8 T Cells Change with Time after Infection, but at a Slower Rate than for 1 o Memory CD8 T Cells Figure 36: Memory CD8 T Cell Ability to Sense Inflammation Increases with Additional Ag Encounters but Decreases with Time after Infection Figure 37: Vaccination Provides Protection to Outbred Mice, but Success is Less Predictable than for Inbred Mice xv

19 LIST OF ABBREVIATIONS Abs ActA Ag AP Att BFA BrdU CCR CD CDK CFSE CFU CXCL CXCR d DCs earlym ECAR ELISPOT FCCP FCS FMO GKO GP GrB GSEA HIV HSV i.p. i.v. ICS IFN IL KLRG1 latem LCMV LM MAPK MHC MIP NFAT NFKB OCR p.i. Antibodies Actin-assembly-inducing protein Antigen Activator protein Attenuated Brefeldin A Bomodeoxyuridine Chemokine receptor Cluster of differentiation Cyclin-dependent kinase Carboxyfluorescein succinimidyl ester Colony forming units Chemokine (C-X-C motif) ligand CXC chemokine receptor Day Dendritic cells Early memory Extracellular acidification rate Enzyme-linked immunospot Fluorocarbonyl cyanide phenylhydrazone Fetal calf serum Fluorescence minus one Gamma knockout Glycoprotein GranzymeB Gene set enrichment analysis Human immune-deficiency virus Herpes simplex virus Intraperitoneally Intravenous Intracellular cytokine staining Interferon Interleukin Killer-cell lectin like receptor G1 Late memory Lymphocytic choriomeningitis virus Listeria monocytogenes Mitogen-activated protein kinase Major histocompatibility complex Macrophage inflammatory protein Nuclear factor of activated T-cells Nuclear factor kappa-light-chain-enhancer of activated B cells Oxygen consumption rate Post infection xvi

20 PBL PBMCs PBS PE PfSPZ PFU Rag RANTES SRC STAT TCF T cm TCR T DIM S T em Tg TNF T rm TSB VacV Vir VSV Peripheral blood lymphocytes Peripheral blood mononuclear cells Phosphate-buffered saline Phycoerythrin Plasmodium falciparum sporozoites Plaque-forming units Recombination-activating gene Regulated on activation, normal T expressed and secreted Spare respiratory capacity Signal transducer and activator of transcription T-cell factor T-central memory T-cell receptor T-death intermediate memory T-effector memory Transgenic Tumor necrosis factor Tissue-resident memory Tryptic soy broth Vaccinia virus Virulent Vesicular stomatitis virus xvii

21 CHAPTER I: INTRODUCTION The CD8 T Cell Response to Infection and/or Vaccination CD8 T cells are specialized cells of the adaptive immune system that play a critical role in combating infections caused by intracellular pathogens including viruses, certain bacteria, and protozoan parasites [1]. The immune system must be prepared to respond to a wide range of infectious microorganisms. Therefore, the number of naïve CD8 T cells that bear T cell receptors (TCRs) recognizing an individual pathogen-derived epitope are low. Recent tetramer enrichment experiments have indicated that these numbers range from as few as 10 to as many as several thousand cells in laboratory mice [2,3], numbers far too low to provide protection to the host from a rapidly spreading infection. However, during infection CD8 T cells mount a pathogen-specific response that is initiated when dendritic cells (DCs) capture foreign antigen (Ag), migrate to draining lymph nodes, and present pathogen-derived Ags to naïve CD8 T cells [4]. Interactions with Ag-presenting DCs in the lymph nodes provide signals 1 (TCR stimulation), 2 (co-stimulation), and 3 (inflammatory cytokines) to naïve CD8 T cells initiating a period of robust proliferation where numbers of Ag-specific effector CD8 T cells increase in numbers by as much as 50,000-fold over a period of 7-8 days (Figure 1) [5,6]. During the vigorous expansion phase, effector CD8 T cells undergo a period of differentiation, acquiring the ability to migrate to infected tissues, to produce inflammatory cytokines including interferon (IFN)-γ and tumor necrosis factor (TNF)-α, and to lyse infected cells, responses which help to eliminate the invading pathogen [7]. Following the expansion phase, CD8 T cells undergo a period of programmed contraction that can be independent of pathogen clearance where 90-98% of the Ag-specific CD8 T 1

22 cells are eliminated through apoptosis [8]. The pool of cells that remain following contraction constitutes the memory CD8 T cell population that is capable of responding and providing enhanced protection during re-infection with the same pathogen. Characteristics of Memory CD8 T Cells What characteristics of memory CD8 T cells account for increased protection compared to naïve CD8 T cells? To answer this question one must take into account I) differences in the numbers of naïve and memory CD8 T cells present in an intact host, and II) potential functional differences between naïve and memory CD8 T cells when these cells are compared on a per-cell basis. Perhaps the biggest difference is seen at the population level, as the precursor frequency of Ag-specific cells is much greater for the memory than the naïve CD8 T cell population of the same Ag specificity. This large pool of memory CD8 T cells is also capable of fast and vigorous secondary expansion following re-infection resulting in a large population of secondary effector CD8 T cells that is able to quickly counter invading microbes [9,10]. On the cellular level, memory CD8 T cells may begin to proliferate more quickly than naïve CD8 T cells as they accumulate pre-activated cyclin D3/cyclin-dependent kinase (CDK) 6 complexes in the cytoplasm and contain lower levels of p27kip inhibitors allowing them to remain poised in the late G1 phase of the cell cycle [11,12]. However, recent work questioned these assumptions showing that naïve CD8 T cells have a lower antigen threshold requirement for cell cycle entry than central memory CD8 T cells. 2

23 Central memory CD8 T cells did not activate Zap70, induce cmyc expression, or degrade p27 in response to Ag levels that triggered these responses in naïve CD8 T cells [13]. Additionally, following primary infection, the memory CD8 T cell population includes cells that reside in peripheral tissues where they can encounter infection faster than naïve CD8 T cells localized predominantly in secondary lymphoid organs [14,15]. In contrast to naïve cells, memory CD8 T cells also can quickly relocate to peripheral tissues in response to inflammation associated with infection [16]. Core 2 O-glycan expression on memory CD8 T cells recently was shown to be altered in response to IL- 15, facilitating interactions with P- and E-selectins and migration into inflamed tissues during infection [17]. Finally, memory CD8 T cells are able to execute effector functions, such as cytokine production and cytolysis, more quickly than naïve CD8 T cells, requiring only a brief 5-6 hour period of stimulation to elaborate effector functions [18-20]. Faster effector responses by memory CD8 T cells are facilitated by preformed stores of cytolytic molecules, enhanced TCR-proximal signaling, and permanent and heritable chromatin remodeling allowing for easier access to gene transcription machinery [21-24]. All together, these characteristics comprise the hallmark attributes of protective memory CD8 T cells. Memory CD8 T Cell Activation and Effector Functions CD8 T cell effector mechanisms are designed to clear the host of infecting microbes. Memory CD8 T cells are able to execute effector functions more rapidly than 3

24 naïve CD8 T cells, and responses can occur in response to reduced Ag levels and are less dependent upon co-stimulatory signals [25,26]. Effector mechanisms are primarily organized upon two fronts, one that results in apoptosis of infected host cells, and another that stimulates the activity of cells of the immune system through cytokines. CD8 T cell cytolysis of infected cells can occur through several pathways. One pathway involves upregulation of FasL on CD8 T cells, which can initiate programmed cell death by aggregation of Fas on infected target cells [27]. The granule exocytosis pathway constitutes another mode of CD8 T cell-mediated killing, in which perforin facilitates the delivery of cytolytic granzyme molecules to infected cells, initiating apoptosis of target cells [28-30]. Activation of CD8 T cells also results in the production and release of cytokines including IFN-γ, TNF-α, and IL-2 that stimulate the microbicidal activity of innate immune cells, lead to increased expression of major histocompatibility complex (MHC) I on target cells, and can act upon adaptive immune cells to regulate adaptive immune responses [7,31-34]. CD8 T cells can also produce chemokines including RANTES, MIP-1α, and MIP-1β that aid in the recruitment of immune cells to sites of infection [35]. CD8 T cell activation classically is thought to be driven by TCR-mediated signaling events (Figure 2). Following infection, short pathogen-derived peptides (8-10 amino acids) that are presented by MHC I on the surface of antigen-presenting cells are recognized by the TCR present on CD8 T cells [36]. Recognition of cognate antigen through the TCR activates downstream signaling cascades resulting in the activation of the Ras-MAP kinase pathway, the protein kinase C pathway, and the calcineurin pathway [37]. Stimulation of these pathways leads to the activation of transcription factors 4

25 including NFAT, AP1, and NFκB that regulate expression of genes necessary for CD8 T cell differentiation, proliferation, and effector functions. It has also become apparent in recent years that memory CD8 T cells can be activated in an Ag-independent manner (Figure 2). Ag-independent activation of effector and memory CD8 T cells has been observed in vitro following incubation with cytokines, and ex vivo following infection with unrelated pathogens (bystander activation) in mice [38-40] and in humans [41-43], and in response to danger-associated molecular patterns including lipopolysaccharide and poly (I:C) [44,45]. Cytokine signaling appears to be the driving force behind Ag-independent CD8 T cell activation, as incubation of memory CD8 T cells with combinations of cytokines including IL-2, IL-12, IL-15, and IL-18 was first shown to induce Ag-independent CD8 T cell IFN-γ production [46], and neutralization of IL-12 and IL-18 following non-specific bacterial infection was shown to ablate bystander CD8 T cell responses [47]. Furthermore, activation of signal transducer and activator of transcription (STAT) 4 downstream of IL-12 signaling recently was shown to be critical for bystander memory CD8 T cell responses following non-related viral infection [48], and memory CD8 T cells generated following chronic Lymphocytic choriomeningitis virus (LCMV) infection display reduced expression of components of the IL-12 and IL-18 receptors and are less responsive to activation induced by IL-12 and IL-18 or during non-related bacterial infection [49]. While the combination of IL-12 and IL-18 causes the most robust cytokine-driven CD8 T cell activation, it was recently shown that many cytokine combinations can elicit Ag-independent CD8 T cell IFN-γ production [50]. Additionally, while IFN-γ is the primary effector molecule produced following non-ag driven activation, expression of the activation markers CD25 and 5

26 CD69 and granzymeb (GrB) also have been shown to increase in response to nonspecific infection and following incubation with IL-15 [39]. Ag-independent activation of memory CD8 T cells during non-specific infection could provide the host with a protective benefit, and studies have shown that IFN-γ produced by memory CD8 T cells transferred into IFN-γ knockout mice provide protection [38,40]. However, it is less clear whether bystander responses by memory CD8 T cells provide a protective benefit to hosts with a full complement of IFN-γ producing cells. I explore this question in Chapter III of this thesis. Memory CD8 T Cell Heterogeneity While memory CD8 T cells possess different properties than naïve CD8 T cells, the memory population is comprised of a heterogeneous group of cells differing from one another in phenotype and functional characteristics [51,52]. Subsets of memory cells have been described based on expression of surface molecules that confer functionality including the ability to survive, to traffic, and to localize within tissues. Classically, memory CD8 T cells have been divided into two subsets, effector memory (T em ) and central memory (T cm ). T cm express chemokine receptor 7 (CCR7) and L-selectin (CD62L), which allow efficient trafficking to lymph nodes, while T em cells do not express these molecules and localize more efficiently to peripheral tissues [53,54]. While T em and T cm are both efficient at producing IFN-γ and TNF-α, T cm are better equipped to produce IL-2, have a greater capacity to persist in the host, and undergo higher 6

27 magnitudes of proliferative expansion upon Ag re-encounter [55]. Therefore, T cm may be important in controlling prolonged and/or systemic infection. In contrast, localized infections in peripheral organs may be better handled by another subset of memory CD8 T cells termed tissue-resident memory cells (T rm ). T rm cells have been identified within tissues based on expression of the α-chain of α E (CD103) β7 integrin and CD69 [56,57], and by intravascular staining to discriminate between circulating cells and cells localized within tissue parenchyma [58]. The classical definition of these cells suggests that T rm cells do not recirculate, but are instead permanent and long-lasting residents of peripheral tissues [15]. They have been reported to exist within the skin [59-61], the gut [56,57], the female reproductive tract [62,63], the brain [64], and within secondary lymphoid organs [65], and they have been shown to provide protection in the skin and female genital tract in response to infection with herpes simplex virus [59,60]. They can also stimulate localized immune responses by producing cytokines and chemokines including IFN-γ and CXCL9 that attract additional immune cells to the site of infection [62]. Thus, T rm cells represent a subset of memory CD8 T cells that are important for providing protection in barrier tissues. Heterogeneity within the memory CD8 T cell population is likely much broader than the T em, T cm, and T rm classifications. Memory CD8 T cells have been divided into additional subsets based on expression of surface receptors including a member of the TNF-receptor family (CD27) and a glycosylated form of sialophorin (CD43), and more recently based on expression of the fractalkine receptor (CX3CR1) [66,67]. CD27 lo CD43 lo populations display similar expression of transcription factors and surface markers compared to T em and T rm subsets, including low expression of CD62L, CD122, 7

28 and CXCR3, intermediate expression of Eomes, and high expression of KLRG1 and T- bet, but also unique expression patterns including high expression of CD127. Despite a reduced ability to proliferate, CD27 lo CD43 lo cells provided better protection against infection with Listeria monocytogenes (LM) due to the ability to localize to the red pulp of the spleen [68]. As a further delineation of memory CD8 T cells, CX3CR1 expression has recently been used to differentiate memory CD8 T cells that have enhanced cytotoxic functions compared to cells not expressing CX3CR1. The representation of CX3CR1+ cells within the Ag-experienced CD8 T cell compartment is greatly reduced in mice and humans under conditions of chronic viral infection, suggesting that this subset of memory CD8 T cell may be important in providing protection against viruses that cause chronic infection [67]. Thus, successful vaccination strategies must take into account whether sufficient numbers of memory CD8 T cells of the appropriate quality will be generated to counter infection. The Process of Memory CD8 T Cell Differentiation Another important consideration for the design of protective vaccines is that differentiation from an effector to memory CD8 T cell is a process that takes considerable time. Gene expression profiles of CD8 T cells continue to change during the transition from effector to memory cell, and a period of time is required for CD8 T cells to acquire characteristics of memory including the ability to self-renew and proliferate in response to Ag [69]. However, the period of time required for memory differentiation is not fixed, and levels of inflammation present during the time of either infection or vaccination appear to strongly influence the size of effector CD8 T cell responses and the size of the 8

29 resulting memory CD8 T cell pool, as well as the rate at which CD8 T cells acquire memory characteristics [70,71]. The cytokines IL-2, IL-12, and type I interferons appear to be highly important in regulating the size of the effector response and the rate of acquisition of memory characteristics [72-76]. Recently, IL-12 and type I interferons were shown to regulate expression of the high affinity subunit of the IL-2 receptor (CD25 or IL-2Rα). Sustained expression of CD25 due to signals transmitted by IL-12 and type I interferons increased sensitivity to IL-2 and allowed for extended division and increased accumulation of activated CD8 T cells [76]. Furthermore, my laboratory in collaboration with the laboratory of John Harty recently demonstrated that IL-2 signaling regulates secondary CD8 T cell responses. Late entry of primary memory CD8 T cells into the immune response impacted the duration of CD25 expression on secondary effector cells, and blocking or enhancing IL-2 signaling altered secondary memory differentiation. [77]. Limiting the duration of bacterial infection through antibiotic treatment (leading to lower levels of inflammation) resulted in the generation of CD8 T cells displaying memory characteristics, such as the ability to undergo vigorous secondary expansion and increased IL-2 production upon Ag re-encounter, within two weeks [78]. Additionally, vaccination using peptide-coated mature DCs that elicit low levels of inflammation resulted in the formation of memory CD8 T cells that could be boosted as soon as 4 days following DC priming [79,80]. Administration of CpG to induce systemic inflammation reversed this accelerated memory differentiation after either DC immunization or antibiotic treatment [78-80]. Thus, inflammation can greatly impact numbers of memory CD8 T cells generated and the rate of memory development following infection and/or 9

30 vaccination. This is an important consideration for the design of vaccines utilizing inflammation-inducing adjuvants. Maintenance and Longevity of Primary Memory CD8 T Cells While most vaccines are intended to prevent either seasonal illnesses or infections that may be encountered in the relatively near future, infection may not occur for long periods of time following the original vaccination. Thus, it is important to establish if CD8 T cell memory that is formed after vaccination is maintained throughout life, and how the function of memory CD8 T cells change over extended lengths of time. These questions are easier studied in animal models where re-exposure to infection can more easily be controlled. However, the durability of CD8 T cell memory following either infection or vaccination in humans has been studied in the context of several acute viral infections that only cause rare infections (measles virus), or are not endemic (vaccinia virus and yellow fever virus) [81,82]. Unlike naïve cells, memory CD8 T cells do not require either TCR signaling or MHC class I molecules for long-term survival [83,84]. Instead, memory CD8 T cells are maintained in vivo through a slow process of basal turnover that is dependent on the prosurvival cytokines IL-7 and IL-15 [85]. Decreased percentages of IL-7Rα deficient CD8 T cells survive and differentiate into memory cells following infection, and memory CD8 T cells generated in either IL-15 or IL-15Rα deficient mice slowly decline in numbers [86,87]. In order to maintain stable numbers of memory CD8 T cells, basal turnover, which results in two daughter cells, must be balanced by an equivalent rate of cell death. 10

31 Recent work from my laboratory in collaboration with the laboratory of John Harty has indicated that basal turnover results in the formation of a subset of memory CD8 T cells termed T death intermediate memory (T DIM ) which are non-functional (i.e. are unable to produce cytokines or internalize TCRs following stimulation with Ag, are destined to die, and presumably serve to keep numbers of memory CD8 T cells stable during basal turnover [88]. Thus, basal proliferation allows the memory CD8 T cell population to be maintained at relatively stable numbers for great lengths of time. Homann et al. examined the longevity of epitope specific memory CD8 and CD4 T cell populations in mice following LCMV infection [89]. Numbers of memory CD4 and CD8 T cell populations were determined in peripheral blood (PBL) by ELISPOT and MHC class I or class II tetramers at various time points for over 900 days following LCMV infection. While numbers of Ag-specific memory CD4 T cells were found to decline with time after infection, numbers of each of the six analyzed epitope-specific memory CD8 T populations remained constant. Memory CD8 T cells were found to retain expression of the anti-apoptotic protein Bcl2 to a greater extent than memory CD4 T cells, potentially leading to an increased resistance to apoptosis. In agreement with the longevity of LCMV-specific CD8 T cell memory, a recent report by Valkenburg et al. examined memory CD8 T cell generation in response to infection with influenza virus [90]. Detectable numbers of memory CD8 T cells specific for two different epitopes of influenza virus were found by intracellular cytokine staining (ICS) greater than 22 months after infection with the H3N2 strain of influenza virus. Collectively, these and other studies indicate that primary CD8 T cell memory established early in life can be maintained for the life of the mouse. 11

32 While mice can be housed in specific pathogen-free facilities, humans are repeatedly exposed to multiple unrelated pathogens. Since the size of the memory CD8 T cell pool is thought to be limited by constraints of space and availability of survival factors such as IL-7 and IL-15, exposure to unrelated infections could lead to attrition of memory CD8 T cell populations. Schmidt et al. found that numbers of memory CD8 T cells specific for a malaria circumsporozoite protein were sharply reduced following infection with four unrelated bacteria and viruses [91]. Additionally, Selin et al. found that LCMV-specific memory CD8 T cells undergo substantial attrition in mice following one or more heterologous virus challenges [92]. Further experiments by Varga et al. determined that attrition of LCMV-specific memory CD8 T cells was independent from CD4 T cells, as numbers of LCMV-specific CD4 T cells remained stable while numbers of memory CD8 T cells declined following heterologous challenges [93]. However, a study by Vezys et al. found that the memory CD8 T cell compartment can grow in size following sequential infections to allow maintenance of pre-existing memory populations. [94]. While they observed attrition of LCMV-specific memory CD8 T cells following a series of three heterologous infections with vesicular stomatitis virus (VSV) strains, the severity of attrition was more modest than indicated by earlier studies [94]. However, these studies indicate that memory CD8 T cells in humans, who are repeatedly exposed to non-related infections, may decrease in number, which could lead to loss of CD8 T cell mediated protection. A number of studies, however, have indicated that memory CD8 T cells in humans are detectable for many years following either infection or vaccination. Ahmed and Akondy have reported that following a live virus vaccination against yellow fever 12

33 (YFV-17D), memory CD8 T cells could be detected for decades following vaccination [95]. Similarly, peripheral blood mononuclear cells (PBMCs) from vaccinia virus (VacV) vaccinated individuals were able to lyse VacV infected target cells 30 or more years after original vaccination [96]. Additionally, VacV specific CD8 T cells persisted for long periods of time following natural exposure to smallpox or after vaccination [97,98]. Detection of IFN-γ- and TNF-α- producing CD8 T cells by ELISPOT suggested that VacV-specific memory CD8 T cells declined with time after vaccination with a half-life between 8 and 15 years. However, 50% of individuals receiving one vaccination possessed detectable CD8 T cell memory at least 20 years after vaccination, and at least one individual possessed memory CD8 T cells 75 years after vaccination [98]. Naniche et al. have also reported that measles virus-specific CD8 T cells were detectable by ICS in individuals up to 34 years after vaccination [99]. Thus, while numbers of memory CD8 T cells may decline with time after infection and/or vaccination in human subjects, these cells can be maintained for the life of the host. Changes in Primary Memory CD8 T Cell Properties with Time after Infection Differentiation of memory CD8 T cells from the effector population requires time, and it appears that memory differentiation is a process that continues for great lengths of time following infection. While a limited number of experiments have examined the functional changes that occur in memory CD8 T cell populations with time after infection, this question remains largely underexplored (Figure 1). This is an important unresolved question, as protection offered by memory CD8 T cells is based both on quantity (numbers) and quality (functional ability) of these cells at the time of re- 13

34 infection. Thus, changes in memory CD8 T cell function that occur with time after infection could directly impact their ability to provide protection from re-infection. Expression of phenotypic markers including CD62L and CD27 has been shown to increase in the memory population with time after infection [100,101]. Based on these markers, the memory population becomes comprised of primarily T cm cells, suggesting that the overall function/responsiveness of memory CD8 T cells may change with time after infection. Jabbari et al. showed that the ability of memory CD8 T cells to produce IL-2, a function that is better suited to T cm compared to T em cells, increased with time up to 80 days after infection with L. monocytogenes. T cm also exhibit an increased ability to undergo proliferative expansion compared to T em cells, and Roberts et al. examined if the ability of memory CD8 T cells to proliferate following Ag re-encounter changes with time after infection [102]. In elegantly designed experiments, they examined proliferative potential of aged (12 months after infection) and recent (1 month after infection) Sendai virus-specific memory CD8 T cells. They found that aged memory proliferated to a greater extent than recent memory indicating that the ability of memory CD8 T cells to proliferate in response to Ag re-encounter increases with time after infection. Similarly, in work that I conducted while a Master s student, using TCR transgenic OT-I cells which recognize the OVA epitope derived from chicken ovalbumin, I generated early memory (1 month), and late memory (8 months) after infection with L. monocytogenes expressing OVA and reported that the ability of memory CD8 T cells to proliferate in response to secondary Ag encounter increased with time after infection [10]. Importantly, higher proliferative potential but indistinguishable kinetics of secondary CD8 T cell responses generated from late versus early primary memory CD8 T 14

