Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phenotypic and functional characterization of T regulatory cells Wang, Adele Y. 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


24-ubc_2012_fall_wang_adele.pdf [ 3.7MB ]
JSON: 24-1.0073309.json
JSON-LD: 24-1.0073309-ld.json
RDF/XML (Pretty): 24-1.0073309-rdf.xml
RDF/JSON: 24-1.0073309-rdf.json
Turtle: 24-1.0073309-turtle.txt
N-Triples: 24-1.0073309-rdf-ntriples.txt
Original Record: 24-1.0073309-source.json
Full Text

Full Text

PHENOTYPIC AND FUNCTIONAL CHARACTERIZATION OF T REGULATORY CELLS  by Adele Y. Wang  B.Sc., The University of British Columbia, 2007  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Experimental Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  August 2012  ©  Adele Y. Wang, 2012  ii ABSTRACT FOXP3 +  T regulatory cells (Tregs) normally function to restrain immune responses, but when their activities go awry diseases such as autoimmunity and cancer can result. Animal models have proven that enhancing or inhibiting the function of Tregs is an effective way to prevent, and in some cases cure, many immune-mediated diseases. Approaches to specifically modulate the activity of Tregs are already being translated to humans, yet we know remarkably little about how Tregs achieve their potent immunosuppressive effects. The aim of this research was to further understand the factors that regulate the molecular phenotype and functionality of Tregs in order to better use them for therapeutic purposes. To achieve this goal, the interaction between Tregs and adenoviral-transduced human monocyte- derived dendritic cells in the context of cancer immunotherapy was explored. I found that these genetically-engineered DCs designed to boost the immune response were still susceptible to Treg suppressive influence. Next, I investigated the biological relevance of chemokine secretion by Tregs and determined that chemokine-mediated active recruitment of their targets of inhibition may be a novel mechanism of action. Finally, I established a human Treg-specific gene signature using Affymetrix microarray technology in order to define better ways to isolate and track these cells. Taken together, these studies have contributed significantly to understanding how Tregs exert their homeostatic control of immunity and revealed potential tactics to manipulate their activity in clinical aspects.  iii PREFACE  Chapter 1 This chapter contained excerpts from a book chapter. I wrote all of the sections following extensive literature review. M.K. Levings supervised the preparation of the manuscript. J. Medin and D. Fowler reviewed the book chapter and provided feedback. This is a re-printed version of the following book chapter with permission from Springer: Wang AY, and MK Levings. (2011). T regulatory cells and cancer immunotherapy. In J. Medin & D. Fowler (Eds.), Experimental and Applied Immunotherapy (pp.207-228). New York: Humana Press.  Chapter 2 Research in this chapter was done in collaboration with J. Medin (Ontario Cancer Institute, University Health Network, Toronto, ON, Canada) and J. Bramson (Department of Pathology and Molecular Medicine Centre for Gene Therapeutics, McMaster University, Hamilton, ON, Canada). I performed all the experiments, analyzed the results and wrote the manuscript. S. Crome optimized the flow cytometry based isolation of Th17 cells and edited the manuscript. Virus production was performed by K. Jenkins and supervised by J. Bramson and J. Medin. J. Bramson and J. Medin also reviewed the manuscript. M.K. Levings supervised the research and the preparation of the manuscript. A version of chapter 2 was published and re-printed with permission from Springer:  iv Wang AY, Crome SQ, Jenkins KM, Medin JA, Bramson JL, and Levings MK. Adenoviral- transduced dendritic cells are susceptible to suppression by T regulatory cells and promote interleukin 17 production. Cancer Immunol Immunother. 2011 Mar;60(3):381-388.  Chapter 3 I performed all the experiments and analyzed the data. Dr. Scott Patterson, a postdoctoral fellow in the lab, helped with the repeats for experiments on chemokine production and migration assays.  Diabetic patient and control samples were provided by Drs. Rusung Tan (Department of Pathology, Child and Family Research Institute, University of British Columbia, Vancouver, BC, Canada) and Constadina Panagiotopoulos (Department of Pediatrics, Child and Family Research Institute, University of British Columbia, Vancouver, BC, Canada).  Dr. M.K. Levings aided the experimental design and supervised the research.  Chapter 4 Research in this chapter was done in collaboration with Dr. Nick Haining (Department of Pediatric Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA). I performed all the experiments and analyzed the microarray data with advice from Dr. Haining. RNA amplification, fragmentation and hybridization on the Affymetrix microarray were performed by The McGill University and Génome Québec Innovation Centre. Dr. M.K. Levings aided the experimental design and supervised the research.  Work performed under approval by the Clinical Research Ethics Board (UBC CREB NUMBER: H03-70062)  Animal work performed under approval by UBC Animal Care and Use Committee (UBC APPLICATION NUMBER: A10-0214, A11-0066)  v TABLE OF CONTENTS  ABSTRACT ................................................................................................................................. ii PREFACE .................................................................................................................................. iii TABLE OF CONTENTS ............................................................................................................... v LIST OF TABLES ..................................................................................................................... viii LIST OF FIGURES ..................................................................................................................... ix LIST OF ABBREVIATIONS ......................................................................................................... xi ACKNOWLEDGEMENTS .......................................................................................................... xiii DEDICATION ........................................................................................................................... xv 1. INTRODUCTION ................................................................................................................. 1 1.1 Phenotypic and functional characteristics of Tregs .................................................. 1 1.1.1 Naturally-occurring Tregs in mice and humans.................................................... 1 1.1.2 Inducible Tregs ..................................................................................................... 4 1.1.3 Subsets of human natural Tregs ............................................................................ 5 CD45RA+FOXP3lo naive Tregs .................................................................... 5 CD45RA-FOXP3hi memory Tregs ................................................................ 6 1.1.4 Key surface markers on human Tregs .................................................................. 7 CTLA-4 ......................................................................................................... 7 CD127 ........................................................................................................... 8 CD39 ............................................................................................................. 8 LAP/ IL-1R/ GARP ...................................................................................... 9 1.2 Tregs in cancer and autoimmunity ............................................................................ 9  vi 1.2.1 Treg-mediated suppression of tumour immunity ................................................ 10 1.2.2 Functional defects of Tregs in T1D .................................................................... 12 1.3 Suppressive mechanisms of Tregs .......................................................................... 13 1.3.1 Inhibitory cytokines ............................................................................................ 13 1.3.2 Cytolytic pathways.............................................................................................. 15 1.3