35 cells led to an increase in secondary memory CD8 T cell numbers, suggesting that memory generation potential of primary memory CD8 T cells is also dependent on the age of the cells. Thus, the limited amount of data on the functional changes that occur over time in the memory CD8 T cell population indicates that memory CD8 T cells regain expression of surface markers including CD27 and CD62L, have an increased capacity to produce IL-2, proliferate more robustly in response to Ag re-encounter, and have a greater memory generation potential. These data obtained in mice suggest the exceptional possibility that the memory CD8 T cell pool continues to differentiate for extended periods of time. The extent to which memory CD8 T cells change with time in humans is not known, but it would be interesting to determine if the process of memory CD8 T cell differentiation continues for great lengths of time in long-lived humans. Additionally, how the functional changes that occur in memory CD8 T cells with time after infection affect their ability to confer protection from infection is largely unknown. The characteristics of memory CD8 T cells that confer protection to some pathogens may not lead to protection from others [103]. Thus, the ability of memory CD8 T cells to confer protection may increase with time for some infections and decrease with time for others. Importantly, the success of booster immunization strategies may be dependent upon the changes in memory CD8 T cells that occur over the period of time between boosts. Recently, utilizing a vaccine strategy involving intravenous injection of cryopreserved Plasmodium falciparum sporozoites (PfSPZ), Seder et al. reported that all human subjects receiving five injections were protected, while not all subjects receiving four or fewer injections were protected upon controlled human malaria infection [104]. 15

36 The authors indicated a number of potential reasons for the increased protection provided using the 5 dose immunization regimen including that these subjects had received the highest dose of PfSPZ, and that there was an increased interval (7 weeks) between the fourth and fifth booster challenges. They argued that the increased length of time between the fourth and fifth dose may have led to greater numbers of CD8 T cells than would have been achieved using a shorter time frame between doses. Thus, changes that occur within the memory CD8 T cell population over periods of time between boosts may directly impact protection achieved through booster immunizations. For these reasons, a better understanding of the full spectrum of changes that occur in the memory CD8 T cell population with time and the implications of these changes for conferring protection from re-infection is needed. This question is addressed in Chapter II of this thesis. Effects of Additional Ag Encounters on Memory CD8 T Cell Properties Humans are often infected multiple times with the same or related pathogens, leading to the formation of memory CD8 T cell populations that have encountered their cognate Ag more than once. Furthermore, the numbers of memory CD8 T cells required to achieve protection from infection may be higher than can be achieved with a single immunization [105], and prime-boost vaccination strategies designed to elicit large numbers of memory cells result in memory CD8 T cells that have encountered Ag multiple times [106]. Therefore, research in the past several years has been devoted to understanding how the properties of memory CD8 T cells are affected by additional Ag encounters. From these studies it has become clear that the gene expression profiles, 16

37 phenotype, function, and protective abilities of memory CD8 T cells change with each subsequent Ag encounter (Figure 3). Work from my laboratory in collaboration with the laboratories of John Harty and Hai-Hui Xue recently showed that gene expression profiles of memory CD8 T cell populations that have encountered Ag more than once differ from one another. Using a serial adoptive transfer model to generate primary (1 o ), secondary (2 o ), tertiary (3 o ), and quaternary (4 o ) memory OT-I cells that differed only in the number of Ag encounters, it was found that a group of genes representing a memory core signature is differentially expressed in naive compared to memory CD8 T cells regardless of the number of Ag encounters [107]. However, gene expression patterns among memory CD8 T cell populations that had encountered Ag 1-4 times differed from one another, and more than 700 genes were differentially regulated between 1 o and 4 o memory cells. Interestingly, the number of genes that were differentially expressed between memory populations and naive CD8 T cells increased with each additional Ag encounter indicating that repetitive Ag stimulation induces stepwise changes in gene expression patterns. Gene-setenrichment-analysis (GSEA) performed on the list of genes that are up-regulated in T em showed no progressive enrichment in effector memory associated genes in 2 o, 3 o, and 4 o memory CD8 T cell populations, suggesting that each round of Ag stimulation further increases the complexity of memory CD8 T cell populations in a manner that may supersede current T em and T cm classifications. Additionally, expression of many genes in memory CD8 T cell populations continued to either increase (Gzmb, Anxa1, Ccr5) or decrease (Actn1, Ccr7, Treml2) in expression with additional Ag encounters indicating that functional differences may become magnified as memory CD8 T cells encounter Ag 17

38 additional times. Genes that were differentially regulated with additional Ag encounters clustered into families of genes regulating effector functions, signaling, migration, adhesion, cell cycling, and apoptosis. Additionally, the expression levels of several transcription factors including Eomes, Prdm1 (Blimp-1), Tbx21 (T-bet), and Tcf7 (TCF- 1) either progressively increased or decreased with additional Ag encounters [107]. In agreement with the transcriptomic data, the phenotype and function of memory CD8 T cells also was shown to be modified following additional Ag encounters. Expression of surface molecules including CD127, CD62L, CD27, and CD122 decreases with each additional Ag encounter [100,101,107]. Consistent with their low expression of CD62L, localization of memory CD8 T cells in lymph nodes decreases with additional Ag encounters while their localization in peripheral tissues increases with additional Ag encounters [100,101,107]. Furthermore, Ag stimulation history also appears to have an impact on proliferation and memory generation following Ag re-encounter. Using the serial adoptive transfer model, my laboratory in collaboration with the laboratories of John Harty and Hai-Hui Xue reported that the ability of memory CD8 T cells to undergo proliferative expansion following infection decreased with subsequent Ag encounters from 1 o to 4 o memory [107]. Additionally, while the period of contraction was delayed for 2 o, 3 o and 4 o memory, contraction among these populations was eventually more vigorous than 1 o memory. Each subsequent Ag encounter led to reduced expression of CD122 (IL-15Rβ), reduced responsiveness to IL-15, and decreased basal turnover of the memory population [100,107]. This suggests that with subsequent Ag encounters memory CD8 T cells may have a reduced ability to be maintained through basal proliferation. 18

39 CD8 T cells combat infection through the production of cytokines to stimulate the immune response and by killing infected target cells primarily through secretion of perforins and granzymes. The evidence suggests that while the ability of memory CD8 T cells to produce IFN-γ and TNF-α is unaffected by the number of Ag encounters, the ability to produce IL-2 decreases with additional Ag encounters while preformed stores of granzymes and ability to kill infected target cells increase with additional Ag encounters [100,101,107]. The enhanced ability to kill infected cells would lead one to believe that memory CD8 T cells that have encountered Ag multiple times would be more protective. However, because different pathogens utilize distinct anatomical niches in which to replicate, perhaps it is not surprising that memory CD8 T cells that have encountered Ag multiple times are more protective than 1 o memory in some instances and less protective in others. 2 o memory CD8 T cells generated through serial adoptive transfers were found to provide enhanced protection compared to an equal number of 1 o memory cells following acute systemic infection with either L. monocytogenes or VacV despite decreased ability to undergo proliferative expansion [100,103]. However, 1 o memory cells were found to be more protective than an equal number of 2 o memory CD8 T cells following chronic infection with LCMV-clone 13. Because clone 13 infection is more efficiently controlled by CD8 T cells that are able to localize to the lymph nodes, differences in anatomical location may account for decreased protection of 2 o memory cells following this chronic infection. However, 2 o memory CD8 T cells generated through heterologous prime boosting of individual mice were at least as protective as 1 o memory cells after infection with clone 13 [108]. As the above data illustrates, careful consideration must be made regarding the nature of the pathogen and the properties of 19

40 memory CD8 T cells generated through prime boost for the design of effective vaccine strategies. LM and LCMV as Model Pathogens Used to Study the CD8 T Cell Response to Infection Infection with either LM or LCMV elicits a robust CD8 T cell response, and both pathogens have been extensively utilized to experimentally examine the immune response following intracellular bacterial and viral infections. LM has been used to study every phase of the CD8 T cell response to intracellular bacterial infection and has provided insight into the factors that influence the primary T cell responses, memory CD8 T cell generation and maintenance, the functions of memory CD8 T cells important for providing protection from infection, and the influence of repeated infection on the properties of memory CD8 T cells [109,110]. Additionally, investigations using the LCMV model have yielded information regarding MHC restriction of T cell responses, immune tolerance, persistent infections, immune pathology in disease, and T cell exhaustion [111,112]. Listeria monocytogenes is a gram-positive facultative intracellular bacterial pathogen that is the causative agent of listeriosis [113]. It was first used experimentally in a murine model of listeriosis by George Mackaness in the early 1960s [114]. Following i.v. injection of LM, most bacteria can be found in the spleen and liver within minutes where they are quickly internalized by resident macrophages [115]. Once inside the host cell, LM escapes the phagosome by secreting the pore forming toxin listeriolysin O, and 20

41 it then gains access to the host cell cytosol. LM replicates within the cytosol, and mobility within the cytosol and cell-to-cell spread of the bacteria is facilitated by actinassembly-inducing protein (ActA). While responses of innate immune cells including neutrophils and macrophages are initially important for controlling infection [116], cellular immunity was shown by Mackaness and colleagues to be critical for control of the infection in mice [114]. Further experiments showed the importance of cellular immunity in the control of LM infection as SCID mice developed a chronic infection [117,118], and the importance of CD4 and CD8 T cells for clearance was demonstrated in MHC deficient mice [119]. However, in vivo depletion and adoptive transfer studies demonstrated that memory CD8 T cells are most effective at mediating protection against LM infection [120,121]. The cytokine IFN-γ is important for controlling primary LM infection, as mice deficient in IFN-γ are highly susceptible to LM infection [122]. However, IFN-γ appears to be less important for protective immunity against re-infection [123]. Perforin is important for providing immune-mediated protection against LM, as perforin deficient mice have a defect in clearing a challenge infection following immunization [124]. However, increasing the number of memory CD8 T cells was able to overcome diminished protection in the absence of perforin [125], suggesting that perforin-independent pathways also contribute to immune-mediated LM protection. CD8 T cell-mediated responses to LM infection have been widely studied due to the ability of LM to survive and replicate within the infected host cell cytosol and to access the endogenous MHC class I pathway, which results in the induction of robust CD8 T cell responses [110]. However, the LM murine model has a number of additional strengths that have served to advance our knowledge of the CD8 T cell response to 21

42 infection with intracellular pathogens. First, LM infection is very reproducible and bacterial loads in the host can be easily enumerated [116]. Second, LM can be easily modified genetically, and this has allowed deletion of individual virulence factors as well as insertion of genes expressing introduced antigens [109,116]. This has allowed for the study of the importance of specific CD8 T cell effector functions in the mice that would normally be highly susceptible to infection, as well as the study of the importance of memory CD8 T cell-mediated protection following vaccination with attenuated LM strains. Expression of model antigens has also expanded the available tools for studying CD8 T cell responses following infection. Lastly, the duration of LM infection can easily by manipulated with antibiotic treatment, which has allowed for the examination of the influence of inflammation and antigen presentation on CD8 T cell responses following infection. Because of these benefits, the LM model has been widely used to study CD8 T cell responses following infection. There are also a number of benefits to the use of LCMV as a model to study CD8 T cell responses including: 1) The T cell response to LCMV is very robust, 2) LCMV Armstrong is a natural acute infection in mice, 3) Strains of LCMV have been isolated that cause a persistent or chronic infection in mice, allowing for the study of the immune response during chronic infections, and 4) there are a number of tools available to study CD4 and CD8 T cell responses following LCMV infection including TCR transgenic (Tg) mice and MHC tetramers that allow for the detection of endogenous Ag-specific T cell responses [112]. LCMV was first identified in the early 1930s in St. Louis by Armstrong and colleagues following an encephalitis epidemic, and in subsequent years it was classified 22

43 within the Arenaviridae virus family [111,112]. LCMV is an enveloped negative-strand RNA virus with a bisegmented genome and life cycle restricted to the cell cytoplasm [111]. Virions are spherical to pleomorphic, ranging in size from 40 to 200 nm, with a dense lipid envelope and a surface layer covered by spike-like surface structures representing the viral glycoprotein (GP) [111]. The GP1 and GP2 proteins mediate virus interaction with host cell surface receptors, and upon receptor binding, virions are delivered to endosomes where the viral ribonucleoprotein enters the host cell s cytoplasm [126]. Viral replication occurs in both peripheral tissues and secondary lymphoid organs [127]. While responses by innate immune cells including DCs, macrophages, and NK cells are important in combating LCMV infection, it is primarily the virus-specific killing of infected cells through a perforin-mediated pathway by CD8 T cells that is responsible for LCMV clearance There are a number of different ways in which LCMV infection can cause a persistent of chronic infection within mice. The Clone 13 variant of LCMV is commonly used to initiate chronic infection in mice, and the principles learned using this model have been extended to studies in related fields including HIV, hepatitis virus, and cancer [112]. Clone 13 was isolated from the spleen of a carrier mouse that sustained a persistent LCMV infection at birth, and interestingly, the Clone 13 and Armstrong virus genomes differ by only 5 out of over 10,000 nucleotides [112]. Following infection of naïve mice with LCMV Clone 13, CD8 T cells undergo loss of function, where they progressively lose the ability to produce effector cytokines, to self-renew, and to proliferate, while displaying increased expression of inhibitory receptors including PD1, LAG3, and CTLA4 [128]. Memory CD8 T cells generated through vaccination or after acute LCMV 23

44 infection are able to provide protection against LCMV Clone 13 infection, and factors including memory CD8 T cell localization to the lymph nodes, ability to proliferate, and metabolic capabilities are important for their protective abilities following Clone 13 infection [103,108]. Throughout this Thesis I have utilized LM and strains of LCMV resulting in acute and chronic infections as model pathogens to examine the CD8 T cell response to infection, as well as memory CD8 T cell-mediated protection against infection. Thesis Rationale and Objectives Over the past century, advancements in our understanding of the immune system have paved the way for the development of vaccines that protect humans from public health threats including measles, smallpox, polio, influenza, and tetanus. However, the success of these vaccines is primarily due to the establishment of antibody responses by B cells, and it is recognized that the design of vaccines to stimulate protective memory CD8 T cells could reduce public health threats due to influenza, malaria, HIV, and hepatitis virus [104, ]. However, design of successful vaccine platforms intended to elicit protective memory CD8 T cell responses against these pathogens has proven challenging. The studies described in this thesis are aimed at understanding the properties of memory CD8 T cells generated following vaccination, and the characteristics of protective memory CD8 T cells that lead to protection against diverse pathogens. Specifically, these studies had two main focuses: 24

45 1) Determine how the phenotype and function of primary memory CD8 T cells change with time after infection and the consequences for CD8 T cell-mediated immunity, which I discuss in Chapter II. 2) Determine if the ability of memory CD8 T cells to sense inflammation and to become activated in an Ag-independent manner is altered with time after infection, how pathogen derived inflammatory cytokine signals contribute to Agdriven memory CD8 T cell activation following infection, and whether Agindependent memory CD8 T cell responses during unrelated infections (bystander activation) provide the host with a protective benefit, which I discuss in Chapter III. 25

46 Figure 1: Model of the CD8 T Cell Response Following 1 o Infection and/or Vaccination Upon encountering antigen, naïve CD8 T cells differentiate into effector cells and undergo robust proliferative expansion in numbers over a period of 5-10 days. The effector response is followed by a period of contraction, and the cells remaining after contraction constitute the memory CD8 T cell pool. Memory CD8 T cells can be maintained for years, however their properties change over the course of time. The full extent of changes that occur in memory CD8 T cells with time after infection is presently unknown. 26

47 27

48 Figure 2: Model of Two Modes of Memory CD8 T Cell Activation (Top) Memory CD8 T cells can be activated in an Ag-dependent manner when the T-cell receptor recognizes cognate Ag presented on MHC I of a target cell leading to the production and release of cytolytic molecules (perforin and granzymes) and cytokines (IFN-γ, etc.). (Bottom) Memory CD8 T cells can also be activated in an Ag-independent manner, which is driven by inflammatory cytokines, leading to the production of granzymes and IFN-γ. 28

49 29

50 Figure 3: Effects of Multiple Ag Encounters on CD8 T Cell Properties The functional abilities of memory CD8 T cells are altered upon additional Ag encounters. Functional abilities and characteristics shown are highest for memory populations at the outermost edges and progressively decrease for memory populations moving inward (i.e. proliferative expansion 1 o >2 o >3 o >4 o ). 30

51 31

52 CHAPTER II: PHENOTYPIC AND FUNCTIONAL ALTERATIONS IN CIRCULATING MEMORY CD8 T CELLS WITH TIME AFTER PRIMARY INFECTION Abstract Memory CD8 T cells confer increased protection to immune hosts upon secondary viral, bacterial, and parasitic infections. The level of protection provided depends on the numbers, quality (i.e. functional ability), and location of memory CD8 T cells present at the time of infection. While primary memory CD8 T cells can be maintained for the life of the host, the full extent of phenotypic and functional changes that occur over time after initial antigen encounter remains poorly characterized. Here I show that critical properties of circulating primary memory CD8 T cells, including location, phenotype, cytokine production, maintenance, secondary proliferation, secondary memory generation potential, and mitochondrial function change with time after infection. Interestingly, phenotypic and functional alterations in the memory population are not due solely to shifts in the ratio of effector (CD62L lo ) and central memory (CD62L hi ) cells, but also occur within defined CD62L hi memory CD8 T cell subsets. CD62L hi memory cells retain the ability to efficiently produce cytokines with time after infection. However, while it was not formally tested whether changes in CD62L hi memory CD8 T cells over time occur in a cell intrinsic manner or are due to selective death and/or survival, the gene expression profiles of CD62L hi memory CD8 T cells change, phenotypic heterogeneity decreases, and mitochondrial function and proliferative capacity in either a lymphopenic environment or in response to antigen reencounter increase with time. Importantly, and in accordance with their enhanced 32

53 proliferative and metabolic capabilities, protection provided against chronic LCMV clone-13 infection increases over time for both circulating memory CD8 T cell populations and for CD62L hi memory cells. Taken together, the data in this study reveal that memory CD8 T cells continue to change with time after infection and suggest that the outcome of vaccination strategies designed to elicit protective memory CD8 T cells using single or prime-boost immunizations depends upon the timing between antigen encounters. Introduction Memory CD8 T cells provide immune hosts with enhanced protection from pathogenic infection due to an increased precursor frequency of Ag-specific cells, widespread localization to both lymphoid and non-lymphoid tissues, and ability to rapidly execute effector functions such as cytokine production and cytolysis compared to naïve CD8 T cells [1,7,15]. Protection provided by memory CD8 T cells is dependent upon the number, quality (i.e. functional abilities), and location of memory CD8 T cells available at the time of infection. Importantly, the quality and location of memory CD8 T cells best suited to combat diverse infections is dependent upon the tropism of the invading pathogen. Memory CD8 T cells consist of a heterogeneous population of cells [51] that were initially categorized into T cm and T em subsets based on CCR7 and CD62L expression, and that differ in anatomical location and functionality [53,55]. Recently, an additional subset of memory CD8 T cells has been described that reside within tissues and that have been called T rm cells [134]. While the relative protection provided by circulating T cm and T em cells differs depending on the nature of infection [55,68,103,135], 33

54 both are better suited to provide protection against systemic infection than T rm cells that provide enhanced protection against infection that occurs within peripheral tissues [59,60,62,63,136]. Several studies have suggested that T rm cells may be long-lived in the skin following VacV or herpes simplex virus (HSV) infection and in mucosal surfaces following intramuscular immunization with adenovirus vectors [60,136,137]. However, other studies examining T rm generated following influenza virus infection have suggested that T rm cell numbers wane following infection [138]. Therefore, longevity of T rm cells likely depends on the infection/vaccination model and the tissue of memory residence. However, circulating memory CD8 T cells persist for great lengths of time following immunization or systemic viral infection. For example, LCMV-specific memory CD8 T cells are maintained at stable numbers in the spleen for the life of the laboratory mouse [89], and detectable numbers of memory CD8 T cells can be found in human peripheral blood lymphocytes (PBL) years after natural exposure to, or vaccination against yellow fever virus, measles virus, and smallpox [95-99]. However, several studies have indicated that some properties of circulating memory CD8 T cells change with time after infection. For example, expression of CD62L and CD27 (markers of central memory cells) increases, indicating that the subset composition of the memory population changes with time after infection. In addition, functions such as cytokine production, proliferation, and memory generation following Ag re-encounter, increase with time [10, ,139]. The full extent of phenotypic and functional alterations that occur within the memory CD8 T cell population with time after infection, however, remains poorly characterized. Furthermore, it is unclear if alterations are due solely to differences in subset composition of memory CD8 T cell populations, or due to changes within defined memory subsets. 34

55 These are important questions to address, as the level of protection provided against systemic infections may change with time following initial infection and/or vaccination. While most vaccines are intended to elicit protection against seasonal illnesses or pathogens that will be encountered in the relatively near future, infection may not occur for long periods of time following the original vaccination. Therefore, changes in memory CD8 T cell function between vaccination and the time of infection may impact the protection provided by memory CD8 T cells. Furthermore, evidence has suggested that the number of memory CD8 T cells required to provide protection against some pathogens may be quite high [104,105,140]. Currently, the best method for eliciting large numbers of memory CD8 T cells involves prime-boost strategies in which a series of vaccinations are administered allowing for a period of time between boosts [106]. Recently, it was reported that all human subjects receiving five injections of cryopreserved radiation attenuated Plasmodium falciparum sporozoites were protected upon controlled infection with P. falciparum, while not all subjects receiving four injections were protected [104]. A number of potential reasons for the increased protection provided by the five-dose regimen were proposed including a longer interval of time between administration of the fourth and fifth boost. Although not tested in their work, the authors argued that increasing the interval of time between boosts could lead to the establishment of greater numbers of memory CD8 T cells and increased protection compared to immunization strategies using a shorter time interval between boosts. Thus, functional changes in the properties of memory CD8 T cells occurring in the time between boosts may directly impact the protection achieved through prime-boost 35

56 vaccination strategies. For these reasons, an understanding of how memory CD8 T cell quality changes with time after infection and/or vaccination is needed. In this study I examined how the properties of circulating memory CD8 T cells change with time following an acute systemic infection with LCMV. I demonstrate that memory CD8 T cell quality changes with time after Ag-encounter in a manner not solely due to shifts in memory subset composition. Importantly, my data suggest that alterations in memory CD8 T cell function that occur with time after Ag-encounter could impact their ability to provide protection against diverse pathogens, and that the generation of memory CD8 T cells through prime boost protocols may depend on the timing between boosts. Materials and Methods Mice, Bacteria, and Viruses C57BL/6 Thy1.2 mice were obtained from the National Cancer Institute (Frederick, MD). B6/SJL (CD45.1), recombination-activating gene (Rag)-/- mice, and P14 mice were bred at the University of Iowa (Iowa City, IA). The Armstrong strain of LCMV, the clone-13 strain of LCMV, VacV expressing the GP 33 epitope, attenuated acta- deficient Listeria monocytogenes expressing the GP 33 epitope [104], and virulent Listeria monocytogenes strain 1043S expressing the GP 33 epitope were grown and quantified as previously described [10,103]. Briefly, stock solutions of bacteria were incubated for 2-3 hours in tryptic soy broth (TSB) at 37 o C with light shaking. Bacterial colonies/ml were determined based upon OD readings at 600nm using a Genesys 10UV 36

57 (Thermo Scientific). Bacteria were re-suspended in saline solution at the desired colonies /ml, and 200µL was intravenous (i.v.) injected per mouse. Stock viral strains were diluted to the desired concentration in phosphate buffered saline (PBS), and 200µL was intraperitoneally (i.p.) injected per mouse. Adoptive Transfers and Generation of earlym and latem P14 and Endogenous Memory Cells For generation of earlym and latem P14 cells, P14 CD8 T cells (specific for the LCMV GP epitope) were isolated from the peripheral blood of young Thy1.1/1.1 or Thy1.1/1.2 P14 mice. Contaminating memory phenotype (CD11a hi /CD44 hi ) P14 cells were always <5%. 5x10 3 P14 cells were transferred retro-orbitally into 6-12 week old naïve C57BL/6 mice, and recipients were infected 24 hours later i.p. with 2x10 5 plaque forming units (PFU) of LCMV Armstrong. All earlym analysis was done between days after infection and latem analysis was done 8+ months after infection. For cotransfer of earlym and latem P14 cells, P14 cells were isolated from the spleens of mice containing Thy disparate earlym and latem P14 cells, mixed at a 1:1 ratio, and 2x10 4 of each were transferred retro-orbitally into naïve C57BL/6 mice followed 24 hours later by i.p. injection of 2x10 5 PFU of LCMV or 3x10 6 PFU of VacV, or by i.v. injection of 5x10 6 colony forming units (CFUs) of LM. For co-transfer of CD62L hi earlym and latem P14 cells, P14 cells were isolated from the spleens of mice containing Thy disparate earlym and latem P14 cells, and cells were surface stained for Thy1.1, CD8, and CD62L (ebioscience) and sorted using a BD FACSAria II (BD Biosciences). Sorted cells were 37

58 mixed at a 1:1 ratio and 1x10 4 of each were transferred retro-orbitally into naïve C57BL/6 mice followed 24 hours later by i.p. injection of 2x10 5 PFU of LCMV. Sorted cells were >95% pure. Input ratios of earlym and latem P14 cells were confirmed by flow cytometry before adoptive transfer. For adoptive transfer of CD62L hi endogenous GP 276 cells, splenocytes of earlym and latem C57BL/6 (CD45.2) mice were stained with phycoerythrin (PE)-anti-CD8 antibodies and purified with anti-pe magnetic bead sorting using standard AutoMacs protocols. Cells were then stained with CD62L, and CD62L hi cells were sorted using a BD FACSAria II (BD Biosciences). Following sorting, cells were stained with GP 276 tetramer to determine the percentage of endogenous GP 276 memory cells in the sorted CD62L hi CD8 T cell population. 2.5x10 3 endogenous CD62L hi earlym or latem GP 276 tetramer positive cells were then transferred into CD45.1 B6/SJL mice followed 24 hours later by i.p. injection of 2x10 5 PFU of LCMV. Lymphocyte Isolation, Quantification of CD8 T Cell Responses and Surface Marker Expression, ICS and Degranulation Analysis, Transcription Factor Staining, and KI67 Staining EarlyM and latem CD8 T cell responses were quantified in peripheral blood by collecting blood via retro-orbital puncture. Red blood cells were lysed with ACK, and P14 cells were surface stained for Thy1.1, Thy1.2, and CD8 (ebioscience), and endogenous memory cells were surface stained for Thy1.1, CD45.2, CD8 (ebioscience), and GP 276 tetramer. Cells were acquired on a FACSCalibur instrument (BD Biosciences), and Thy expression was used to distinguish between earlym and latem P14 cells. For isolation of lymphocytes from tissues, anesthetized mice were perfused through the left 38

59 ventricle with cold PBS for 1-2 minutes and single-cell suspensions from the lung, spleen, liver, and inguinal lymph nodes were prepared by mashing tissues over a metal filter [10]. Surface marker expression and heterogeneity among earlym and latem P14 cells was determined by 8 color staining of isolated lymphocytes for Thy1.1, CD8, CD27, CD122, killer-cell lectin like receptor G1 (KLRG1), CD11b (ebioscience), CD62L, and CD127 (biolegend), or 8 color staining of isolated lymphocytes for Thy1.1, CD8, CD27 (ebioscience), CXCR3, CD127, CCR5, and CD43 (biolegend). Surface marker expression and heterogeneity among endogenous GP 33 and GP 276 earlym and latem cells was determined by 8 color staining of isolated splenocytes for GP 33 or GP 276 tetramer, Thy1.1, CD8, CD27, CD122, KLRG1 (ebioscience), CD62L and CD127 (biolegend). Cells were acquired on a LSR II instrument (BD Biosciences), and gates were set using fluorescence minus one (FMO) staining. Ex vivo cytokine detection [141] was determined by mixing splenocytes containing earlym and latem P14 cells and incubating with 200nM GP peptide at 37 o C for 5 hours in the presence of Brefeldin A (BFA) for 5 hours or 1 hour. To prevent CD62L cleavage, cells were pre-incubated for ½ hour in the presence of 100µM TAPI-2 (Peptides International) [142]. To determine functional avidity, splenocytes were incubated as described above in the presence of the indicated concentrations of GP peptide [143]. To determine the time required to produce IFN-γ, TNF-α, or IL-2, splenocytes containing earlym and latem P14 cells were mixed together and incubated as described above in the presence of 200nM GP peptide for the indicated periods of time. Cells were surface stained for Thy1.1, Thy1.2, CD8, and CD62L (ebioscience), then permeabalized and stained intracellularly for IFN-γ, TNF-α, or IL-2 production. Some samples were also surface stained for expression of CD25, 39

60 CD69, and CD122. For detection of degranulation, cells were pre-incubated with TAPI-2 for ½ hour then in the presence of 200nM GP peptide plus monensin and anti- CD107a antibodies (BD Biosciences) for 5 hours at 37 o C prior to surface staining as previously described [143]. For detection of cycling, cells were pre-incubated with TAPI- 2 for ½ hour then in the presence of 200nM GP for 5-24 hours at 37 o C. Cells were then surface stained for CD8, Thy1.1, Thy1.2, and CD62L (ebioscience), then permeabilized using a Foxp3 staining kit (ebioscience) and stained intracellularly with antibodies (Abs) against KI67 (BD Pharmingen). Cells were acquired on a FACSCanto instrument (BD Biosciences). For detection of polyfunctional cytokine production, earlym and latem P14 cells from spleens were incubated in the presence of 200nM GP peptide as described above. Cells were surface stained for CD8, Thy1.1, and CD62L (ebioscience) then permeabilized and stained intracellularly for IFN-γ, TNF-α, and IL-2 production. Cells were acquired on a LSR II instrument (BD Biosciences), and gates were set using cells incubated in the absence of GP peptide. To determine expression of Eomes and T-cell factor 1 (TCF1), splenocytes were surface stained for CD8, Thy1.1, and CD62L, permeabilized using a Foxp3 staining kit (ebioscience), and stained intracellularly with Abs against Eomes (ebioscience) or TCF1 (Cell signaling). Detection of BrdU Uptake, Responsiveness to IL-15, and Homeostatic Proliferation in Rag-/- Mice For detection of basal proliferation, mice were i.p. injected with 2mg bromodeoxyuridine (BrdU) and given 0.8 mg/ml BrdU in drinking water for an 40

61 additional 8 days. P14 cells isolated from peripheral blood were surface stained for CD8 and Thy1.1 (ebioscience) followed by fixation and permeabilization procedures as recommended in the BrdU flow kit (BD Biosciences). Anti-BrdU mab (ebioscience) was used for intracellular staining to detect BrdU uptake. Cells were acquired on a FACSCalibur instrument (BD Biosciences). For determination of responsiveness to IL- 15, splenocytes containing earlym and latem P14 cells were mixed together, washed three times in PBS, and carboxyfluorescein succinimidyl ester (CFSE) labeled by incubating 10 7 splenocytes/ml in room temperature PBS for 15 minutes at 37 o C in the presence of 5µM CFSE. CFSE-labeled cells were incubated on ice for 5 minutes with 1mL of fetal calf serum (FCS) and washed three times with RPMI 1640 containing 10% FCS. CFSE labeled cells were incubated for 3 days at 37 o C in the presence or absence of the indicated concentrations of recombinant mouse IL-15 (Biolegend). Cells were surface stained for Thy1.1, Thy1.2, CD8, and CD62L (ebioscience) and acquired on a FACSCanto instrument (BD Biosciences). Cells incubated without IL-15 were used to set gates for CFSE dilution. For detection of homeostatic proliferation in Rag-/- mice, P14 cells were isolated from spleens of mice containing Thy disparate earlym and latem P14 cells, and cells were surface stained for Thy1.1, CD8, and CD62L (ebioscience) and sorted using a BD FACSAria II (BD Biosciences). Sorted cells were mixed at a 1:1 ratio and 3x10 4 of each were transferred retro-orbitally into Rag-/- mice. Sorted cells were >95% pure. Input ratios of earlym and latem P14 cells were confirmed by flow cytometry before adoptive transfer. Rag-/- mice were sacrificed on d15 after transfer, and ratios of earlym and latem P14 cells in spleens were determined by surface staining for 41

62 Thy1.1, Thy1.2, and CD8. Cells were run on a FACSCalibur instrument (BD Biosciences), and earlym and latem P14 cells were distinguished based on Thy disparity. Measure of Bacterial and Viral Clearance Total splenocytes from mice containing earlym or latem P14 cells were stained with PE-anti-Thy1.1 antibodies and purified with anti-pe magnetic bead sorting with standard AutoMacs protocols. CD62L hi cells were further surface stained for CD8 and CD62L (ebioscience) and sorted using a BD FACSAria II (BD Biosciences). For determination of protection based on bacterial clearance, 7x10 4 PE-selected earlym or latem populations or sorted CD62L hi earlym or latem P14 cells were transferred retroorbitally into naïve C57BL/6 mice followed 24 hours later by i.v. injection of 1x10 5 CFU of virulent Listeria monocytogenes expressing the GP 33 epitope. Three days after infection, spleens were harvested and placed in sterile deionized water containing 0.2% IGEPAL and disrupted using a tissue homogenizer. Samples were plated on TSB-agar plates containing streptomycin and incubated at 37 o C for 24 hours, then CFUs were counted. For determination of protection based on viral clearance, 5x10 4 PE selected earlym or latem populations or sorted CD62L hi earlym or latem P14 cells were transferred retro-orbitally into naïve C57BL/6 mice followed 24 hours later by i.v. infection of 2x10 6 PFU of LCMV clone days after infection spleens were obtained and homogenized, and viral titers were quantified with standard plaque assaying on VERO cells [103]. Sorted cells were >95% pure. Control naïve mice did not receive adoptive transfer of P14 cells. 42

63 Microarray Data Acquisition and Analysis CD62L hi earlym and latem P14 cells were isolated from spleens and cells were surface stained for Thy1.1, CD8, and CD62L (ebioscience) and sorted using a BD FACSAria II (BD Biosciences). Samples from three individual mice were obtained for each group, and sorted cells were >95% pure. RNA was extracted using the RNEasy Kit (QIAGEN), and 1-5ng of mrna was used for microarray analysis. RNA quality was assessed using the Agilent Model 2100 Bioanalyzer. mrna for the microarray was processed using the NuGEN WT-Ovation Pico RNA Amplification System along with the NuGEN WT-Ovation Exon Module. Samples were hybridized and loaded onto Affymetrix GeneChip Mouse 1.0 ST arrays. Arrays were scanned with the Affymetrix Model 7G upgraded scanner, and data were collected using the GeneChip Operating Software. Data from the Affymetrix Mouse Exon 1.0 ST arrays were first quantile normalized and median polished using Robust Multichip Average background correction with log2 adjusted values. Probe sets for exons were then summarized for a specific gene using the median value. After obtaining log2 expression values for genes, significance testing was performed using analysis of variance (ANOVA). Functional assignment of genes was performed using the Functional Annotation Tools in DAVID bioinformatics resources ( following recommended protocols [144]. Enrichment of genes in known pathways was analyzed using the KEGG pathway tool in the DAVID database, and GSEA was performed using GSEA software ( [145,146]. The microarray data were deposited in the NCBI Gene Expression Omnibus with the accession number GSE

64 Quantitative RT-PCR Total splenocytes from mice containing earlym or latem P14 cells were stained with PE-anti-Thy1.1 antibodies and purified with anti-pe magnetic bead sorting with standard AutoMacs protocols. CD62L hi cells were further surface stained for CD8 and CD62L (ebioscience) and sorted using a BD FACSAria II (BD Biosciences). Sorted CD62L hi earlym and latem cells were then incubated with or without 200nM GP peptide at 37 o C for 5 hours. Total RNA was reverse-transcribed using a QuantiTech Reverse Transcription Kit (Qiagen). The resulting cdna was analyzed for expression of different genes by quantitative PCR using SYBR Advantage qpcr pre-mix (Clontech) on an ABI 7300 Real Time PCR System (Applied Biosystems). Relative gene expression levels in each sample were normalized to that of a housekeeping gene, hypoxanthine phosphoribosyltranserase 1 (Hprt1) [147]. The primers used in quantitative RT-PCR were as follows: Ifng: 5 -GCGTCATTGAATCACACCTG and 3 -TGAGCTCATTGAATGCTTGG; Tnfa: 5 -TAGCCCACGTCGTAGCAAAC and 3 -GCAGCCTTGTCCCTTGAAGA; Il2: 5 -AACCTGAAACTCCCCAGGAT and 3 -CGCAGAGGTCCAAGTTCATC; Xcl1: 5 - ATGGGTTGTGGAAGGTGTGG and 3 -TGATCGCTGCTTTCACCCAT; Ccl3: 5 - CATATGGAGCTGACACCCCG and 3 -GTCAGGAAAATGACACCTGGC; Ccl5: 5 - GACAGCACATGCATCTCCCA and 3 -GTGTCCGAGCCATATGGTGA; Fasl: 5 - GCAGAAGGAACTGGCAGAAC and 3 -TTAAATGGGCCACACTCCTC; Klrg1: 5 -TCCTCTGGACGAGGAATGGT and 3 -ACAGCTTCACTCCCTGGTTG; Il2ra: 5 -GGTGCATAGACTGTGTTGGC and 3 -GCAAGAGAGGTTTCCGAAGAC; 44

65 Cd69: 5 -ACATCTGGAGAGAGGGCAGA and 3 -AAGGACGTGATGAGGACCAC; Ccr5: 5 -CCCCTACAAGAGACTCTGGCTC and 3 -TTTTGGCAGGGTGCTGACAT; Il12rb2: 5 -GTGTCTGCAGCCAACTCAAA and 3 -AGGCTGCCAGGTCACTAGAA; Il18rap: 5 -GCAGGCTTACTCACCATTTCA and 3 - GCTTGTGCATCTTTATCCACGG; Cx3cr1: 5 -AAGTTCCCTTCCCATCTGCT and 3 -CGAGGACCACCAACAGATTT; Tbx21: 5 -TCAACCAGCACCAGACAGAG and 3 -CCACATCCACAAACATCCTG; Eomes: 5 -GGAAGTGACAGAGGACGGTG and 3 -AGCCGTGTACATGGAATCGT; Tcf7: 5 -CAATCTGCTCATGCCCTACC and 3 - CTTGCTTCTGGCTGATGTCC; Prdm1: 5 -CCTGCCAACCAGGAACTTCT and 3 - GTTGCTTTCCGTTTGTGTGAGA; Foxm1: 5 -CGAGCACTTGGAATCACAGC and 3 -GGATGGGCACCAGGTATGAG; Id2: 5 -CATCAGCATCCTGTCCTTGC and 3 - GTGTTCTCCTGGTGAAATGG; Id3: 5 -TGATCTCCAAGGACAAGAGGA and 3 - TGAAGAGGGCTGGGTTAAGA; Bcl2: 5 -GGAGGCTGGGATGCCTTTGT and 3 - TGCACCCAGAGTGATGCAG; Bcl6: 5 -CCTGAGGGAAGGCAATATCA and 3 - CGGCTGTTCAGGAACTCTTC; Foxo3: 5 -CTCATGGATGCTGACGGGTT and 3 - CGTCAGTTTGAGGGTCTGCT; Stat3: 5 -TGGTGTCCAGTTTACCACGA and 3 - TGTTCGTGCCCAGAATGTTA; Stat4: 5 -TTTTGACGCTGCAAGAAATG and 3 - TCCAGTCCTGCAGCTCTTCT; Myc: 5 -GTACCTCGTCCGATTCCACG and 3 - GCCTCTTCTCCACAGACACC; Ccnd2: 5 -TCAGTGTGGGTGATCTTGGC and 3 - CAGACCTTCATCGCTCTGTG; Ccnd3: 5 -GGACACTCGCTTTGTTTGGG and 3 - AGCATTTCAGGGCGAGCTTA; Ccne1: 5 -GTGGAGCTTATAGACTTCGCAC and 3 -ACTTACCTGAGAGATGAGCACT; Ccne2: 5 -AGAGTCGATGGCTAGAATGC and 3 -TGTCCAGTAACAGTCATCTCCT; Ccna2: 5-45

66 GGTGAAGGCAGGCTGTTTAC and 3 -AGAAGCTCAAGACTCGACGG; Ccnb1: 5 - CCTGAGCCTGAACCTGAACT and 3 -ACGTCACTCACTGCAAGGAT; Ccnb2: 5 - GCAGAGCAGAGCATCAGAGA and 3 -CAGCCTCTGTGAAACCAGTG; Cdk1: 5 - TCAAGTCTCTGTGAAGAACTCG and 3 -TCCATGGACCTCAAGAAGTACC; Cdk2: 5 -CAATGCAGAGGGGTCCATCA and 3 -ACACACTAGGTGCATTTCAGC; Cdk4: 5 - CAGGTAGGAGTGCTGCAGG and 3 -AGTCAGTGGTGCCAGAGATG; Cdk5: 5 -GGATCTTCCGACTGCTAGGG and 3 -GCTGCACAGGGTTACACTTC; Cdk6: 5 -GCATCGTGATCTGAAACCGC and 3 -GTGACGACCACCGAGGTAAG. Metabolism Assays Both earlym and latem populations and CD62L hi subsets were analyzed for metabolic function. Populations of earlym and latem cells were isolated from total splenocytes from mice containing earlym or latem P14 cells by staining with PE-anti- Thy1.1 antibodies and purifying with anti-pe magnetic bead sorting using standard AutoMacs protocols. Additional cells were sorted for CD62L hi subsets after AutoMacs purification by surface staining for CD8 and CD62L (ebioscience) and sorting using a BD FACSAria II (BD Biosciences). 2x10 5 purified earlym or latem populations or CD62L hi cells were plated in XF media, and oxygen consumption rates (OCR) and extracellular acidification rates (ECAR) were measured under basal conditions and in response to 1 µm oligomycin, 1.5mM fluorocarbonyl cyanide phenylhydrazone (FCCP), and 0.5 µm rotenone + 1µM antimycin with the XF-96 Extracellular Flux Analyzer (Seahorse Bioscience) [148]. 46

67 Results Changes Occur in Memory CD8 T Cell Location, Phenotype, Function, and Maintenance with Time after Infection Ag-specific CD8 T cells can be analyzed at the level of the population (every CD8 T cell in the host), the subset level (CD8 T cells expressing a marker or combination of phenotypic markers), or the level of single cells. T cm and T em subsets differ in anatomical location and functionality [53,55]. Thus, differences in function between memory populations could be due to alterations in subset composition that occur with time after primary antigen recognition. Additionally, individual cells within the population and within subsets can differ in phenotype and function from one another. However, because the level of protection is determined by the quality of all memory CD8 T cells present at the time of re-infection, I first examined how circulating memory CD8 T cells change with time after infection when analyzed on the population level. I adoptively transferred low numbers (5,000 P14 cells/mouse) of naïve Thy1.1 or Thy1.1/1.2 Tg P14 CD8 T cells specific for the GP epitope derived from LCMV into Thy1.2 C57BL/6 recipients and infected recipients with LCMV 24 hours later. Low numbers of naïve P14 cells were used for adoptive transfer, because the magnitude of CD8 T cell responses and memory differentiation have been shown to be unaffected by adoptive transfer of low numbers of naïve CD8 T cells [149]. I then analyzed memory P14 cells on the population level (i.e. all memory cells present in the examined organs) days (earlym) or 8+ months (latem) later. Although similar numbers of P14 cells were found in the spleen (Figure 4A), earlym cells were found in greater proportions in the lungs following perfusion while 47

68 latem cells were found in greater proportions in the inguinal lymph nodes (Figure 4B). Surface marker profiles also differed with time, with expression of CD127, CD62L, CD27, and CD122 increasing and expression of KLRG1 decreasing with time after infection (Figure 4C). Similar patterns of surface marker expression were also seen for endogenous earlym and latem GP 33 - and GP 276 specific CD8 T cell responses (Figure 4D). To determine if cytokine production or degranulation changes with time, ICS was performed on earlym and latem cells that were mixed and incubated with GP peptide for 5 hours. The percentages of IFN-γ and TNF-α producing P14 cells and degranulation as measured by surface CD107a expression did not change with time. However, the percentage of memory CD8 T cells able to produce IL-2 increased with time (Figure 4E). Consistent with these data, the percentage of cells capable of polyfunctional cytokine production (IFN-γ, TNF-α, and IL-2) increased with time after infection (Figure 4F). Finally, to determine if maintenance of memory CD8 T cells changes with time, basal proliferation of earlym and latem cells was examined with BrdU incorporation during an eight-day period. A higher percentage (p<0.01) of latem cells incorporated BrdU during the eight-day interval (Figure 4G), indicating that the rate of basal proliferation increases in memory CD8 T cells with time after infection. Basal proliferation is partially dependent upon IL-15 signaling [85,87,150], and increased expression of IL- 15Rβ (CD122) in latem cells (Figure 4C) suggested that sensitivity to IL-15 could change with time. To test this, Thy disparate earlym and latem cells were labeled with CFSE, mixed and incubated with increasing concentrations of IL-15, and dilution of CFSE was determined after 3 days. latem cells proliferated to a greater extent in response to exogenous IL-15 (Figure 4H), indicating that sensitivity to IL-15 increases over time. 48

69 Expression of CD122 was also greater in latem compared to earlym endogenous GP 33 and GP 276 tetramer positive cells (Figure 4D), suggesting that endogenous latem cells may be more responsive to IL-15 compared to earlym cells. These data show that while memory CD8 T cells analyzed on the population level persist at stable numbers in the spleen after infection, time changes their relative distribution, surface marker expression, Ag-driven cytokine production, and ability to respond to homeostatic cues in the environment. Antigen-driven Proliferation and Memory Generation Potential Increase in Memory CD8 T Cells over Time Primary memory CD8 T cells robustly proliferate and generate secondary effector and memory populations after Ag re-encounter [9,10]. Proliferation by primary memory CD8 T cells was reported to increase with time after initial Ag exposure following secondary infection with Sendai virus or Listeria monocytogenes [10,102]. To verify that time-dependent changes in Ag-driven proliferation and secondary memory generation of primary memory CD8 T cells analyzed on the population level are not dependent upon the type of infection or Ag-specificity, we set up adoptive co-transfer experiments. Thy disparate earlym and latem P14 cells were mixed (Figure 5A) and 2x10 4 of each were transferred into naïve C57BL/6 recipients followed by infection with pathogens expressing GP 33 including LCMV, LM, and VacV. This experimental setup ensured that secondary responses generated from earlym and latem cells were subject to the same in vivo environmental conditions throughout the response. Secondary responses generated from earlym and latem P14 cells after LCMV Armstrong, LM, and VacV infection were 49

70 tracked longitudinally in PBL, and a significantly greater percentage of secondary effector cells was found for responses generated from latem compared to earlym P14 cells after each infection (Figure 5B,C). Additionally, when progeny were examined at a secondary memory time point, I found a 5-7 fold increase in the frequency of secondary memory cells generated from latem compared to earlym P14 cells after each infection (Figure 5C). Furthermore, a greater percentage of secondary effector and memory cells was found for responses generated from latem compared to earlym P14 cells in multiple organs following LCMV infection (Figure 5D). These results indicate that reduced progeny generated from earlym cells was not due to restriction of responses by primarily CD62L lo earlym cells to non-lymphoid tissues. Taken together, these data indicate that Ag-driven proliferation and memory generation potential of primary memory CD8 T cells analyzed on the population level increases with time after infection irrespective of the type of infection and for memory CD8 T cells recognizing epitopes beyond OVA and those expressed by Sendai virus. Phenotypic Heterogeneity of CD62L hi Memory CD8 T Cells Decreases with Time Historically, circulating memory CD8 T cells have been characterized as CD62L lo T em or CD62L hi T cm cells based on expression of CD62L and localization in peripheral tissues and lymphoid organs [53,55]. Functional differences between T em and T cm subsets also have been demonstrated, with T cm having a greater capacity to produce IL-2, increased proliferative potential, and the ability to provide increased protection following some, but not all infections [55,135]. Therefore, the functional changes observed in the memory CD8 T cell population with time could be due solely to changes in subset 50

71 composition. Alternatively, the properties of defined subsets of memory CD8 T cells could change with time. To test the hypothesis that changes in memory CD8 T cell properties with time after infection are not due solely to shifts in subset composition, I began to examine the properties of earlym and latem CD62L hi cells. I chose to focus on CD62L hi memory subsets for two reasons. First, with time after acute infection memory CD8 T cells convert to a population that is primarily CD62L hi [10,100,101,139] (Figure 6A). Second, with time the CD62L lo population becomes enriched for T DIM S that arise from homeostatic division, are non-functional, and are destined to die [88]. To further document this, I examined IFN-γ production by CD62L hi and CD62L lo earlym and latem P14 cells following stimulation with cognate Ag. The capacity of CD62L hi earlym and latem cells to produce IFN-γ was similar (Figure 6B). In contrast, a higher percentage of CD62L lo latem compared to earlym cells were unable to produce IFN-γ (Figure 6B), consistent with an enrichment of T DIM S. At present, there is no reliable phenotypic marker that can be used to distinguish T DIM S from normally functioning memory CD8 T cells within the CD62L lo population. Thus, by focusing my analysis on CD62L hi memory CD8 T cells, I was able to examine how a well-characterized subset of memory CD8 T cells changes with time after infection while avoiding complications in analyzing CD62L lo populations. Surface marker expression has been used to characterize memory CD8 T cells with different functional abilities. I rationalized that if CD62L hi earlym and latem cells displayed changes in phenotype, they also were likely to display other differences as well. Therefore, I began to explore to what extent the phenotype of CD62L hi memory 51

72 populations was influenced by time by examining the expression of CD27, CD127, and CD122, markers highly expressed by T cm cells, on CD62L hi earlym and latem P14 cells (Figure 7A). The percentage of CD62L hi cells expressing CD27, CD127, and CD122 was greater in latem compared to earlym P14 cells in the PBL and spleen indicating that the phenotype of CD62L hi memory CD8 T cells continues to change with time. This pattern was also seen for expression of CD122 and KLRG1 on gated CD62L hi endogenous GP 33 and GP 276 earlym and latem cells (Figure 7B). The percentage of cells expressing CD27, CD127, and CD122 among CD62L hi earlym P14 cells also differed between organs. Percentages of cells expressing CD27 and CD127 in the lungs were lower compared to the PBL, spleen, and inguinal lymph nodes, while percentages of cells expressing CD122 were lower in the inguinal lymph nodes compared to the PBL, spleen, and lungs. In contrast, percentages of CD62L hi latem cells expressing CD27, CD127, and CD122 were uniformly high and were similar regardless of anatomical location except for a reduction in the percentage of cells expressing CD122 in the inguinal lymph nodes. This suggested that phenotypic heterogeneity within the CD62L hi memory CD8 T cell subset decreases with time after infection. To further test this, I co-stained for expression of CD27, CD127, CD122, CD11b, and KLRG1 on CD62L hi earlym and latem P14 cells. This strategy allowed me to differentiate 32 different subpopulations of CD62L hi memory P14 cells, and the percentage of each subpopulation of total CD62L hi earlym and latem P14 cells is shown in Figure 8A. To examine heterogeneity within CD62L hi earlym and latem cells, the number of subpopulations (out of 32) comprising greater than 1% of the total CD62L hi memory pool was counted. In each of the organs examined, earlym contained 2-3 fold 52

73 greater numbers of subpopulations than latem cells (Figure 8B). Similar patterns were seen for endogenous GP 33 and GP 276 tetramer positive memory cells (Figure 9A,B), and for CD62L hi earlym and latem P14 cells co-stained with CD27, CD127, CD43, CXCR3, and CCR5 (Figure 9C). Taken together, these data indicate that the phenotype of CD62L hi memory CD8 T cells continues to change while heterogeneity of memory CD8 T cells decreases with time after infection. Gene Expression Patterns among CD62L hi Memory CD8 T Cells Change with Time Differences in phenotype among CD62L hi earlym and latem P14 cells suggested that gene expression and functional changes could also occur with time. In order to determine the extent of changes in gene expression that occur with time within CD62L hi memory subsets and to determine if genes regulating memory CD8 T cell functions were differently expressed, I examined the transcriptomes of CD62L hi earlym and latem P14 cells. I detected 3,494 genes that were differentially expressed (p<0.05) between CD62L hi earlym and latem P14 cells, and Figure 10A shows a heat map of genes with significantly different expression between CD62L hi earlym and latem P14 cells at fold differences >1.25. To determine if gene families with function in T cell biology are differentially expressed between CD62L hi earlym and latem P14 cells, I used DAVID bioinformatics resources [144] to assign biological functions and to group genes into function-related classes genes were differentially expressed between CD62L hi earlym and latem P14 cells at fold >1.25, and this number of genes was too great to be processed using the DAVID database. To reduce the number of genes to a size that the DAVID database was able to process, genes with significantly different expression 53

74 (p<0.05) and fold >1.5 (474 total genes) were used for this analysis. This revealed that mrna expression for genes involved in many cellular processes are differentially expressed over time after infection in CD62L hi memory cells (Table 1). Prominently among these changes, CD62L hi latem cells showed increased expression of cytokine receptors including Il2rb and IL15ra (encoding IL-2Rβ and IL-15Rα respectively), and decreased expression of killer-cell lectin-like receptors including Klrg1. Furthermore, latem cells showed alterations in genes regulating cell cycle progression and ribosome biogenesis, suggesting that the proliferative potential of memory CD8 T cells may increase with time after infection not only at the population level (Figure 5), but also within defined subsets. Additionally, I used KEGG pathway analysis [151] to examine regulation of biological pathways in CD62L hi earlym and latem cells. This analysis revealed that multiple pathways are differentially regulated with time in CD62L hi memory cells including cell cycle and ribosome pathways (Figure 10B). It is recognized that metabolic pathways are dynamically regulated during CD8 T cell responses, and both naïve and memory CD8 T cells catabolize fatty acid and utilize oxidative phosphorylation to meet their energy requirements [148, ]. The DAVID and KEGG analyses revealed that metabolism-related genes/pathways including metabolism of fatty acid, fructose and mannose metabolism, and oxidative phosphorylation were altered with time after infection in CD62L hi memory CD8 T cells, and suggested that CD62L hi latem cells might be more metabolically fit than CD62L hi earlym cells (Table 1 and Figure 10B). Taken together, these data further indicate that transcriptomic alterations occur within CD62L hi memory CD8 T cells over time after infection. 54

75 Cytokine Production, Degranulation, and Functional Avidity of CD62L hi Memory CD8 T Cells is not Affected by Time Memory CD8 T cells rapidly respond to infection with the production of cytokines and the release of cytolytic molecules including perforin and granzymes [7]. While microarray data did not indicate differences in mrna expression of effector molecules between resting earlym and latem P14 cells, this did not rule out the possibility that effector functions of CD62L hi memory CD8 T cells change with time in response to Ag-stimulation. To examine if cytokine production by CD62L hi memory CD8 T cells changes with time, Thy disparate earlym and latem P14 cells were mixed together and incubated with GP peptide for 5 hours followed by ICS for detection of IFN-γ, TNF-α, IL-2, and CD107a as a measure of degranulation. No differences in the production of any cytokines or surface CD107a expression as a measure of degranulation were observed between earlym and latem CD62L hi P14 cells (Figure 11A,B). To determine if sensitivity to Ag changes with time in CD62L hi memory CD8 T cells, I performed ICS as described above using decreasing concentrations of GP peptide. No differences in Ag sensitivity were detected between CD62L hi earlym and latem P14 cells based upon functional avidity curves and the effective concentration of peptide required to induce 50% of cells to produce IFN-γ (EC50) (Figure 11C,D). Additionally, when earlym and latem P14 cells were incubated with GP peptide for decreasing lengths of time, no differences in the time required to produce IFN-γ, TNF-α, or IL-2 were observed (Figure 11E), suggesting that there is a similarly poised state of earlym and latem CD62L hi cells to produce cytokines upon Ag recognition. As suggested from the microarray, these data indicate that effector functions of CD62L hi 55

76 memory CD8 T cells including cytokine production, secretion of cytolytic molecules, Ag sensitivity, and time required to produce cytokines does not change with time after infection. To gain a broader appreciation for the poised state of earlym compared to latem CD62L hi cells, I incubated CD62L hi earlym and latem cells for 5 hours in the presence or absence of GP peptide and examined expression of a number of effector molecules, surface markers, transcription factors, and cell cycle associated genes known to be dynamically regulated following Ag-encounter [69,153,159,160]. As I had previously noted by flow cytometry for expression of IFN-γ, TNF-α, and IL-2 (Figure 11A,B), mrna expression (Figure 12A) and protein expression (Figure 12B) of effector molecules was similar between CD62L hi earlym and latem cells following incubation for 5 hours with cognate Ag. While expression of some genes including Klrg1 and Cx3Cr1 were differently expressed between resting CD62L hi earlym and latem cells (Figure 13A) as was indicated from the microarray data (Table 1), mrna (Figure 13A) and surface protein (Figure 13B) expression of activation markers was similar for CD62L hi earlym and latem cells following the 5 hour incubation period, indicating that CD62L hi earlym and latem cells are similarly activated following pathogen re-encounter. Additionally, mrna (Figure 14A) and protein levels (Figure 14B) of transcription factors were regulated to a similar extent in CD62L hi earlym and latem cells following 5 hour peptide incubation. Interestingly, while the microarray data indicated that CD62L hi latem cells regulate expression of cell cycle related genes differently compared to earlym cells (Figure 10B and Table 1), mrna levels of cyclins and cyclin dependent kinases was similarly regulated in CD62L hi earlym and latem cells following 5 hour incubation 56

77 with cognate Ag (Figure 15A). Taken together, these data indicate that CD62L hi earlym and latem cells display a similarly poised state to respond following Ag re-encounter. Ability to Undergo Homeostatic and Ag-driven Proliferation Increases with Time among CD62L hi Memory CD8 T Cells While mrna expression of cell cycle genes following 5 hour incubation with Ag indicated that genes regulating cell cycling were similarly regulated in CD62L hi earlym and latem cells following Ag re-encounter (Figure 15A), KEGG pathway analysis indicated that genes regulating cell cycle pathways are differentially expressed in CD62L hi memory CD8 T cells with time (Figure 10B and 16A). Additionally, GSEA [146] comparing gene expression patterns of CD62L hi earlym and latem P14 cells with existing gene sets revealed that genes highly expressed by CD62L hi latem P14 cells were enriched in gene sets involved in cell cycle pathways (Figure 16B). Furthermore, while cell cycle entry of CD62L hi earlym and latem cells incubated with cognate Ag in vitro was similar after 5 hours based on KI67 expression, a greater percentage of CD62L hi latem compared to earlym cells entered the cell cycle following incubation with cognate Ag for 24 hours (Figure 17A). Taken together, these data suggested that CD62L hi latem CD8 T cells might possess enhanced abilities to undergo homeostatic proliferation and/or Ag-driven proliferative expansion following re-infection. Basal and homeostatic proliferation is partially dependent upon IL-15 [85,87,150]. IL-15 bound to IL-15Rα is trans-presented to CD8 T cells and signals through the common receptor γ chain (γ c, CD132) and IL-2Rβ (CD122) [161]. mrna 57

78 expression of CD122 as determined from microarray data (Figure 18A) and surface expression of CD122 as detected by flow cytometry (Figure 7A and 18B), increased with time in CD62L hi memory P14 cells, suggesting that sensitivity to IL-15 could increase in CD62L hi memory CD8 T cells with time after infection. To test this, Thy disparate earlym and latem P14 cells were labeled with CFSE, mixed and incubated with increasing concentrations of IL-15, and dilution of CFSE was determined after 3 days. CD62L hi latem P14 cells proliferated to a greater extent in response to exogenous IL-15 (Figure 18C). Endogenous GP 33 and GP 276 CD62L hi latem cells also displayed increased expression of CD122 compared to earlym cells (Figure 7B) suggesting they might possess increased sensitivity to IL-15 compared to earlym CD62L hi cells. Taken together, these data indicate that proliferation in response to IL-15 increases over time in CD62L hi memory CD8 T cells. Differences in responsiveness to IL-15 suggested that CD62L hi earlym and latem CD8 T cells could differ in their ability to undergo homeostatic proliferation. To test this, Thy disparate CD62L hi earlym and latem P14 cells were sorted, mixed in equal numbers (Figure 18D left panel), injected into lymphopenic Rag-/- mice (3x10 4 each), and the percentage of CD62L hi earlym and latem cells was determined in the spleens of recipients 15 days after transfer. A higher percentage of latem cells was found with the ratio of latem to earlym cells increasing approximately two-fold (Figure 18D right panel), indicating that ability to undergo homeostatic proliferation increases with time in CD62L hi memory CD8 T cells. Differences in cell cycle regulation also could indicate that the ability to undergo Ag-driven proliferation changes with time among CD62L hi memory CD8 T cells. To test 58

79 this, Thy disparate CD62L hi earlym and latem P14 cells were sorted, mixed in equal numbers (Figure 19A), and 1x10 4 of each were transferred into naïve C57BL/6 recipients followed by infection with LCMV 24 hours later. An increased percentage of progeny were generated from CD62L hi latem P14 cells during the effector phase (Figure 19B-D), indicating that with time CD62L hi memory CD8 T cells have an increased ability to undergo secondary expansion. Adoptive transfer of endogenous CD62L hi GP 276 -specific earlym and latem cells (Figure 20A) also showed increased numbers of 2 o effector cells generated from latem cells compared to earlym cells in PBL and spleen (Figure 20B,C), and nearly 4 times the number of 2 o effector cells generated from CD62L hi latem cells compared to earlym cells were recovered in spleens 7 days post LCMV infection (Figure 20C). Additionally, when progeny generated from CD62L hi earlym and latem P14 cells were examined at memory time points in PBL (Figure 19C,D) or in peripheral tissues and secondary lymphoid organs (Figure 19E), a greater percentage of secondary memory cells were generated from CD62L hi latem compared to earlym, indicating that with time, memory generation potential increases for CD62L hi memory CD8 T cells. As suggested by the microarray data, these results indicate that with time after infection, the ability to undergo Ag-driven proliferation and generate secondary memory cells increases within CD62L hi memory CD8 T cells. Protective Capacity of Memory CD8 T Cells Changes with Time in a Pathogen Dependent Manner The goal of vaccination is establishment of memory populations that will provide increased protection upon infection, and studies have indicated that the quantity, quality, 59

80 and localization of memory CD8 T cells required for protection differ depending upon the nature of the pathogen [55,68,103,108,135,162]. The functional differences that we observed suggested that CD62L hi earlym and latem cells might provide differing levels of protection following infection. To determine if per cell protective capacity of CD62L hi memory CD8 T cells against an acute systemic infection changes with time, CD62L hi earlym and latem P14 cells were sorted, and 7x10 4 cells were transferred into naïve C57BL/6 recipients followed by infection with virulent LM expressing GP 33. Both CD62L hi earlym and latem P14 cells provided protection, as significantly decreased (p<0.05) CFUs of LM were detected in spleens of recipient mice three days after infection compared to mice not receiving adoptive transfer. However, per cell protective capacity of CD62L hi earlym and latem P14 cells did not differ (Figure 21A). The most dramatic alteration in functional ability between CD62L hi earlym and latem CD8 T cells that we observed was the ability to undergo Ag-driven proliferation, and some studies have indicated that the proliferative abilities of memory CD8 T cells are less important than localization and killing ability for providing protection from LM [68,103,135]. However, studies have shown that proliferative abilities of memory CD8 T cells are crucial for clearance of infection with LCMV clone-13, which causes a chronic infection in mice [55,103,108]. To determine if per cell protective capacity of CD62L hi memory CD8 T cells against a chronic infection changes with time, we sorted CD62L hi earlym and latem P14 cells and transferred 5x10 4 cells into naïve C57BL/6 recipients followed by infection with LCMV clone-13. Mice that received adoptive transfer of CD62L hi latem cells had reduced viral titers eight days following infection compared to mice not receiving transferred cells, and this level of protection was significantly greater than that 60

81 provided by CD62L hi earlym cells (Figure 21B). Increased protection provided by latem CD62L hi cells correlated with enhanced magnitudes of proliferative expansion as a greater percentage of progeny generated from CD62L hi latem compared to earlym cells was found in the PBL (Figure 21C), and higher numbers of progeny generated from CD62L hi latem were found in the spleens of recipient mice 8 days post infection (p.i.) (Figure 21D). Taken together, these data indicate that protection provided by CD62L hi memory CD8 T cells changes with time after infection in a pathogen-dependent manner. Protection provided against infection is mediated by all memory cells present at the time of re-infection. When analyzed on the population level (i.e. no sorting based on CD62L expression), latem CD8 T cells proliferated to a greater extent following acute infection compared to earlym populations (Figure 5), suggesting that protection provided by populations of memory CD8 T cells may also differ with time. To determine if per cell protective capacity of memory CD8 T cell populations changes with time, cells were isolated from the spleens of mice containing earlym or latem P14 cells, and 7x10 4 cells were transferred into naïve C57BL/6 recipients followed by infection with virulent LM expressing GP 33. Both earlym and latem populations provided protection against LM, but no difference in the level of protection was observed (Figure 21E). To determine if protection provided by memory CD8 T cell populations against a chronic infection changes with time after infection, 5x10 4 earlym or latem cells were transferred into naïve C57BL/6 recipients followed by infection with LCMV clone-13. LateM populations provided enhanced protection against chronic LCMV clone-13 infection compared to earlym populations (Figure 21F). Enhanced protection provided by latem CD8 T cells also correlated with enhanced secondary expansion in PBL and greater numbers of P14 61

82 cells recovered from spleens of infected mice 8 days following infection (Figure 21G,H). Taken together, these data indicate that the ability of memory CD8 T cells to provide protection against chronic infection increases with time after infection in a manner not due solely to shifts in T em to T cm subsets. Mitochondrial Function of Memory CD8 T Cells Improves with Time after Infection Alterations in metabolic function including enhanced fatty acid oxidation, increased mitochondrial mass, and increased ability to perform oxidative phosphorylation have been shown to enhance memory CD8 T cell development and favor rapid recall responses following Ag re-encounter [148, ,163]. Additionally, memory CD8 T cells that displayed an enhanced ability to undergo oxidative phosphorylation and that possessed increased spare respiratory capacity (SRC), which is the reserve ATP generation capacity of cellular mitochondria, proliferated to a greater extent and provided enhanced protection against chronic infection with LCMV clone-13 [108]. Functional annotation and KEGG pathway analysis of our microarray data revealed that genes regulating mitochondrial function and metabolic pathways including oxidative phosphorylation; fructose, mannose, and glucose metabolism; and fatty acid oxidation were differently regulated between CD62L hi earlym and latem CD8 T cells (Table1, Figure 10B and 22A). Additionally, GSEA analysis revealed that latem CD62L hi CD8 T cells were enriched in genes sets involved in oxidative phosphorylation (Figure 22B). This suggested that CD62L hi memory CD8 T cells might alter their metabolic programs with time after infection, which would impact their ability to proliferate and to provide protection against LCMV clone-13 infection. 62

83 To determine if the metabolic function of CD62L hi memory CD8 T cells is altered with time after infection, I sorted earlym and latem CD62L hi memory CD8 T cells and performed extracellular flux analysis using a Seahorse bioanalyzer. Comparison of the basal OCR, a measure of oxidative phosphorylation, to ECAR, a measure of aerobic glycolysis, revealed that compared to CD62L hi earlym cells, CD62L hi latem cells rely to a greater extent on oxidative phosphorylation for the generation of ATP (Figure 22C). Due to the demands of cell sorting, I was unable to determine SRC of CD62L hi earlym and latem cells, as they did not respond to the FCCP inhibitor. However, latem CD8 T cell populations also proliferated to a greater extent than earlym populations and provided enhanced protection against LCMV clone-13, and these populations could be isolated without using cell sorting. To examine if the metabolic function of CD8 T cell memory populations is altered with time after infection I performed extracellular flux analysis of isolated earlym and latem CD8 T cell populations. latem cells displayed increased basal OCR levels compared to earlym cells, and both cell types responded to metabolic inhibitors (Figure 22D). Comparing the ratio of OCR to ECAR for earlym and latem populations reveled that latem cells rely more heavily on oxidative phosphorylation for ATP production compared to earlym populations (Figure 22E). Analysis of the SRC, calculated as the highest OCR after addition of FCCP over the basal OCR, for earlym and latem revealed a trend for higher SRC in populations of latem cells compared to earlym cells (Figure 22F). Taken together, these data indicate that mitochondrial function of memory CD8 T cells improves with time after infection in a manner that is not solely due to shifts in T cm and T em subsets. The enhanced mitochondrial function of latem populations and CD62L hi cells likely provides latem 63

84 cells with a metabolic advantage enabling robust proliferation and enhanced protection against chronic infection with LCMV clone-13. Discussion Protection provided by memory CD8 T cells is dependent upon their numbers, functional ability (i.e. quality), and location at the time of infection [129]. I have shown that the quality of the circulating memory CD8 T cell population differs with time after infection in a manner not solely due to shifts in memory subset composition. Some functions of memory CD8 T cells analyzed on the population level, such as ability to produce IL-2, increased with time after infection, but were no different in CD62L hi memory cells early or late after infection. However, other qualitative aspects of memory CD8 T cells including proliferation in response to the homeostatic cytokine IL-15 or to Ag, and mitochondrial function increased with time after infection when both the memory CD8 T cell population, and defined CD62L hi subsets were analyzed. Thus, while some alterations in the functional abilities of memory CD8 T cells with time after infection can be attributed to shifts in subset composition, other qualitative changes cannot be wholly attributed to shifts in subset composition. Interestingly, as a consequence of these functional changes, the protection provided by memory CD8 T cell populations and CD62L hi memory CD8 T cells against a chronic viral infection increased over time. Importantly, my data suggests that the outcome of vaccination schemes designed to elicit protective memory CD8 T cells will depend on the timing between booster immunizations, and on the timing of re-infection following vaccination. 64

85 While circulating memory CD8 T cells are best suited to provide protection against systemic infections, tissue resident memory CD8 T cells provide a first line of defense against pathogens encountered in peripheral tissues [59,60,62,63,134,136]. While the longevity of T rm cells relative to circulating memory CD8 T cells is unclear at present, some studies have indicated that T rm cells remain in mice for at least 300 days following infection with vaccinia virus [136], herpes virus [60], or vesicular stomatitis virus [54]. However, other studies examining T rm formation following infection with influenza virus have indicated that the T rm CD8 T cell population wanes following infection [138]. Thus, longevity of T rm CD8 T cells may vary depending on factors including the nature of the primary infection and/or vaccination and the tissue of residence. Additionally, studies examining the longevity and phenotypic and functional changes that might occur in T rm CD8 T cells over time following infection will be complicated due to a lack of phenotypic markers that definitively identify T rm cells, as CD103 and CD69, markers used to identify T rm cells, are not expressed on all T rm cells [65,134,164]. However, T rm CD8 T cells likely would play an important role in providing protection against human pathogens that infect at peripheral tissues, including HIV, herpes viruses, and influenza viruses. Therefore, determining the longevity of T rm cells and whether the phenotype, function, and protective abilities of T rm cells differs with time after infection, as I have shown for circulating memory CD8 T cells, is an important goal. While I have provided evidence that changes in memory CD8 T cell phenotype and function seen on the population level are not due solely to conversion to CD62L hi cells with time after infection, a question still remains as to how the progressive changes in phenotype and function seen with time after infection in both memory CD8 T cell 65

86 populations and within CD62L hi memory CD8 T cells occurs. My microarray data suggests that genes important for memory CD8 T cell function are differentially regulated in CD62L hi memory CD8 T cells with time after infection, but it is unclear if differences in the transcriptional program with time after infection are due to 1) cell-intrinsic changes in gene regulation, or 2) a subset of memory CD8 T cells within the earlym population selectively survives and comes to constitute the latem pool. To ideally address these possibilities, earlym CD8 T cell subsets would be transferred into naïve mice, and the phenotype, function, protective abilities, and transcriptional regulation of the transferred cells would be analyzed for the latem cells derived from the transferred populations. However, due to the low number of CD62L hi memory CD8 T cells present early following infection, loss of cells upon adoptive transfer, and further loss of cells upon isolation from recipient mice, these experiments are difficult to execute. I hope that improvements in cell isolation technology will allow me to perform these experiments in the future. I considered the possibility that transcription factors that facilitate memory CD8 T cell formation such as Tbet, Eomes, or TCF1 [147,165,166] could differentially regulate survival of subsets of memory cells displaying phenotypic markers that were present in earlym cells but not present in latem cells as seen in Figure 8. However, I was unable to find conclusive evidence that expression of these transcription factors regulated survival of subsets within CD62L hi earlym and latem cells. The rate at which phenotypic and functional changes of memory CD8 T cells occur following infection and/or vaccination is likely to be influenced by a number of factors. In this study I examined Tg memory P14 cells primarily localized within the spleen following acute infection with LCMV. Previous studies have indicated that the 66

87 rate of acquisition of effector functions and expression of surface markers associated with central memory CD8 T cells is influenced by the number of transferred Tg T cells [139,149,167]. However, expression of CD62L following adoptive transfer of low numbers of Tg P14 cells as used in my study has been shown to be similar between Tg and endogenous D b -GP 33 restricted CD8 T cells [139,167]. On the other hand, the rate of surface CD62L expression and ability to produce IL-2 has been shown to differ between endogenous T cell populations of different LCMV epitope specificities and among CD8 T cells localized within secondary lymphoid organs or peripheral tissues [139]. Additionally, the nature of the infecting pathogen and/or inflammation elicited during either infection or vaccination has been shown to influence the phenotype and function of memory CD8 T cells generated during the response. Infection with LCMV, vaccinia virus, or influenza virus leads to the formation of memory CD8 T cells with distinct phenotypic and functional qualities [168,169], while administration of inflammation inducing toll-like receptor agonists during DC immunizations abrogates the rapid acquisition of memory characteristics seen during low-inflammatory DC immunizations [78-80]. Because of these considerations, the extent of changes that occur within the CD8 T cell memory population with time, and thus their functional and protective abilities during re-infection will likely depend upon conditions elicited during the primary infection and/or immunization. Therefore, vaccine design should include considerations of how the vaccine strategy may influence changes in memory CD8 T cells with time. Unlike mice housed in specific pathogen free facilities, humans are infected with many non-related pathogens, and co-infections or chronic infections could influence the development and/or differentiation of primary memory CD8 T cells, or the properties of 67

88 already established memory CD8 T cells. A recent report showed that established chronic infections in mice influence the development and differentiation of primary memory CD8 T cells, but that the impact of chronic infections on pre-established primary memory CD8 T cells was less severe [170]. However, pre-established memory CD8 T cells examined in their study were generated 1 year prior to the chronic infection. It will be interesting to examine if differentiation of more recently established memory CD8 T cells is also minimally impacted by chronic or repeated unrelated infections. Studies have indicated that the quantity, quality, and localization of memory CD8 T cells required for protection differ depending upon the nature of the pathogen [55,68,103,108,135,162]. I found that the protective abilities of memory CD8 T cells changes with time in a pathogen-dependent manner. CD62L hi central memory cells have been described as being more effective at providing protection against chronic infections due to localization within the lymph nodes and increased mitochondrial function leading to an enhanced ability to proliferate [55,103,108]. I showed that the ability of memory CD8 T cell populations and CD62L hi central memory CD8 T cells to provide protection from LCMV clone-13 infection increases with time, and that this increased protection correlated with enhanced mitochondrial function and proliferative abilities following infection. Some studies [68,103,135], however, have indicated that memory CD8 T cells with effector memory characteristics provide increased protection against acute infection with L. monocytogenes or localized infection with vaccinia virus, and the localization of the memory population to sites of infection is important in these instances. Therefore, increased protection provided by latem cells likely does not apply to all infections. 68

89 The numbers of memory CD8 T cells required to achieve protection against certain pathogens including Plasmodium species which cause malaria is quite high, and prime boost protocols have been established in order to achieve high numbers of memory CD8 T cells [104,105,140]. My data indicate that higher numbers of memory CD8 T cells may be achieved through prime boost protocols by increasing the length of time between boosts. However, my study analyzed primary memory cells, and recent studies have indicated that the properties of memory CD8 T cells including magnitude of proliferative expansion, duration and degree of contraction, cytotoxicity, IL-2 production, basal proliferation and long-term survival, memory generation potential, lymph node homing, and transcriptome diversification change sequentially with each additional Ag encounter [107,171,172]. While little is known about how the number of Ag encounters influences the changes in memory CD8 T cell functions that occur with time after infection, studies indicate that the phenotype of memory CD8 T cells that have encountered Ag multiple times changes with time after infection, but at a slower rate than in primary memory CD8 T cells [100,101]. As with primary memory, changes with time in the properties of memory CD8 T cells that have encountered Ag more than once could influence their ability to provide protection against infection and/or affect the outcome of prime boost immunizations requiring multiple boosts. My results firmly establish that memory CD8 T cells continue to change with time after infection. The results indicate that the function of memory CD8 T cells continues to change with time after infection, and that protection provided by memory CD8 T cells changes with time in a pathogen-dependent manner. Because of this, experimental investigation of memory CD8 T cell quality and/or protection following 69

90 either infection or vaccination should include analysis of memory CD8 T cells at multiple time points following the infection and/or vaccination. 70

91 Figure 4: Localization, Phenotype, and Function of Memory CD8 T Cells Changes with Time after Infection Memory P14 cells were generated by adoptively transferring 5x10 3 naïve P14 T cells into Thy disparate naïve recipients followed by i.p. injection of LCMV 24 hours later. Analysis was performed on earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells. (A) Numbers of earlym and latem P14 cells found in the spleens of recipient mice. (B) Ratios of the percentage of earlym or latem P14 cells in the indicated organs following perfusion out of total lymphocytes to the percentage of earlym or latem P14 cells in PBL out of total lymphocytes. (C) Representative histograms showing CD127, CD62L, CD27, CD122, and KLRG1 expression on gated earlym (open histograms) and latem (grey histograms) P14 cells isolated from spleens. (D) Representative histograms showing CD127, CD62L, CD27, CD122, and KLRG1 expression on gated earlym (open histograms) and latem (grey histograms) endogenous GP 33 or GP 276 tetramer positive cells isolated from spleens. (E) Percentage of gated earlym or latem P14 cells producing IFN-γ, TNF-α, or IL-2 as measured by ICS, or undergoing degranulation as measured by surface CD107a expression following 5 hour incubation with GP peptide. (F) Percentage of gated earlym or latem P14 cells producing one (IFN-γ), two (IFN-γ and TNF-α), or three (IFN-γ, TNF-α, and IL-2) cytokines as measured by ICS following 5 hour incubation with GP peptide. (G) Representative dot plots of BrdU staining for gated earlym or latem P14 cells in PBL 8 days after BrdU injection and drinking water administration. (H) Representative histograms of CFSE dilution on gated earlym (open histograms) or latem (grey histograms) P14 cells after 3 day culture in the presence or absence (black histograms) of the indicated concentrations of IL-15. NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Representative data from one of three individual experiments with 3 mice per group per experiment. Error bars represent the standard error of the mean. 71

92 72

93 Figure 5: Proliferation and Memory Generation Potential in Response to Agreencounter Increases with Time after Infection Thy disparate earlym and latem P14 cells were mixed 1:1 and injected into naïve recipients followed by i.p. injection of LCMV or VacV, or i.v. injection of LM 24 hours later. (A) Dot plot showing the master mix of earlym (Thy1.1/1.1) and latem (Thy1.1/1.2) P14 cells used for adoptive transfer. (B) Representative plots showing the response of progeny of earlym (Thy1.1/1.1) and latem (Thy1.1/1.2) P14 cells in PBL d7 after the indicated infections. (C) Kinetic analysis of secondary responses generated from earlym and latem P14 cells in PBL at the indicated days post LCMV, LM, or VacV infection. Fold difference in percentage of progeny generated from earlym and latem in PBL at d90 is indicated. (D) Ratio of latem to earlym cells generated from transferred P14 cells in the indicated organs after LCMV infection at the height of the effector response (d8- left) or at a memory time point (d30-right). Dotted lines indicate the starting ratio of latem to earlym P14 cells. R: ratio of latem to earlym P14 cells. * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Representative data from one of three individual experiments with 5 mice per group per experiment. Error bars represent the standard error of the mean. 73

94 74

95 Figure 6: With Time, Memory CD8 T Cells Convert to CD62L hi and Become Enriched for T DIM S Analysis was performed at the indicated days post LCMV infection or on CD62L hi earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells. (A) Representative histograms showing CD62L expression on gated P14 cells isolated from spleens at the indicated days p.i. (B) Representative histograms showing IFN-γ production by gated CD62L lo or CD62L hi earlym (open histograms) or latem (shaded histograms) P14 cells isolated from spleens as determined by ICS following 5 hour incubation with or without GP peptide. Numbers inside plots indicate the percentage of earlym (top) or latem (bottom) cells not producing IFN-γ. Representative data from one of three individual experiments with 3 mice per group per experiment. 75

96 76

97 Figure 7: Phenotype of CD62L hi Memory CD8 T Cells Changes with Time after Infection (A) Representative histograms of CD27, CD127, and CD122 expression on gated CD62L hi earlym (open histogram) and latem (grey histogram) P14 cells in the indicated organs. Numbers inside histograms indicate the percentage of CD62L hi earlym (top) or latem (bottom) cells staining positive for the indicated marker. (B) Representative histograms of CD127, CD27, CD122, and KLRG1 expression on gated CD62L hi endogenous GP 33 or GP 276 tetramer positive earlym (open histogram) and latem (grey histogram) cells in the spleen. Numbers inside histograms indicate the percentage of CD62L hi earlym (top) or latem (bottom) cells staining positive for the indicated marker. Representative data from one of three individual experiments with 3 mice per group per experiment. 77

98 78

99 Figure 8: Phenotypic Heterogeneity of Memory CD8 T Cells Decreases with Time after Infection earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells from the indicated organs were co-stained for CD62L, CD27, CD122, CD127, KLRG1, and CD11b. (A) Percentages of subpopulations out of total CD62Lhi earlym or latem P14 cells. Surface marker expression patterns for the 32 possible subpopulations are indicated in the figure legend. (B) Number of subpopulations (out of 32 possible) comprising greater than 1% of the total CD62L hi P14 population for earlym and latem P14 cells in the indicated organs. * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Representative data from one of three individual experiments with 3 mice per group per experiment. Error bars represent the standard error of the mean. 79

100 80

101 Figure 9: Phenotypic Heterogeneity of Endogenous and P14 Memory CD8 T Cells Decreases with Time after Infection (A) Percentages of subpopulations out of total CD62L hi endogenous GP 33 or GP 276 tetramer positive earlym and latem cells co-stained for CD27, CD122, CD127, and KLRG1. Surface marker expression patterns for the 16 possible subpopulations are indicated in the figure legend. (B) Number of subpopulations (out of 16 possible) comprising greater than 1% of the total CD62L hi endogenous GP 33 or GP 276 population for earlym and latem cells in the spleen. (C) P14 cells were co-stained for CD62L, CD27, CxCr3, CD127, CCR5, and CD43. Percentages of subpopulations out of total CD62L hi earlym or latem P14 cells. Surface marker expression patterns for the 32 possible subpopulations are indicated in the figure legend. * statistically significant (p<0.05) as determined by Student t-test. Representative data from one of three individual experiments with 3 mice per group per experiment. Error bars represent the standard error of the mean. 81

102 82

103 Figure 10: Gene Expression Patterns among CD62L hi Memory CD8 T Cells Change with Time after Infection mrna was isolated from sorted CD62L hi earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells and used for microarray hybridization. (A) Heat map of genes with significantly different mrna expression between earlym and latem CD62L hi P14 cells with a fold change >1.25. (B) Biological pathway analysis of genes with significant mrna changes of fold >1.5 in CD62L hi earlym and latem P14 cells was generated using the KEGG pathway tool in DAVID bioinformatics resources., genes with increased expression in CD62L hi latem compared to earlym P14 cells., genes with decreased expression in CD62L hi latem compared to earlym P14 cells. Data were generated from six separate microarray hybridizations performed using mrna extracted from sorted earlym and latem P14 cells from three individual mice per group. 83

104 84

105 Table 1: Functional Annotation of Genes Differentially Expressed between CD62L hi earlym and latem CD8 T Cells Genes with significantly different gene expression at a fold of >1.5 between CD62L hi earlym and latem cells were analyzed using the functional annotation tools in DAVID bioinformatics resources. Genes were categorized based on biological function. Positive fold changes indicate higher expression in CD62L hi latem compared to earlym cells, while negative fold changes indicate higher expression in CD62L hi earlym compared to latem cells. 85

106 86

107 87

108 88

109 Figure 11: Cytokine Production, Degranulation, and Functional Avidity of CD62L hi Memory are not Influenced by Time Analysis was performed on CD62L hi earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells. (A) Representative histograms from gated CD62L hi earlym or latem P14 cells showing IFN-γ, TNF-α, or IL-2 production as measured by ICS, or surface CD107a expression as a measure of degranulation following 5 hour incubation with GP peptide. (B) Percentage of gated CD62L hi earlym or latem P14 cells producing IFN-γ, TNF-α, or IL-2 or undergoing degranulation. (C) Functional avidity curves for CD62L hi earlym (open squares) and latem (black squares) P14 cells was determined by ICS following 5 hour incubation of cells with the indicated concentrations of GP peptide. (D) EC50 for CD62L hi earlym and latem P14 cells as determined from functional avidity curves. (E) Time required for CD62L hi earlym (open squares) and latem (black squares) P14 cells to produce IFN-γ, TNF-α, or IL-2 was determined by ICS following incubation with GP peptide for the indicated lengths of time. NS not statistically significant as determined by Student t-test. Representative data from one of three individual experiments with 3 mice per group per experiment. Error bars represent the standard error of the mean. 89

110 90

111 Figure 12: CD62L hi earlym and latem Cells Exhibit a Similarly Poised State for Effector Responses following Ag Re-encounter (A) mrna was extracted from sorted CD62L hi earlym and latem P14 cells that were incubated for 5 hours in the presence (+) or absence (-) of GP peptide. mrna expression of the indicated effector molecules. Expression is relative to HPRT1. (B) earlym and latem P14 cells were mixed and incubated for 5 hours in the presence (+) or absence (-) of GP peptide. Percentage of earlym or latem P14 cells producing the indicated cytokines. Data from one experiment with 3 mice per group per experiment. Error bars represent the standard error of the mean. 91

112 92

113 Figure 13: CD62L hi earlym and latem Cells Regulate Expression of Surface/Activation Markers Similarly following Ag Re-encounter (A) mrna was extracted from sorted CD62L hi earlym and latem P14 cells that were incubated for 5 hours in the presence (+) or absence (-) of GP peptide. mrna expression of the indicated surface/activation markers. Expression is relative to HPRT1. (B) earlym and latem P14 cells were mixed and incubated for 5 hours in the presence (+) or absence (-) of GP peptide. Percentage of earlym or latem P14 cells expressing the indicated surface/activation markers. Data from one experiment with 3 mice per group per experiment. Error bars represent the standard error of the mean. 93

114 94

115 Figure 14: CD62L hi earlym and latem Cells Regulate Expression of Transcription Factors Similarly following Ag Re-encounter (A) mrna was extracted from sorted CD62L hi earlym and latem P14 cells that were incubated for 5 hours in the presence (+) or absence (-) of GP peptide. mrna expression of the indicated transcription factors. Expression is relative to HPRT1. (B) earlym and latem P14 cells were mixed and incubated for 5 hours in the presence (+) or absence (-) of GP peptide. Percentage of earlym or latem P14 cells expressing (top), and per cell expression based on gmfi (bottom), of the indicated transcription factors. Data from one experiment with 3 mice per group per experiment. Error bars represent the standard error of the mean. 95

116 96

117 Figure 15: CD62L hi earlym and latem Cells Regulate Expression of Cell Cycle/Survival Genes Similarly following Ag Re-encounter (A) mrna was extracted from sorted CD62L hi earlym and latem P14 cells that were incubated for 5 hours in the presence (+) or absence (-) of GP peptide. mrna expression of the indicated cell cycle/survival associated genes. Expression is relative to HPRT1. Data from one experiment with 3 mice per group per experiment. Error bars represent the standard error of the mean. 97

118 98

119 Figure 16: Cell Cycle Pathways are Differentially Regulated in CD62L hi Memory CD8 T Cells with Time mrna was isolated from sorted CD62L hi earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells and used for microarray hybridization. (A) Biological pathway analysis of genes with significant mrna changes of fold >1.5 in CD62L hi earlym and latem P14 cells was generated using the KEGG pathway tool in DAVID bioinformatics resources. Shown is relative gene expression for genes involved in the KEGG cell cycle pathway that are differentially expressed between CD62L hi earlym and latem P14 cells. Positive fold changes represent genes with increased expression in CD62L hi latem compared to earlym P14 cells while negative fold changes represent genes with decreased expression in CD62L hi earlym compared to latem P14 cells. (B) Gene set enrichment analysis was performed comparing expression of genes in CD62L hi earlym and latem P14 cells to existing gene sets. latem P14 cells showed enrichment for several gene sets involved in cell cycling. 99

120 100

121 Figure 17: CD62L hi latem Cells Enter Cell Cycle Faster than earlym Cells following Ag Re-encounter (A) earlym and latem P14 cells were mixed and incubated in the presence of GP peptide. Percentage of earlym or latem P14 cells staining positive for KI67 following incubation with GP peptide for the indicated lengths of time. ** statistically significant (p<0.01) as determined by Student t-test. Data from one experiment with 3 mice per group per experiment. Error bars represent the standard error of the mean. 101

122 102

123 Figure 18: Sensitivity to IL-15 and Ability to Undergo Homeostatic Proliferation for CD62L hi Memory Changes with Time Analysis was performed on CD62L hi earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells. (A) A comparison of relative mrna expression of the indicated components of the IL-15R complex between CD62L hi earlym and latem P14 cells from microarray data. Positive fold changes represent genes with increased expression in CD62L hi latem compared to earlym P14 cells. (B) Representative histograms of CD122 expression on gated CD62L hi earlym (open histogram) and latem (grey histogram) P14 cells. (C) Representative histograms of CFSE dilution on gated CD62L hi earlym (open histograms) or latem (grey histograms) P14 cells after 3 day culture in the presence or absence (black histograms) of the indicated concentrations of IL-15. (D) CD62L hi earlym and latem P14 cells were sorted and co-transferred into Rag-/- mice. Left. Dot plot showing the master mix of earlym (Thy1.1/1.2) and latem (Thy1.1/1.1) P14 cells used for adoptive transfer. Right. Representative dot plot of transferred CD62L hi earlym (Thy1.1/1.2) and latem (Thy1.1/1.1) P14 cells isolated from spleens of Rag-/- mice d15 after adoptive transfer. R: ratio of CD62L hi latem to earlym P14 cells. 103

124 104

125 Figure 19: Proliferation and Memory Generation Potential of CD62L hi Memory Cells Increases with Time after Infection Sorted CD62L hi Thy disparate earlym and latem P14 cells were mixed and injected into naïve recipients followed by i.p. injection of LCMV Armstrong 24 hours later. (A) Dot plot showing the master mix of CD62L hi earlym (Thy1.1/1.1) and latem (Thy1.1/1.2) P14 cells used for adoptive transfer. (B) Representative plot showing the response of progeny of sorted CD62L hi earlym (Thy1.1/1.1) and latem (Thy1.1/1.2) P14 cells in PBL d8 after LCMV infection. (C) Kinetic analysis of secondary responses generated from sorted CD62L hi earlym (open squares) and latem (black squares) P14 cells in PBL at the indicated days post LCMV infection. Fold difference in percentage of progeny generated from CD62L hi earlym and latem in PBL at d91 is indicated. (D) Ratio of cells generated from sorted CD62L hi latem to earlym P14 cells in PBL at the indicated days post LCMV infection. Doted line indicates the starting ratio of CD62L hi latem to earlym P14 cells. (E) Ratio of cells generated from sorted CD62L hi latem to earlym P14 cells in the indicated organs at a memory time point (d140) following LCMV infection. Doted line indicates the starting ratio of CD62L hi latem to earlym P14 cells. R:, ratio of CD62L hi latem to earlym P14 cells or progeny of CD62L hi latem to earlym P14 cells.* statistically significant (p<0.05) as determined by Student t-test. Representative data from one of two individual experiments with 5 mice per group per experiment. Error bars represent the standard error of the mean. 105

126 106

127 Figure 20: Ability to Undergo Ag-driven Proliferation Increases in Endogenous Memory CD8 T Cells with Time after Infection (A) Experimental design. 2.5x10 3 endogenous GP 276 tetramer positive cells among sorted CD62L hi CD8 T cells from the spleens of mice infected 1 month (earlym) or >8 months (latem) previously with LCMV were transferred into naïve C57B6/SJL mice. Recipient mice were infected 24 hours later with LCMV. (B) Representative dot plots from the spleens of recipient mice 7 days post LCMV infection showing gating of transferred cells (CD45.2) and 2 o effector GP cells generated from transferred CD62L hi 1 o earlym or latem GP cells. (C) (left) Percentage of 2 o effector cells out of total lymphocytes generated from transferred CD62L hi earlym or latem endogenous GP 276 tetramer positive cells in PBL of recipient mice 7 days post LCMV infection. (right) Total numbers of 2 o effector cells generated from transferred CD62L hi earlym or latem endogenous GP 276 tetramer positive cells recovered from the spleens of recipient mice 7 days post LCMV infection. Fold difference in numbers of progeny generated from CD62L hi earlym and latem cells detected in the spleen is indicated. Data from one of two individual experiments with 3-5 mice per group per experiment. Error bars represent the standard error of the mean. 107

128 108

129 Figure 21: latem Cells Provide Better Protection than earlym Cells following LCMV Clone-13 Infection (A-D) FACS sorted CD62L hi earlym (30-45 days p.i.) or latem (8+ months p.i.) P14 cells were injected into naïve recipients followed by i.v. injection of LM or LCMV clone hours later. (E-F) earlym (30-45 days p.i.) or latem (8+ months p.i.) populations of P14 cells positively selected on AutoMacs columns were injected into naïve recipients followed by i.v. injection of LM or LCMV clone hours later. (A and E) Bacterial CFUs in the spleen were measured three days after LM infection for mice that received no cells (naïve), or adoptive transfer of 7x10 4 sorted CD62L hi earlym or latem P14 cells (A) or 7x10 4 positively selected earlym or latem P14 cell populations (E). (B and F) Viral PFUs in the spleen were measured 8 days after LCMV clone-13 infection for mice that received no cells (naïve), or adoptive transfer of 5x10 4 sorted CD62L hi earlym or latem P14 cells (B) or 7x10 4 positively selected earlym or latem P14 cell populations (F). (C and G) Kinetic analysis of secondary responses generated from sorted CD62L hi earlym (open squares) or latem (black squares) P14 cells (C) and secondary responses generated from positively selected earlym (open squares) or latem (black squares) P14 cell populations (G) in PBL at the indicated days post LCMV clone-13 infection. (D and H) Numbers of progeny P14 cells generated from transferred CD62L hi earlym or latem cells (D) and transferred earlym or latem cell populations (H) found in the spleens of recipient mice 8 days post LCMV clone-13 infection. NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test or ANOVA with a Bonferroni post-test. Representative data from one of two to three individual experiments with 3 mice per group per experiment. Error bars represent the standard error of the mean. 109

130 110

131 Figure 22: Mitochondrial Function of Memory CD8 T Cells Increases with Time after Infection (A) Biological pathway analysis based on microarray data of genes with significant mrna changes of fold >1.5 in CD62L hi earlym and latem P14 cells was generated using the KEGG pathway tool in DAVID bioinformatics resources. Shown is relative gene expression for genes involved in the KEGG oxidative phosphorylation pathway that are differentially expressed between CD62L hi earlym and latem P14 cells. Positive fold changes represent genes with increased expression in CD62L hi latem compared to earlym P14 cells while negative fold changes represent genes with decreased expression in CD62L hi latem compared to earlym P14 cells. (B) Gene set enrichment analysis was performed comparing expression of genes in CD62L hi earlym and latem P14 cells to existing gene sets. latem cells showed enrichment for oxidative phosphorylation gene sets. (C-F) earlym (30-45 days p.i.) and latem (8+ months p.i.) populations or CD62L hi P14 cells were purified and 2x10 5 cells were plated in XF media and analyzed with an XF-96 extracellular flux analyzer (Seahorse Bioscience). (C) Ratio of O 2 consumption rates (OCR) to extracellular acidification rates (ECAR) measured under basal conditions for CD62L hi earlym or latem P14 cells. (D) OCR was measured under basal conditions and in response to the indicated mitochondrial inhibitors for earlym or latem P14 cell populations. (E) Ratio of OCR to ECAR measured under basal conditions for earlym or latem P14 cell populations. (F) SRC of earlym or latem P14 cell populations as indicated by maximum OCR following FCCP injection as a percentage of maximum OCR under basal conditions. NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Combined data from four individual experiments that provided similar results with 3 mice per group per experiment. Error bars represent the standard error of the mean. 111

132 112

133 CHAPTER III: ANTIGEN-DEPENDENT AND INDEPENDENT CONTRIBUTIONS TO PRIMARY MEMORY CD8 T CELL ACTIVATION AND PROTECTION FOLLOWING INFECTION Abstract When memory CD8 T cells detect infection, they become activated and produce effector molecules including IFN-γ and granzymeb that can contribute to clearance of invading pathogens. Activation can be induced not only by Ag-dependent signals through the TCR, but also by pathogen derived inflammatory cytokines in an Ag-independent manner. However, it is currently unclear if the ability of memory CD8 T cells to sense inflammation and to become activated in an Ag-independent manner is altered with time after they are generated, how pathogen derived inflammatory cytokine signals contribute to Ag-driven memory CD8 T cell activation following infection, and whether Agindependent memory CD8 T cell responses during unrelated infections (bystander activation) provide the host with a protective benefit. To investigate these questions, I examined activation of virus-specific primary memory CD8 T cells in vivo following infection with L. monocytogenes either expressing or not expressing cognate Ag, and in vitro in response to inflammatory cytokines. I show that the ability of memory CD8 T cells to sense inflammation and to become activated either in response to inflammatory cytokines in vitro, or in response to unrelated infection in vivo decreases with time after memory generation. However, Ag and inflammation acted synergistically in vitro to induce memory CD8 T cell activation, and activation of earlym and latem CD8 T cells was equally robust in response to Ag and inflammation. As was suggested by in vitro evidence of synergy between Ag and inflammation for CD8 T cell activation, in vivo, I 113

134 found that when memory CD8 T cells significantly contribute to clearance of infection, early activation and continued responses by memory CD8 T cells are enhanced by cognate Ag recognition. Additionally, I show that bystander responses by memory CD8 T cells are dependent upon the dose of infection and inflammation elicited, and that they provide protection in IFN-γ-deficient mice, but not in immuno-competent hosts. The data suggest that Ag-independent memory CD8 T cell activation is more tightly regulated with time after infection, and elucidate the requirements for Ag-driven memory CD8 T cell activation following infection and the protective role of bystander memory CD8 T cell responses. Introduction Following infection, memory CD8 T cells become activated and produce effector molecules providing immune hosts with enhanced protection against invading microorganisms [7]. Therefore, understanding the requirements for memory CD8 T cell re-activation may aid in the design of protective vaccines. It is well established that memory CD8 T cell production of cytokines and lytic molecules is induced downstream of TCR signaling upon cognate Ag recognition. However, memory CD8 T cells can be activated and produce IFN-γ and GrB in a cytokine-dependent, Ag-independent manner [38,44-47] (Figure 2). It was recently reported that exhausted CD8 T cells down-regulate the IL-18 receptor and become unresponsive to inflammatory cytokines and bacterial coinfections [49], suggesting that Ag-independent functions of memory CD8 T cells can be regulated in a CD8 T cell intrinsic manner. I recently reported that many functions of 114

135 memory CD8 T cells change with time after infection [173]. However, whether the Agindependent functions of memory CD8 T cells are altered with time after memory generation is currently unknown. Furthermore, because memory CD8 T cells are capable of being activated in both Ag-dependent and Ag-independent fashions, activation following infection with pathogens expressing Ags recognized by memory CD8 T cells could be driven by either cognate Ag recognition or by inflammatory cytokines. Recent studies have come to differing conclusions as to the relative importance of cognate Ag recognition for eliciting memory CD8 T cell functions. One recent report indicated that early activation of circulating memory CD8 T cells occurs independently of cognate Ag recognition [39]. However, a different study indicated that the ability of tissue-resident memory CD8 T cells to sense infection resulting in the production of IFN-γ, local chemokine production, and the recruitment of immune cells, requires cognate Ag recognition [62]. While these conclusions appear to be contradictory, they both could be true if the influence of cognate Ag on memory CD8 T cell activation is context dependent. However, these studies indicate that the contribution of Ag-dependent and independent signals to memory CD8 T cell reactivation has yet to be clearly defined. Cytokine production and targeted killing of infected cells promote clearance of pathogens recognized by memory CD8 T cells following infection. In the same manner, effector cytokines and lytic molecules produced by memory CD8 T cells activated in an Ag-independent manner (i.e. bystander activation) could provide protection against nonrelated infections. Bystander responses have been shown to provide protection against infection with L. monocytogenes or S. typhimurium in IFN-γ deficient or Natural Killer 115

136 (NK) cell depleted hosts [38,40,174,175], suggesting that bystander responses could represent an important protective immune response upon infection with diverse pathogens. However, experiments examining the protective role of bystander CD8 T cell responses in immuno-competent hosts have yielded conflicting results. One report has indicated that bystander responses following bacterial infection are protective [39], while another report has indicated that they provide little to no protection [176]. Additionally, multiple studies examining viral infection have indicated that only memory CD8 T cells that recognize Ag due to TCR cross-reactivity are able to provide protection against infection with unrelated viruses [177]. Therefore, it is unclear if bystander responses by memory CD8 T cells provide protection in immuno-competent hosts. In this study I address the influence of time after memory generation on Agindependent memory CD8 T cell responses, the contribution of Ag and inflammation to memory CD8 T cell activation, and protection provided by virus-specific bystander memory CD8 T cells following LM infection. I show that with time after memory generation, the ability of memory CD8 T cells to sense and respond to inflammation in an Ag-independent manner decreases. However, I show that Ag and inflammatory cytokines synergize in vitro to induce memory CD8 T cell activation. Interestingly, both earlym and latem CD8 T cells were able to robustly respond in the presence of both Ag and inflammation, suggesting that activation-induced effector functions are more tightly regulated in memory CD8 T cells with time after their generation. Furthermore, as was suggested by the in vitro observation that Ag and inflammation synergize to induce memory CD8 T cell activation, in vivo responses by memory CD8 T cells that significantly contributed to clearance of infection were enhanced upon infection with 116

137 pathogens expressing cognate Ag. Memory CD8 T cell bystander responses during unrelated infection were dependent upon the dose of infection and the level of inflammation elicited following infection, and in agreement with previous literature, they were protective in IFN-γ deficient hosts. However, bystander memory CD8 T cell responses in immuno-competent hosts provided no protection. My findings show that with time after their generation the ability of memory CD8 T cells to respond to nonspecific infection decreases while their ability to respond to infections bearing Ags they recognize is retained, and elucidate the role of Ag in memory CD8 T cell activation, and protection provided by bystander memory CD8 T cell responses following non-related bacterial infections. Materials and Methods Mice C57BL/6 (B6), IFN-γ knockout, and P14 mice were bred and maintained in the animal facilities at the University of Iowa at the appropriate biosafety levels. All animal studies were approved by the University of Iowa Institutional Animal Care and Use Committee and met stipulations of the Guide for the Care and Use of Laboratory Animals (National Institutes of Health). All animal studies were carried out in accordance with these approved guidelines. 117

138 Bacterial and Viral Infections, Memory Generation, Adoptive Transfers, and Ampicillin Treatment The Armstrong strain of LCMV and attenuated acta- deficient (Att LM) strain DP-L1942 or virulent Listeria monocytogenes (Vir LM) strain 1043S expressing and not expressing the GP 33 epitope were grown and quantified as previously described [10,103]. Briefly, stock solutions of bacteria were incubated for 2-3 hours in TSB at 37 o C with light shaking. Bacterial colonies/ml were determined based upon OD readings at 600nm using a Genesys 10UV (Thermo Scientific). Bacteria were re-suspended in saline solution at the desired colonies /ml, and 200µL was i.v. injected per mouse. Stock viral strains were diluted to the desired concentration in PBS, and 200µL was i.p. injected per mouse.1 o memory P14 cells were generated by adoptively transferring 5x10 3 P14 cells obtained from peripheral blood of naïve P14 mice (Thy1.1) into B6 recipients (Thy1.2) followed by infection with either Vir LM-GP 33 (1x10 4 CFU i.v.) or LCMV (2x10 5 PFU i.p.). All earlym analysis was done between days after infection and latem analysis was done 8+ months after infection. 2 o infections with Att LM (5x10 6 CFU/mouse) or Vir LM (1x10 4, 1x10 5, or 1x10 6 CFU/mouse) either expressing or not expressing GP 33 were administered i.v. >30 days after 1 o infection. For adoptive transfer of 1 o memory cells, splenocytes from mice infected >30 days previously with LCMV were stained with PEanti Thy1.1 antibodies and purified with anti-pe magnetic bead sorting using standard AutoMacs protocols. B6 or GKO mice received adoptive transfer of 400,000 1 o memory P14 cells i.v. As a measure of protection, body weight was monitored daily, and CFUs were determined in the spleens and liver at the indicated times post 2 o infection [141]. 118

139 Ampicillin treated mice received i.p. injection (2mg in H 2 O) and drinking water (2mg/ml) for 24 hours [178]. Detection of Memory Cells and ex vivo P14 Cell Analysis Total memory CD8 T cells were detected using the surrogate activation markers CD8 and CD11a [179]. Briefly, cells were surface stained for CD8 and CD11a, and Ag experienced cells were detected based upon up-regulation of CD11a and down-regulation of CD8. Endogenous GP 33 -specific memory CD8 T cells were detected with MHC class I GP 33 tetramers, and memory P14 cells were gated on CD8/Thy1.1+ cells. Ex vivo activation status following infection was determined by incubating splenocytes at 37 o C for 1 hour in the presence of BFA followed by intracellular staining with Abs against either IFN-γ or GrB or extracellular staining with Abs against either CD25 or CD69. All cells for flow cytometry were analyzed using a FACSCanto flow cytometer and analyzed using FlowJo software. In vitro Activation of P14 Cells Splenocytes were isolated from mice containing memory P14 cells generated following LCMV infection and were incubated for 4 hours in the presence of media, GP 33 peptide (0.01 or 200nM), ril-12 and IL-18 or IL-12 and TNF-α or IL-18 and IFN-β (0.5ng/mL each), or 0.01nM GP 33 peptide and 0.5ng/mL each of recombinant cytokine combinations. Cells were incubated for 1 additional hour in the presence of BFA before being stained extracellularly with Abs against CD25 or CD69, or intracellularly with Abs 119

140 against IFN-γ. Thy disparate earlym and latem CD8 T cells were combined and incubated together and distinguished based on expression of the Thy marker (Thy1.1/1.1 or Thy1.1/1.2) Detection of Serum IL-12 Blood samples were collected by retroorbital bleeding from mice 4, 8, and 24 hours post 2 o infection and from mice that did not receive 2 o infection, and serum was collected after centrifugation. IL-12 was measured using a mouse IL-12 platinum ELISA (ebioscience), and absorbance values (450 nm) were measured and assessed using Gen5 software (BioTek). Microarray Data Acquisition and Analysis earlym and latem P14 cells were isolated from spleens and cells were surface stained for Thy1.1 and CD8 (ebioscience) and sorted using a BD FACSAria II (BD Biosciences). Samples from three individual mice were obtained for each group, and sorted cells were >95% pure. RNA was extracted using the RNEasy Kit (QIAGEN), and 1-5ng of mrna was used for microarray analysis. RNA quality was assessed using the Agilent Model 2100 Bioanalyzer. mrna for the microarray was processed using the NuGEN WT-Ovation Pico RNA Amplification System along with the NuGEN WT- Ovation Exon Module. Samples were hybridized and loaded onto Affymetrix GeneChip Mouse 1.0 ST arrays. Arrays were scanned with the Affymetrix Model 7G upgraded scanner, and data were collected using the GeneChip Operating Software. Data from the 120

141 Affymetrix Mouse Exon 1.0 ST arrays were first quantile normalized and median polished using Robust Multichip Average background correction with log2 adjusted values. Probe sets for exons were then summarized for a specific gene using the median value. After obtaining log2 expression values for genes, significance testing was performed using analysis of variance (ANOVA). Statistical Analysis Statistical analyses were performed using GraphPad Prism software version 6 (GraphPad Software Inc., San Diego, CA). Statistical comparisons of two groups were done using the Mann-Whitney test and the unpaired t test. Statistical comparisons of more than two groups were done using one-way ANOVA with the Bonferroni posttest. R squared values were calculated from linear regression analysis. Results Ability of Memory CD8 T Cells to Undergo Bystander Activation Decreases with Time after Memory Generation Memory CD8 T cells that were functionally exhausted due to chronic viral infection were recently shown to have a decreased ability to respond to inflammatory cytokines or during un-related infections [49], suggesting that Ag-independent functions of memory CD8 T cells are regulated in a cell-intrinsic manner based on the quality of the memory CD8 T cell. I recently demonstrated that many functions of memory CD8 T 121

142 cells change with time after infection [173]. Microarray analysis of the transcriptomes of populations of earlym and latem CD8 T cells indicated that components of the IL-12 and IL-18 receptors decrease with time after infection (Figure 23), suggesting that the ability of memory CD8 T cells to sense inflammation and to undergo bystander activation during unrelated infections may change with time after memory generation. To examine this, I developed an in vivo bystander activation model in which endogenous LCMVspecific and Tg P14 memory CD8 T cells were generated in response to LCMV infection. At either 1 month (earlym) or 8+ months (latem) post LCMV infection, mice were given a secondary infection with Vir LM to induce bystander memory CD8 T cell responses, and bystander activation of endogenous LCMV-specific and transferred Tg P14 memory CD8 T cells was monitored based on ex vivo IFN-γ production 20 hours after LM infection (Figure 24A). Down-regulation of CD8α and up-regulation of CD11a has been used to identify Ag experienced cells [179], and both Tg memory P14 and endogenous CD11a hi /CD8a lo memory CD8 T cells were able to undergo bystander activation while endogenous CD11a lo /CD8a hi naïve CD8 T cells were not (Figure 24B,C). However, a decreased percentage of both latem Tg P14 and endogenous LCMV-specific memory CD8 T cells compared to earlym cells produced IFN-γ in response to unrelated LM infection (Figure 24B,C). These data suggest that the ability of memory CD8 T cells to undergo bystander activation decreases with time after their generation. 122

143 Ability of Memory CD8 T Cells to Sense Inflammation Decreases with Time after Infection Microarray data showed that mrna levels for components of receptors recognizing the inflammatory cytokines IL-12 and IL-18 decrease in memory CD8 T cells with time after infection (Figure 23), suggesting that the decreased ability of latem CD8 T cells to respond during unrelated infections may be due to a decreased ability to sense inflammation. To test this, I co-incubated Thy disparate earlym and latem P14 cells with inflammatory cytokines known to elicit activation of memory CD8 T cells [38,44-47,50]. A significantly decreased percentage of latem compared to earlym cells produced IFN-γ in response to all cytokine combinations tested (Figure 25A,B), suggesting that the ability of memory CD8 T cells to generally sense inflammation decreases with time after infection. Taken together, the data in Figures show that the ability of memory CD8 T cells to sense inflammation and to respond in an Agindependent manner during unrelated infections decreases with time after their generation. Ag and Inflammation Act Synergistically in vitro to Induce Memory CD8 T Cell Activation Upon infection with pathogens bearing relevant memory CD8 T cell epitopes, CD8 T cell activation could be driven by either cognate Ag recognition or by inflammation elicited during the infection. Furthermore, due to the decreased ability of latem CD8 T cells to sense inflammation, their activation following relevant infections 123

144 could be compromised if inflammation is important for driving Ag-specific memory CD8 T cell responses. To first determine how Ag and inflammation might interact to influence memory CD8 T cell activation during infection, I devised an in vitro system that allowed me to examine their effects on memory CD8 T cell activation separately, or in combination. At the very onset of infection, Ag and inflammation are present at low levels. I therefore incubated latem P14 cells with low concentrations of inflammatory cytokines, low concentrations of cognate Ag, or a combination of cytokines and Ag. Less than 10% of memory CD8 T cells that were capable of responding to Ag (Figure 26A left panels) became activated following incubation with low concentrations of GP 33 peptide or recombinant (r)il-12 and IL-18 alone (Figure 26A,B). However, a large percentage of memory CD8 T cells produced IFN-γ and expressed the activation markers CD25 and CD69 when incubated with low levels of GP 33 peptide and ril-12 and IL-18 (Figure 26A,B), or ril-12 and TNF-α or ril-18 and IFN-β (Figure 26C). These data suggest that Ag and inflammation have the capacity to synergize to induce CD8 T cell activation. Activation of latem CD8 T Cells is more Tightly Regulated, but they Remain Fully Capable of Responding to Infection Expressing Recognized Ags Because earlym cells had an increased ability to sense inflammation compared to latem CD8 T cells, I next sought to determine if activation of earlym and latem cells in response to Ag and inflammation was similarly regulated. I therefore co-incubated earlym and latem cells with inflammatory cytokines, with low concentrations of GP 33 peptide, or with a combination of cytokines and GP 33 peptide. As I had previously noted 124

145 (Figure 25), earlym cells have an increased ability to sense inflammation based on the higher percentage of earlym compared to latem cells that produce IFN-γ following incubation with inflammatory cytokines (Figure 27A,B). However, upon incubation with inflammatory cytokines and with low concentrations of cognate Ag that induce only a small percentage of memory CD8 T cells to produce IFN-γ, earlym and latem CD8 T cells both robustly produced IFN-γ, and latem cells were equally capable of IFNγ production as earlym cells (Figure 27A,B). Thus, while latem CD8 T cells are less capable of responding to inflammatory cytokines by themselves or during unrelated infections, they remain capable of robust responses in the presence of inflammation and cognate Ag. This data suggests that activation of memory CD8 T cells is more tightly regulated with time after they are generated, but that they are fully capable of robust responses following infection with pathogens expressing recognized Ags. Early Activation of Memory CD8 T Cells that do not Significantly Contribute to Clearance of Infection is not Influenced by Cognate Ag A recent study concluded that early activation of memory CD8 T cells during infection is not influenced by the presence of cognate Ag [39]. However, my in vitro findings suggested that cognate Ag and inflammation might synergize to enhance memory CD8 T cell responses during re-infection. In order to further explore the contribution of Ag and inflammation to memory CD8 T cell activation and to attempt to explain why cognate Ag does not influence early activation of memory CD8 T cells using a system similar to that used by Soudja et al. [39], I generated memory P14 cells by 125

146 adoptively transferring small numbers of naïve P14 cells into naïve B6 mice that were then infected with LM expressing the GP 33 epitope. Because I had shown that earlym cells are more responsive to activation induced by inflammatory cytokines, and because I was attempting to produce similar data to that which concluded that inflammation rather than Ag drives memory CD8 T cell activation, mice were given a heterologous infection at an earlym time point with LM either expressing or not expressing GP 33 (Figure 28A). In this system, P14 cell memory responses occur in the presence of LM-specific memory responses and in the context of an infection either expressing cognate Ag (LM +Ag), or in the absence of cognate Ag (LM Ag). Similarly to Soudja et al., I found that the percentage of cells producing IFN-γ or GrB, or expressing the activation markers CD25 or CD69 was similar regardless of the presence of cognate Ag (Figure 28B). However, in this system, memory P14 cells constituted a minor fraction of the total pathogen-specific response (Figure 28C). Additionally, while prior infection with LM led to a reduction in bacterial burdens, responses by P14 cells did not significantly enhance LM clearance beyond that mediated by the large LM-specific memory CD8 T cell population (Figure 28D). These data suggest that cognate Ag recognition is not required for activation of memory CD8 T cells that do not significantly contribute to clearance of infection. Cognate Ag and Continued Infection Drive Early Activation and Sustained Responses of Memory CD8 T Cells that Contribute to Pathogen Clearance In contrast to bacterial and/or viral infection which elicits a large polyclonal memory population that is capable of responding and contributing to pathogen clearance during re-infection, subunit vaccination results in the generation of memory CD8 T cells 126

147 recognizing one or a small number of Ags present in invading microbes. To examine how Ag and inflammation influence the activation of a population of memory CD8 T cells specific for a single Ag that respond in the absence of a large polyclonal memory response, I generated memory P14 cells following infection with LCMV. At a memory time point, mice were given a 2 o infection with Vir LM either expressing or not expressing GP 33 (Figure 29A). In this system, P14 cells and endogenous GP 33 -specific cells are the only Ag-specific memory cells present, and P14 cell responses occur in the context of an infection either expressing (LM +Ag) or not expressing (LM Ag) cognate Ag. Clearance of LM was mediated by memory P14 cells, as mice infected with LM expressing GP 33 displayed significantly lower CFUs in the spleen 16 and 24 hours post infection (Figure 29B). Importantly, IFN-γ production by memory P14 cells occurred faster in response to LM expressing GP 33, and responses waned as infection was cleared (Figure 29C,D). P14 responses were truly driven by the infection, as less than 4% of P14 cells from mice that did not receive a 2 o LM infection stained positive for IFN-γ, GrB, CD25, or CD69 (Figure 29E). These data suggest that early activation of memory CD8 T cells is enhanced by cognate Ag recognition, and that continued responses by memory CD8 T cells are dependent upon continued infection. During the initial stages of infection with Vir LM, bacterial levels are low, resulting in low levels of bacterial Ag [180,181]. While memory CD8 T cell responses were significantly enhanced early following infection with Vir LM expressing GP 33 compared to infection not expressing GP 33, only a small percentage of memory CD8 T cells were activated at early time points (Figure 29C,D). In contrast to Vir LM infection, CD8 T cell responses peak earlier following infection with Att acta deficient LM, and 127

148 initial levels of bacteria and Ag are higher [180, ]. In order to examine the effects of Ag and inflammation on early memory CD8 T cell responses during an infection where levels of Ag are abundant, I generated memory P14 cells following LCMV infection and at a memory time point infected mice with Att LM either expressing or not expressing GP 33 (Figure 30A). While levels of bacteria were similar early after Att LM infection (Figure 30B), a greater percentage of memory CD8 T cells responding in the presence of cognate Ag were activated at early time points, and responses waned as infection was cleared (Figure 30C,D). Taken together, these data suggest that early activation of memory CD8 T cells is enhanced by cognate Ag recognition. Bystander Memory CD8 T Cell Responses are Influenced by Dose of Infection and Amount of Inflammation Elicited upon Infection I noticed that early (8 hours post infection) bystander activation of memory CD8 T cells was enhanced following Att LM infection (Figure 30C,D) compared to Vir LM infection (Figure 29C,D). Because the dose of infection differed in the two experiments, I reasoned that the magnitude of bystander memory CD8 T cell responses might be dependent upon the dose of infection. To test this, I infected LCMV-immune mice containing memory P14 cells with 5x10 6 CFU Att LM or from 1x10 4-1x10 6 CFU Vir LM. 30 minutes following infection, the recoverable CFUs from the spleens and livers of mice infected with 5x10 6 CFU Att LM was significantly higher than that of mice infected with 1x10 5 CFU Vir LM (Figure 31A). Corresponding to the higher initial dose, bystander IFN-γ production was detectable in mice infected with Att LM by 8 hours post infection while it was not detectable in mice infected with 1x10 4 or 1x10 5 CFU Vir LM 128

149 (Figure 31B). By 24 hours post infection when bystander responses were detectable in mice infected with Vir LM, the percentage of memory P14 cells producing IFN-γ in a bystander fashion increased with increasing dose of infection (Figure 31B). However, while the level of infection was not significantly different 8 hours post infection between mice that were infected with 1x10 6 CFU Vir LM, and 5x10 6 Att LM (Figure 31A), the percentage of bystander activated P14 cells was higher in mice infected with Att LM (Figure 31B). Because inflammatory cytokines have been shown to drive bystander activation of memory CD8 T cells in vitro [38,46], I next reasoned that inflammation elicited following infection might impact bystander responses of memory CD8 T cells in vivo. When I examined the systemic levels of IL-12 in the serum of infected mice in relation to mice that did not receive a 2 o LM infection, levels of IL-12 were significantly greater in mice that received Att LM infection than mice that received any dose of Vir LM infection at 4 and 8 hours post infection (Figure 31C). By 24 hours post infection when bystander IFN-γ production by memory CD8 T cells was highest in mice that received 5x10 6 CFU Att LM and 1x10 6 CFU Vir LM infection (Figure 31B), serum IL-12 levels were significantly greater in mice that received either of these two infections compared to mice that received 1x10 4 or 1x10 5 CFU Vir LM infection (Figure 31C). When linear regression analysis was used to examine the correlations between the dose of infection, the percentage of P14 cells producing IFN-γ in a bystander manner, and the level of systemic IL-12, statistically significant correlations were observed between the CFUs of bacteria recovered from the spleens 24 hours post infection and the percentage of P14 cells producing IFN-γ, the percentage of P14 cells producing IFN-γ and the levels of serum IL-12, and the CFUs of bacteria recovered from the spleens 24 hours post 129

150 infection and the levels of serum IL-12 (Figure 31D). Taken together, these data indicate that bystander responses by memory CD8 T cells are influenced by the dose of infection and the level of systemic cytokines elicited upon infection. Bystander Memory CD8 T Cell Responses Provide Protection in IFN-γ Deficient, but not Immuno-competent Hosts Non-Ag driven bystander responses by memory CD8 T cells persisted up to 24 hours following LM infection (Figures 29-31). To examine if bystander responses continue beyond 24 hours, IFN-γ and GrB production by P14 cells in LCMV immune mice was examined from 1-5 days after infection with Vir LM not expressing GP 33. Infection persisted up to 5 days following infection (Figure 32A), and IFN-γ and GrB producing bystander activated P14 cells could be detected at all time points examined (Figure 32B). However, the percentage of IFN-γ producing P14 cells decreased from day 1 to 5. Bystander responses were dependent on the level of infection, as reducing the level of infection through antibiotic treatment resulted in a decreased percentage of P14 cells producing IFN-γ (Figure 32C). Thus, bystander CD8 T cell responses persist for several days but wane following un-related infections, and responses depend on the level of infection. To examine if bystander responses by memory CD8 T cells are capable of providing protection during infection with unrelated pathogens, memory P14 cells were either transferred or not transferred into normal B6 or IFN-γ deficient mice that were infected with LM not expressing GP 33. As has been reported [122,123], IFN-γ is 130

151 important for clearance of primary LM infection, as bacterial burdens were 2-3 logs higher in the livers of IFN-γ knockout mice compared to B6 mice 2 days following infection (Figure 32D). Bacterial CFUs were significantly reduced in GKO mice that received transferred memory P14 cells indicating that in the absence of normal IFN-γ production, bystander CD8 T cells provide protection against unrelated infection. However, bacterial CFUs were not significantly different between B6 mice that received or did not receive transferred memory P14 cells (Figure 32D), indicating that bystander CD8 T cell responses provide minimal protection in mice with a full complement of IFNγ producing cells. Following infection, the number of memory CD8 T cells present in an intact mouse is greater than that present following adoptive transfer. To determine if bystander responses by a large population of memory CD8 T cells are protective against non-related infection with LM, naïve or LCMV immune mice were infected with LM expressing or not expressing GP 33. An approximately 1LD 50 dose of LM (1x10 4 CFU) was used in order to enable survival and analysis over an extended period of time. Ag-directed memory CD8 T cell responses provided protection, as LM-GP 33 infected mice lost less weight and cleared infection faster than either naïve mice or LCMV immune mice that were infected with LM not expressing GP 33 (Figure 30E). However, bystander memory CD8 T cell responses provided no significant protection, as naïve and LCMV-immune mice infected with LM not expressing GP 33 experienced similar weight loss and displayed similar bacterial burdens in the spleens 1-5 days post infection. IFN-γ production by bystander activated CD8 T cells may not provide any additional benefit beyond IFN-γ produced by other cells of the immune system following 131

152 non-lethal infection. In the case of a lethal infection, however, it is possible that IFN-γ in addition to that produced by other cell types could provide a survival advantage. To examine if bystander responses by memory CD8 T cells provide any protective benefit during a lethal infection, naïve mice or mice containing LCMV-specific memory CD8 T cells were infected with a high dose LM infection (~ 100 LD 50 ). As with low dose infection, a high percentage of bystander activated memory CD8 T cells were present 1 day following high dose LM infection, and responses persisted but waned after day 1 (Figure 33A). However, while Ag-directed memory CD8 T cell responses protected mice from weight loss and bacterial growth in the spleen, bystander memory CD8 T cell responses did not protect against weight loss or bacterial growth compared to naïve mice (Figure 33B,C). Taken together, these data indicate that while bystander activated memory CD8 T cells are able to provide protection in hosts lacking a fully functional immune system, bystander responses provide no significant protection during non-related systemic infection with LM in immuno-competent hosts. Discussion I have shown that Ag-independent activation of memory CD8 T cells is altered with time after their generation. mrna expression of components of the IL-12 and IL-18 receptors decreased in memory CD8 T cells with time after infection, and consequently, the ability of memory CD8 T cells to sense inflammation and to respond with cytokine production in vitro and their ability to become activated during unrelated infections in vivo decreased with time after their generation. Because infection occurs in the context of inflammatory cytokines produced by innate immune cells, sensing inflammation could be 132

153 an important mechanism for the activation of Ag-specific memory CD8 T cells following infection with pathogens expressing memory CD8 T cell cognate Ag. Indeed, my data show that Ag and inflammation act synergistically in vitro to induce memory CD8 T cell activation. This would suggest that the ability of memory CD8 T cells to respond to infections expressing Ags recognized by memory CD8 T cells would decrease with time after their generation. However, I showed that Ag and inflammation act synergistically to induce memory CD8 T cell activation, and that latem CD8 T cell responses were equally as robust as earlym responses in the presence of Ag and inflammation. Therefore, rather than a global defect in the ability of memory CD8 T cells to respond to infection with time after their generation, memory CD8 T cells appear to become programmed not to respond during unrelated infections, but to robustly respond to infections expressing Ags that they recognize. How might Ag and inflammation synergize to induce memory CD8 T cell activation and IFN-γ production? TCR signaling leads to increased transcription and expression of both chains of the IL-12 receptor [185]. Thus TCR signaling as the result of low levels of Ag could lead to increased expression of the IL-12 receptor and enhanced sensitivity to IL-12. Furthermore, expression of the IL-12 receptor subunits increases following exposure to IFN-γ and TNF-α [185], and production of IFN-γ and TNF-α by T cells following TCR signaling could lead to increased IL-12 receptor expression and enhanced sensitivity to IL-12. In addition, expression of components of the IL-18 receptor increases after exposure to cytokines including IL-2, and production of IL-2 following TCR signaling could lead to increased expression of the IL-18 receptor and enhanced sensitivity to cytokines. STAT4 activation downstream of IL-12 signaling 133

154 enhances the binding activity of transcription factors that regulate IFN-γ transcription and that are induced upon TCR signaling including AP1 and NFκB [186,187]. Therefore, an additional possibility is that cytokine signaling regulates transcription factors induced following TCR signaling. Furthermore, work conducted at the University of Iowa has demonstrated that the signal-transduction capacity of TCR proximal molecules is enhanced by inflammatory cytokines, which reduces the antigen density required to trigger antimicrobial functions [143]. Thus, cytokine signals may lead to activation of memory CD8 T cells following exposure to lower Ag levels. Moreover, Richer et. al. also showed that IL-15 signaling, which was induced upon type I IFN exposure, induced memory CD8 T cell cycle progression, resulting in greater memory CD8 T cell proliferation following Ag exposure [188]. While this study examined the effects of cytokine signaling on memory CD8 T cell proliferation rather than effector cytokine production, it suggests that exposure to inflammatory cytokines may prepare memory CD8 T cells for optimal responses upon Ag encounter. While additional possibilities likely exist, these are some of the possible ways that Ag and inflammation may synergize to induce memory CD8 T cell activation. I have also shown that early activation of memory CD8 T cells in vivo is enhanced by cognate Ag. When analyzing 2 o memory CD8 T cell activation following homologous infection with LM either expressing or not expressing cognate Ag using a model similar to that described in recent reports [39], I found that early activation of memory CD8 T cells responding in the presence of a large polyclonal CD8 T cell response was similar regardless of cognate Ag expression by the invading pathogen. However, I demonstrated that in this model the Ag-specific memory cells for which 134

155 activation was examined did not significantly contribute to clearance of infection, suggesting that activation was primarily driven by inflammation rather than cognate Ag. Conversely, using a heterologous infection model where Ag-specific memory CD8 T cells responded in the absence of other responding memory populations and significantly contributed to pathogen clearance, memory CD8 T cells responded faster in response to pathogens expressing cognate Ag, and responses waned as the infection was cleared. These data are in agreement with my in vitro data that showed a synergistic effect of Ag and inflammation for memory CD8 T cell activation and suggest that early activation of memory CD8 T cells responding to pathogens expressing cognate Ag is likely enhanced in vivo by a synergistic effect of low levels of inflammation and Ag. Furthermore, I showed that bystander activation of memory CD8 T cells is influenced by the infection dose and the level of inflammation elicited following infection. This suggests that the ability of memory CD8 T cells to sense inflammation will impact their ability to become activated following infection with a pathogen not expressing cognate Ag. I have shown that the ability of memory CD8 T cells to sense inflammation is dependent upon time after infection, but what additional parameters, might influence the ability of memory CD8 T cells to sense inflammation? Exhausted CD8 T cells have been shown to downregulate the IL-18 receptor and to become unresponsive to unrelated infections [49]. Thus, the nature of the initial infection may influence the ability of memory CD8 T cells to sense inflammation and to become activated following re-infection. Additionally, research has shown that the phenotype, function, and transcriptome, including expression of genes coding for cytokine receptors, of memory CD8 T cells is altered in a stepwise fashion with each additional Ag 135

156 encounter [107]. Analysis of microarray data from this study shows that expression of components of the IL-12 and IL-18 receptors increases with each subsequent Ag encounter. Thus, activation of memory CD8 T cells in response to Ag and inflammation may be influenced by Ag stimulation history. Non Ag-specific, inflammation driven bystander responses during unrelated infections could provide the host with a protective benefit. Memory CD8 T cells specific for unrelated viruses become activated in humans following HIV [42,43] or Epstein-Barr virus infections [41]. Therefore, determining whether bystander CD8 T cell responses play a protective role during unrelated infections could be of significance to the human population. I have shown that pre-existing virus-specific memory CD8 T cells respond in a bystander manner but provide no significant protection following unrelated LM infection. However, LM results in acute infection characterized by systemic inflammatory cytokine responses [116], and it is possible that cytokines produced by bystander activated CD8 T cells do not significantly add to systemic levels of cytokines produced by innate cells during the early response to LM. In contrast to systemic infections, immune responses to localized infections are initiated by a smaller number of cells that survey peripheral tissues. Ag-driven responses by resident memory CD8 T cells in barrier tissues provide enhanced protection to localized infections [59,60,63,136,189], but it is unclear if tissue resident memory CD8 T cells can be activated following unrelated infections. This is an important question, as bystander responses by memory CD8 T cells in areas of the body where levels of inflammatory cytokines are low may provide the host with a protective benefit following infection with unrelated pathogens. If bystander responses by resident memory CD8 T cells were found to be protective 136

157 following localized infection, it would be interesting to see whether this protection provided changes with time for the memory CD8 T cells. In summary, my data elucidate how Ag-independent functions of memory CD8 T cells are influenced by time (after generation), the factors influencing Ag-driven activation of memory CD8 T cells, and the protective role of bystander memory CD8 T cells. The data indicate that the ability of memory CD8 T cells to be activated independent of Ag-decreases with time after their generation, that memory CD8 T cell reactivation is enhanced by Ag and inflammation, and that bystander memory CD8 T cell responses to unrelated systemic bacterial infections are influenced by infection dose and amount of inflammation elicited, but are not protective in immuno-competent hosts. These data have implications for the development of vaccination strategies designed to generate protective memory CD8 T cells against diverse pathogens. 137

158 Figure 23: Expression of Components of the IL-12 and IL-18 Receptors Decreases with Time after Infection Analysis was performed on populations of earlym (30-45 days p.i.) and latem (8+ months p.i.) P14 cells. (A) A comparison of relative mrna expression of the indicated components of the IL-12 and IL-18 receptors between populations of earlym and latem P14 cells from microarray data. Fold changes represent decreased expression in latem compared to earlym P14 cells. 138

159 139

160 Figure 24: Ability to Undergo Bystander Activation Decreases with Time after Memory Generation (A) Experimental design. Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 5 CFU Vir LM days (earlym) or 8+ months (latem) later. (B) Representative histograms of bystander IFN-γ production by earlym or latem P14 cells (top), endogenous polyclonal earlym or latem cells (CD11a hi -middle), or Ag-inexperienced cells (CD11a lo -bottom) from earlym or latem mice 20 hours post 2 o infection. (C) Percentage of earlym or latem P14 cells (left) and endogenous polyclonal earlym or latem cells (CD11a hi ) (right) producing IFN-γ 20 hours post 2 o infection. * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Data shown are the mean +SEM of one representative experiment out of greater than three independent experiments with three mice per group. 140

161 141

162 Figure 25: Memory CD8 T Cell Ability to Respond to Inflammatory Cytokines Decreases with Time after Infection (A) Representative histograms showing IFN-γ production by earlym (top) or latem (bottom) P14 cells incubated for 5 hours in the presence of 10ng/mL each ril-12 and IL- 18, ril-12 and TNF-α, or ril-18 and IFN-β. (B) Percentages of earlym or latem P14 cells producing IFN-γ after 5 hour incubation in the presence of 10ng/mL each of the indicated recombinant cytokines. ** statistically significant (p<0.01) as determined by Student t-test. Data shown are the mean +SEM of one representative experiment out of greater than three independent experiments with three mice per group. 142

163 143

164 Figure 26: Ag and Inflammation Act Synergistically in vitro to Induce Memory CD8 T Cell Activation (A) Representative dot plot showing IFN-γ production by P14 cells incubated for 5 hours in the presence of the indicated concentrations of GP 33 peptide and/or the indicated concentrations of ril-12 and IL-18. (B) Percentages of P14 cells producing IFN-γ after 5 hour incubation in the presence (+) or absence (-) of GP 33 peptide (0.01nM) and/or ril-12 and IL-18 (0.5ng each) or (C) IL-12 and TNF-α or IL-18 and IFN-β. Data shown are the mean +SEM of one representative experiment out of greater than three independent experiments with three mice per group. 144

165 145

166 Figure 27: Synergistic Effects of Ag and Inflammation Upon Memory CD8 T Cell Activation Become more Pronounced with Time after Infection (A) Representative histograms showing IFN-γ production by earlym (top) and latem (bottom) P14 cells incubated for 5 hours in the presence of the indicated concentrations of GP 33 peptide and/or the indicated concentrations of ril-12 and IL-18. (B) Percentages of earlym and latem P14 cells producing IFN-γ after 5 hour incubation in the presence (+) or absence (-) of GP 33 peptide (0.01nM) and/or ril-12 and IL-18, IL-12 and TNF-α, or IL-18 and IFN-β (0.5ng/mL each). NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Data shown are the mean +SEM of one representative experiment out of greater than three independent experiments with three mice per group. 146

167 147

168 Figure 28: Cognate Ag Does Not influence Early Activation of Memory CD8 T Cells that Do Not Significantly Contribute to Clearance of Infection (A) Experimental design. Following adoptive transfer of naïve P14 cells, mice were infected with 1x10 4 CFU Vir LM-GP 33. Mice were given a 2 o infection with 1x10 6 CFU Vir LM (LM Ag) or Vir LM-GP 33 (LM +Ag) >30 days later. (B) Percentage of P14 cells producing IFN-γ or granzymeb, or expressing CD25 or CD69 8 hours post 2 o infection. (C) Percentage of memory CD8 T cells out of total CD8 T cells (left) and P14, GP 33 tetramer, or LM-specific memory CD8 T cells out of total memory CD8 T cells (right) prior to 2 o infection. (D) Bacterial CFUs in the spleen 8 hours post 1 o or 2 o infection. NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test or ANOVA with a Bonferroni post-test. Data shown are the mean +SEM of one representative experiment out of two independent experiments with three mice per group. 148

169 149

170 Figure 29: Ag and Continued Infection Promote Early Activation and Prolonged Responses by Memory CD8 T Cells that Contribute to Clearance of Infection (A) Experimental design. Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 5 CFU Vir LM (LM Ag) or Vir LM-GP 33 (LM +Ag) >30 days later. (B) Bacterial CFUs in the spleen at the indicated times post 2 o infection. (C) Representative histograms of IFN-γ production by P14 cells infected with Vir LM (LM (-Ag)- clear histograms) and Vir LM- GP 33 (LM (+Ag)- grey histograms) at the indicated times post 2 o infection. (D) Percentage of P14 cells producing IFN-γ at the indicated times post 2 o infection. (E) Mice either received 2 o infection with 1x10 5 CFU Vir LM or did not receive secondary infection. Representative histograms of IFN-γ, GrB, CD25, and CD69 expression on splenocytes. NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Data shown are the mean +SEM of one representative experiment out of three independent experiments with three mice per group. 150

171 151

172 Figure 30: Ag Influences Early Activation of Memory CD8 T Cells that Contribute to Clearance of Infection Regardless of LM Virulence (A) Experimental design. Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 5x10 6 CFU Att LM (LM Ag) or Att LM-GP 33 (LM +Ag) >30 days later. (B) Bacterial CFUs in the spleen at the indicated times post 2 o infection. (C) Representative histograms of IFN-γ production by P14 cells infected with Att LM (LM (-Ag)- clear histograms) and Att LM- GP 33 (LM (+Ag)- grey histograms) at the indicated times post 2 o infection. (D) Percentage of P14 cells producing IFN-γ at the indicated times post 2 o infection. NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test. Data shown are the mean +SEM of one representative experiment out of two independent experiments with three mice per group. 152

173 153

174 Figure 31: Magnitude of Bystander Memory CD8 T Cell Responses Correlates with the Level of Infection and Inflammation Elicited upon Infection (A-D) Mice containing memory P14 cells generated following LCMV infection (2x10 5 PFU) were given a 2 o infection with Att LM (5x10 6 CFU) or Vir LM (1x10 4, 1x10 5 or 1x10 6 CFU) not expressing GP 33 (LM Ag) >30 days later. (A) Bacterial CFUs in the spleens and livers following the indicated 2 o infections at the indicated times post infection. (B) Representative histograms of IFN-γ production by P14 cells at the indicated times post 2 o infection. (C) Fold increase in the level of serum IL-12 following the indicated 2 o infections at the indicated times post infection relative to mice not receiving 2 o infection (dashed line). (D) CFUs of bacteria recovered from the spleens of mice plotted vs. the percentage of IFN-γ+ P14 cells 24 hours after infection (top), percentage of IFN-γ+ P14 cells plotted vs. the concentration of serum IL hours after infection (middle), and CFUs of bacteria recovered from the spleens of mice plotted vs. the concentration of serum IL-12 8 and 24 hours after infection (bottom). NS not statistically significant; * statistically significant (p<0.05); ** statistically significant (p<0.01) as determined by Student t-test or ANOVA with a Bonferroni post-test. R squared values were significant based on linear regression analysis. Data shown are the mean +SEM of one representative experiment out of two independent experiments with three to four mice per group. 154

175 155

176 Figure 32: Bystander Responses by Memory CD8 T Cells Provide Protection Against LM in IFN-γ Deficient, but not IFN-γ Sufficient Hosts (A-B) Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 4 CFU Vir LM (LM Ag) >30 days later. (A) Bacterial CFUs in the spleen of LCMV immune mice at the indicated day post 2 o infection. (B) Percentage of P14 cells producing IFN-γ or GrB at the indicated day post 2 o infection. (C) B6 mice received adoptive transfer of naïve P14 cells, and mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 6 CFU Vir LM (LM Ag) >30 days later. A group of mice were treated with ampicillin 24 hours post 2 o infection. Left, Bacterial CFUs in the spleen at the indicated times post 2 o infection, and right, percentage of P14 cells producing IFN-γ 48 hours post 2 o infection. (D) 4x10 5 memory P14 cells were transferred (+) or not (-) into IFN-γ deficient or normal B6 (wt) mice, and mice were infected with 1x10 4 CFU Vir LM. Bacterial CFUs in the liver 2 days following infection. (E) Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 4 CFU Vir LM (LM Ag) or Vir LM-GP 33 (LM +Ag) >30 days later. Naïve mice received no adoptive transfer of P14 cells or LCMV infection. Morbidity (left) and bacterial CFUs in the spleen (right) at the indicated day post LM infection (1x10 4 CFU Vir LM or Vir LM- GP 33 ). NS not statistically significant; ** statistically significant (p<0.01) as determined by Student t-test Data shown are the mean +SEM of one representative experiment out of two independent experiments with three to five mice per group. 156

177 157

178 Figure 33: Bystander Memory CD8 T Cell Responses Fail to Provide Protection Against High-dose LM Infection (A) Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 6 CFU Vir LM (LM Ag) >30 days later. Percentage of P14 cells producing IFN-γ (black squares) or GrB (open squares) at the indicated day post 2 o infection with Vir LM (1x10 6 CFU). (B-C) Following adoptive transfer of naïve P14 cells, mice were infected with LCMV (2x10 5 PFU). Mice were given a 2 o infection with 1x10 6 CFU Vir LM (LM Ag) or Vir LM- GP 33 (LM +Ag) >30 days later. Naïve mice received no adoptive transfer of P14 cells or LCMV infection. Morbidity (B) and bacterial CFUs in the spleen (C) at the indicated day post LM infection. NS not statistically significant; ** statistically significant (p<0.01) as determined by Student t-test or ANOVA with a Bonferroni post-test. Data shown are the mean +SEM of one representative experiment out of two independent experiments with three mice per group. 158

179 159

180 CHAPTER IV: DISCUSSION AND FUTURE PERSPECTIVES Changes in Memory CD8 T Cells with Time after Infection CD8 T cells play a critical role in the clearance of invading viral, bacterial, and protozoan pathogens. Following the primary effector response to infection, a pool of long-lived Ag-specific cells remains as memory CD8 T cells that persist at higher numbers than the naïve repertoire, are localized throughout the body, and are able to more rapidly execute effector functions than naïve CD8 T cells. Because of these properties, memory CD8 T cells have the potential to provide enhanced protection upon re-infection. These principals provide the rationale for the design of vaccines intended to elicit protective memory CD8 T cells that could reduce public health threats including those caused by influenza, malaria, HIV, and hepatitis virus [104, ]. However, current vaccine platforms designed to elicit protective memory CD8 T cell responses are lacking. Therefore, a better understanding of the properties of memory CD8 T cells generated following infection and/or vaccination, and the characteristics of memory CD8 T cells that lead to protection against diverse pathogens is needed. Research has indicated that the numbers of memory CD8 T cells needed to provide protection against some infections are higher than can be established following a single vaccination, and prime-boost protocols have been utilized to increase numbers of memory CD8 T cells [106,140]. Research, including that conducted in my laboratory has been devoted to understanding how the properties of memory CD8 T cells are affected by additional Ag encounters. From this research it was concluded that the magnitude of CD8 T cell expansion, the kinetics of the CD8 T cell response, and the extent of contraction following the proliferative expansion phase changes for memory CD8 T cells with each 160

181 additional Ag encounter [100,101,107] (Figure 34). However, it remained unknown how proliferative and memory generation potential compared between naïve and primary memory CD8 T cells. I addressed this knowledge gap as a Master s student and determined that on a per cell basis, naïve CD8 T cells have a greater capacity than primary memory CD8 T cells to undergo proliferative expansion [10]. While I found that initially, contraction of the resulting primary effector population was more severe than for the secondary effector population, contraction of the secondary response was prolonged, and a larger memory pool was generated from naïve than from primary memory CD8 T cells. This research firmly established that the size of the effector population and the level of memory CD8 T cells generated following infection and/or vaccination decreases following each Ag encounter starting with the first Ag encounter. However, this work and the work of others showed that the magnitude of secondary effector responses and the level of memory CD8 T cells generated following re-exposure to cognate Ag was dependent upon the age of the memory CD8 T cells relative to the initial antigen encounter [10,102]. Memory CD8 T cells that rested for longer periods of time were able to undergo more robust proliferative expansion and generated an increased number of secondary memory CD8 T cells following re-infection. Because the level of CD8 T cell mediated protection is dependent upon both the numbers and quality (i.e. functional abilities) of memory CD8 T cells present at the time of reinfection, this data suggested that protection provided by the memory CD8 T cell pool may change with time following initial Ag encounter. However, the full extent of changes that occur in memory CD8 T cells with time after infection, and whether these 161

182 time dependent changes in memory CD8 T cell function translate into differences in protection provided remained unknown (Figure 34). I addressed this knowledge gap in chapter II of this thesis by comparing the phenotype and function of memory CD8 T cells generated 1 month (earlym) and 8 months (latem) following infection with a systemic viral pathogen. I demonstrated that properties of circulating primary memory CD8 T cells, including location, phenotype, cytokine production, maintenance, secondary proliferation, secondary memory generation potential, and mitochondrial function change with time after infection. I further demonstrated that phenotypic and functional alterations in the memory population were not due solely to shifts in memory subsets that occur with time, but also occur within defined CD62L hi memory CD8 T cell subsets. I found that while CD62L hi memory cells retain the ability to efficiently produce cytokines with time after infection, the gene expression profiles of CD62L hi memory CD8 T cells change, phenotypic heterogeneity decreases, and mitochondrial function and proliferative capacity in either a lymphopenic environment or in response to antigen re-encounter increase with time. Importantly enhanced proliferative and metabolic capabilities of memory CD8 T cells that occurred with time after infection lead to increased protection against chronic LCMV clone-13 infection for both circulating memory CD8 T cell populations and for CD62L hi memory cells. The data reveal that memory CD8 T cells continue to change with time after infection, which impacts their ability to provide protection from re-infection, and suggest that the outcome of prime-boost vaccination strategies will be impacted by the timing between boosts. 162

183 In the discussion section of Chapter II, I described a number of future directions related to this project. These included: 1) determining the longevity of T rm cells, and if the phenotype, function, and protective abilities of T rm cells change with time after infection, 2) determining how progressive changes in memory CD8 T cells with time after infection occur (i.e. cell intrinsic changes in memory CD8 T cells or selective survival of a subset of memory CD8 T cells), 3) determining if the rate of progressive changes in memory CD8 T cells with time after infection is influenced by the nature of the infecting pathogen and/or inflammation elicited during infection and/or vaccination, 4) determining how prior infection and/or infection with pathogens causing chronic infection influences alterations of memory CD8 T cells with time, and 5) determining if protection provided against infections other than LM and LCMV clone-13 are altered with time after infection. While I will not discuss these further, they are all important future directions arising out of this work that will further our knowledge of the factors influencing the generation of protective memory CD8 T cell responses. However, there are a number of additional future directions arising from the work described in Chapter II that I will now describe in more detail. Genes and Proteins Regulating Memory CD8 T Cell Function Previous studies examining gene expression profiles of naïve, effector, and memory CD8 T cells [69], of memory CD8 T cells that have encountered Ag multiple times [107], and of T rm cells [64] have improved our understanding of proteins integral for the regulation of memory CD8 T cell function, and how expression of these proteins is contextually regulated following infection and/or vaccination. Such microarray studies 163

184 have provided many target genes and proteins for further examination of how they regulate CD8 T cell function, and how they may be manipulated for the generation of memory CD8 T cells with characteristics necessary to provide protection against diverse pathogens. In Chapter II I performed microarray analysis that examined how the gene expression profiles of CD62L hi memory CD8 T cells change with time after infection, and I have also performed microarray analysis of how the gene expression profiles of memory CD8 T cells change with time on the population level. This analysis showed that several thousand genes, and gene networks regulating functional abilities of memory CD8 T cells, are differentially regulated with time after infection. Many target genes that could potentially regulate memory CD8 T cell function such as proliferation, the ability to sense and respond to inflammatory signals, and metabolic processes have been suggested from this data for further study. Transcription factors including T-bet, Eomes, Blimp-1, TCF1, LEF-1, Id2, Id3, Bcl-6, Foxo1, Foxo3, STAT3, and STAT4 have been shown to play an important role in regulating memory CD8 T cell functional abilities [147,153,165,166, ]. My microarray data has indicated that expression of transcription factors known to regulate CD8 T cell function are dynamically regulated with time after infection, and microarray data previously published from our laboratory has suggested that expression of many of the same transcription factors are regulated with additional Ag encounters [107]. One candidate transcription factor that has arisen out of this analysis for further study is c-myc (Myc). Myc is a basic helix-loop-helix zipper transcription factor that can both activate and repress gene expression, and it has been shown to regulate many physiological processes including cell growth, proliferation, loss of differentiation, and apoptosis 164

185 [202,203]. Expression of Myc from our microarray data was shown to increase with time after infection and decrease with additional Ag encounters, suggesting that it may play a role in regulating CD8 T cell functional abilities. The role of Myc in regulating CD8 T cell biology has not been well characterized, although one study reported that it is involved in regulating homeostasis of Ag-experienced CD8 T cells downstream of IL-15 [204], and another study suggested that is involved in metabolic reprogramming following activation [154]. To further examine how Myc might regulate CD8 T cell functions, I am working in collaboration with the laboratory of Dr. Howard Xue. They have generated Tg P14 mice with inducible expression of Myc under control of the estrogen receptor. I will adoptively transfer these cells into naïve mice and generate memory P14 cells following LCMV infection. At a memory time point, tamoxifen will be injected to induce Myc expression, and I will examine how over-expression of Myc affects memory CD8 T cell functions including the ability survive, to proliferate following re-infection, and to produce cytokines following Ag encounter. This will be examined for primary memory CD8 T cells and for memory CD8 T cells that have encountered Ag multiple times. This analysis will further our knowledge of proteins important in regulating CD8 T cell function and may provide an additional target which can be manipulated to improve the efficacy of CD8 T cell directed vaccines. Changes in Phenotype and Function of Multiply Stimulated Memory CD8 T Cells with Time after Infection As I showed in Chapter II, primary memory CD8 T cells gradually re-acquire expression of surface molecules including CD127, CD62L, CD27, and CD122 with time 165

186 after infection. Interestingly, expression of these surface molecules in populations of memory CD8 T cells further decreases with each additional Ag encounter, and some functional abilities such as the ability to proliferate following re-infection decrease with additional Ag encounters, while other functional attributes such as levels of pre-formed cytolytic molecules increase with additional Ag encounters [100,101,107]. However, evidence suggests that expression of CD27 and CD62L increases with time after infection in memory CD8 T cells that have encountered Ag more than once, although the period of time required for re-expression of these molecules appears to be prolonged compared to primary memory CD8 T cells [100,101]. That expression of these markers increases with time indicates that as with primary memory, functions of memory CD8 T cells that have encountered Ag multiple times, such as maintenance, cytokine production, and the ability to proliferate and generate memory, may change with time after infection. On the other hand, slower reacquisition of expression of these markers suggests that changes in properties of memory populations that have encountered Ag multiple times may occur at a slower rate than for primary memory CD8 T cells. Because humans are often infected multiple times with the same pathogen, and because prime-boost strategies that lead to the formation of memory CD8 T cell populations that have encountered Ag multiple times are often-times necessary to generate sufficient numbers of memory CD8 T cells to combat infection, an understanding of how the functions of memory CD8 T cells that have encountered Ag multiple times change with time after infection is necessary. The most challenging aspect to studying this question is the time required to generate latem cells that have encountered Ag multiple times. Between the series of adoptive transfers and the amount of resting time required after infection and before 166

187 analysis, a full year is required before real analysis can begin. However, I have generated 1 o and 3 o latem cells using the experimental model described in Figure 35A, and I have begun to analyze how the phenotype and function of 3 o memory CD8 T cells change with time, and how these alterations compare to the progressive changes that occur with time in 1 o memory CD8 T cells. As has been shown before [101,107], expression of phenotypic markers is influenced by the number of Ag encounters, as 3 o earlym cells display reduced expression of CD127, CD62L, CD27, and CD122 and increased expression of KLRG1 compared to expression for 1 o latem cells (Figure 35B). However, as I showed for 1 o memory cells, the phenotype of 3 o memory CD8 T cells changes with time after infection, with expression of CD127, CD62L, CD27, and CD122 increasing with time and expression of KLRG1 decreasing with time. While the phenotype of 3 o memory CD8 T cells does change with time, these changes appear to occur more slowly than for 1 o memory CD8 T cells, as 1 o latem CD8 T cells of the same age relative to infection compared to 3 o latem cells display increased expression of CD62L and decreased expression of KLRG1. To examine if the function of 3 o memory CD8 T cells change with time after infection, I adoptively co-transferred 3 o earlym and 3 o latem cells, or 1 o and 3 o latem cells in equal numbers and infected recipient mice with LCMV (Figure 35C,D). A greater percentage of progeny generated from 3 o latem compared to 3 o earlym cells was found in the PBL at the peak of the effector response and at a memory time point (Figure 35E), indicating that proliferative and memory generation potential of 3 o memory CD8 T cells increases with time after infection. However, the percentage of progeny generated from 1 o latem cells was greater at the effector time point in PBL compared to 3 o latem cells (Figure 33E), suggesting that increased proliferative abilities 167

188 that occur with time after infection occur at a slower rate for 3 o memory compared to 1 o memory. Future experiments will examine whether additional functional abilities including the ability to respond to homeostatic cytokines and to produce effector cytokines and cytolytic molecules changes with time after infection for 3 o memory CD8 T cells, and how these functional changes that occur with time for multiply stimulated memory CD8 T cells may impact their ability to provide protection against diverse pathogens. Regulation of Antigen-dependent and independent Memory CD8 T Cell Activation Following activation, CD8 T cells produce cytokines to stimulate the immune response and cytolytic molecules that result in the death of infected cells [7]. Therefore, understanding the regulation of CD8 T cell activation may aid in the development of more effective CD8 T cell based vaccines. In Chapter III I examined how Ag-dependent and independent activation of CD8 T cells is regulated, and the consequences of Agindependent bystander responses for immune protection. While it is known that CD8 T cells can be activated in an Ag independent manner [38,46], I showed that the ability of memory CD8 T cells to become activated in response to inflammatory cytokines and during unrelated infections decreases with time after their generation. However, I showed that Ag and inflammation act synergistically to induce memory CD8 T cell activation, and that over time memory CD8 T cells remain fully capable of responding to Ag in the presence of inflammation. That memory CD8 T cells acted synergistically with inflammatory cytokines to induce activation suggested that, contrary to recent evidence, Ag matters for CD8 T cell activation following infection with pathogens expressing 168

189 cognate Ag. I went on to show that when memory CD8 T cells significantly contribute to clearance of infection, CD8 T cell responses are enhanced in the presence of cognate Ag. I also addressed whether or not bystander CD8 T cell responses provide protection during unrelated infection and found that bystander memory CD8 T cell responses provided no protection following unrelated LM infection. The data described in Chapter III further our understanding of how time affects the ability of memory CD8 T cells to become activated in an Ag-independent manner, the role of Ag and inflammation in driving CD8 T cell activation following infection, and the role of bystander CD8 T cell responses in providing protection following unrelated infections. There are several future directions to pursue arising from this work that I will discuss in further detail. Expression of Cytokine Receptors and Regulation of Cytokine Signaling Cytokine signaling has been shown to be responsible for inducing Agindependent CD8 T cell activation [38]. In addition, expression of cytokine receptors has been shown to be regulated in CD8 T cells based on the quality of memory CD8 T cells, as exhausted CD8 T cells down-regulated expression of components of the IL-12 and IL- 18 receptors, and this correlated with a decreased ability to respond to inflammatory cytokines and during unrelated infections [49]. My microarray data indicates that expression of Il12rb2 and Il18rap decrease with time after infection, suggesting a possible mechanism for why memory CD8 T cells show a decreased responsiveness to inflammatory cytokines and a reduced ability to respond to unrelated infection with time after their generation. However differences in expression of IL-12rβ2 and IL-18rap need to be validated at the level of gene expression by RT-PCR, and at the level of protein 169

190 expression by flow cytometry or by western blot. Furthermore, additional experimentation will determine the mechanisms downstream of IL-12 receptor signaling that lead to reduced IFN-γ production by latem CD8 T cells in response to inflammatory cytokines and during non-related infections. IL-12 signaling is known to induce dimerization of STAT-4, which regulates transcription of IFN-γ [185,186]. In addition, STAT-4 has been shown to be indispensible for inducing bystander CD8 T cell responses following un-related viral infections [48]. This suggests that STAT-4 may be differentially regulated downstream of cytokine receptor signaling in earlym and latem CD8 T cells. To examine this, I will incubate earlym and latem CD8 T cells with IL-12 and IL-18 and determine phosphorylation of STAT-4 by flow cytometry for earlym and latem CD8 T cells. These experiments will provide valuable insight into the regulation of effector responses induced by inflammatory cytokine signaling, and may allow me to determine how Ag-independent memory CD8 T cell responses are regulated temporally following infection. Protection Provided by Bystander CD8 T Cell Responses In Chapter III I showed that bystander CD8 T cell responses provide no protective benefit following un-related LM infection. However, infection with LM causes a systemic cytokine response [116], and it is possible that bystander IFN-γ produced does not add in any meaningful way to levels of systemic cytokines induced upon LM infection. However, infections arising at barrier tissues elicit a localized inflammatory response, and it is possible that bystander memory CD8 T cell responses would provide a 170

191 protective benefit during localized infections. While Ag-dependent responses by T rm CD8 T cells have been shown to provide a protective benefit to the host [62], Ag-independent bystander responses by T rm cells have either been not tested or not reported. Therefore, I will first determine if T rm cells are able to undergo bystander activation following unrelated infection. The establishment of skin T rm cells following VacV infection has been used as a model to study T rm cells [59,205], and our laboratory recently established skin T rm cells following VacV infection. Therefore, I will use this model to examine if skin T rm cells generated in response to VacV undergo bystander activation following unrelated HSV infection, clearance of which has been shown to be facilitated by Agspecific T rm cell responses [59,60]. If it is found that T rm cells are able to undergo bystander activation, I will then determine if bystander responses by T rm cells provide protection against non-related HSV infection. This experimental line will further our understanding of how bystander CD8 T cell responses are regulated, and will allow us to test an additional model where bystander CD8 T cell responses may provide the host with a protective benefit. Regulation of Antigen-independent Functions of Memory CD8 T Cells that have Encountered Antigen Multiple Times In Chapter III I showed that the ability of memory CD8 T cells to sense inflammation and to respond in a bystander manner during unrelated infections decreases with time after their generation. This correlated with decreased expression of components of the IL-12 and IL-18 cytokine receptors. Microarray data from our laboratory has shown that expression of IL12rb2 and IL18rap increase with additional Ag encounters 171

192 [107], suggesting that the ability of memory CD8 T cells to sense inflammation and to respond in a bystander manner during unrelated infections may increase with additional Ag encounters. To examine this, I generated 1 o and 3 o memory CD8 T cells of the same age relative to Ag encounter (8+ mo. p.i.), and co-incubated the two populations with cytokines that elicit Ag-independent IFN-γ production by memory CD8 T cells. Figure 36A shows that while 1 o and 3 o memory CD8 T cells are both capable of responding to cognate Ag, a greater percentage of 3 o compared to 1 o memory P14 CD8 T cells produce IFN-γ after incubation with IL-12 and IL-18, IL-12 and TNF-α, or IL-18 and IFN-β. This data indicates that the ability of memory CD8 T cells to sense and respond to inflammatory cytokines increases with additional Ag encounters. As I had shown that the ability of memory CD8 T cells to sense inflammation decreases with time after infection for primary memory CD8 T cells, I also wanted to determine if this function of memory CD8 T cells decreases with time after infection for memory CD8 T cells that have encountered Ag multiple times. To do this, I generated 3 o earlym (1 month p.i.) and 1 o and 3 o latem P14 cells and incubated cells with inflammatory cytokine combinations. A greater percentage of 3 o earlym compared to 3 o latem cells produced IFN-γ in response to all cytokine combinations tested (Figure 36B,C), suggesting that the decreased sensitivity of memory CD8 T cells to inflammatory cytokines with time after infection is a universal trait of memory CD8 T cells regardless of Ag stimulation history. However, a greater percentage of 3 o compared to 1 o latem cells produced IFN-γ in response to IL-12 and IL-18 (Figure 36B,C), suggesting that decreased sensitivity to inflammation that occurs with time after infection occurs at a slower rate in memory CD8 T cells that have encountered Ag multiple times than for primary memory CD8 T cells. Future 172

193 experimentation will determine if the ability of memory CD8 T cells to respond in a bystander manner during unrelated infections increases for memory CD8 T cells that have encountered Ag multiple times, and if a model in which bystander responses by memory CD8 T cells provide protection can be found, whether protection provided by bystander memory CD8 T cell responses is dependent upon Ag stimulation history. Outbred Mice as a Model to Study CD8 T Cell Responses As a final thought, I would like to discuss the models that we use to study CD8 T cell responses. It has been argued by some that mouse models poorly mimic responses seen in humans [206,207]. However, others have argued that animal models in some cases accurately portray responses seen in humans, and have pointed out that checkpoint blockade therapies determined in mice have translated well as a cancer therapy in humans [ ]. While we have learned a great deal about the immune system using animal models, we should always strive to include models that more accurately represent the human condition. The work presented in this thesis and the majority of animal research conducted utilizes one strain of inbred mice. While models utilizing inbred mouse strains have afforded many tools that allow for detailed analysis of the immune system and lead to a high degree of reproducibility between and within laboratories, this analysis may not accurately represent responses seen in the genetically diverse human population. Therefore, we should strive to diversify our experimental systems to include outbred organisms. 173

194 Tg T cells and MHC tetramers allow us to easily track epitope-specific CD8 T cell responses in inbred mice. However, tracking CD8 T cell responses in outbred mice is a challenge because adoptive transfer of T cells is not possible due to rejection, and tracking endogenous Ag-specific CD8 T cells requires knowledge of host MHC restriction, which will be different in individual outbred mice. However, our laboratory recently reported that polyclonal CD8 T cell responses can be tracked in inbred and outbred mice using the surrogate activation markers CD8α and CD11a [179]. Therefore, using this surrogate activation approach will allow us to examine CD8 T cell responses in genetically diverse organisms that may more accurately represent the human population. I have begun to use this system to determine if some of our seminal notions about CD8 T cells learned using inbred mice are accurately represented in outbred models. It has been established that vaccination can provide hosts with enhanced protection upon reinfection, that vaccination dose correlates with the size of the memory CD8 T cell pool generated, that memory CD8 T cell numbers correlate with the level of protection, and that boosting can be used as a platform to increase the level of CD8 T cell memory and protection provided against re-infection [9,69,105,141,182]. To examine if these principals are similar in outbred organisms, I vaccinated inbred B/6 and outbred NIH Swiss mice with either low dose or high dose Att LM, tracked the level of memory CD8 T cells generated using the surrogate activation markers CD8 and CD11a, and examined the level of protection that these vaccination strategies offered (Figure 37A). As has been shown before, the level of memory CD8 T cells generated following vaccination was dependent on the vaccination dose for inbred mice, and responses were similar in individual inbred mice (Figure 37B). However, the level of memory CD8 T cells 174

195 generated was highly variable in individual outbred mice, and vaccination dose did not correlate with the level of memory CD8 T cells generated in all outbred mice (Figure 37B). Both low dose and high dose vaccination provided immunized B/6 mice with protection, as immunized mice lost less weight than vaccinated mice and lower levels of bacteria were recovered in the spleens of mice four days following challenge (Figure 37C). Furthermore, the level of memory CD8 T cells generated following vaccination correlated with the level of protection provided (Figure 37C). On the other hand, while vaccination was successful in inbred mice, as a decreased percentage of vaccinated NIH Swiss mice died after challenge compared to unvaccinated mice, a vaccine platform that was able to protect all inbred mice was unable to protect all outbred mice (Figure 37D). Furthermore, there was no correlation between the level of memory CD8 T cells generated and the protection provided in outbred mice (Figure 37D). This raises the interesting question of why vaccination was successful in some inbred mice and not in others. While this model indicates that the principal of vaccination to increase levels of memory CD8 T cells and protection provided against re-infection hold true in outbred organisms, it suggests that success of vaccination might vary between individuals in the human population. I will use this model in the future to test if other principals of CD8 T cell biology that we have learned using inbred mouse models are applicable to outbred systems and attempt to determine the reasons why vaccination is successful in some outbred mice but not in others. 175

196 Figure 34: Regulation of Memory CD8 T Cells following Multiple Ag Encounters and with Time after Infection- a Model Following Ag encounter, naïve and memory CD8 T cells (1 o, 2 o, 3 o ) undergo proliferative expansion, contraction, and generate memory CD8 T cell populations. The properties of memory CD8 T cells including their ability to undergo proliferative expansion, duration/degree of contraction, memory generation potential, cytotoxicity, Ag-driven IL- 2 production, basal proliferation and long-term survival, lymph node homing, and transcriptome diversification are influenced by the number of Ag encounters. I have shown that the properties of 1 o memory CD8 T cells including their ability to undergo Ag-dependent and -independent proliferation, ability to generate 2 o memory populations, cytokine production in response to Ag and/or inflammatory cytokines, localization, phenotypic heterogeneity, metabolic properties, and ability to provide protection against diverse infections are altered with time after infection (box). 176

197 177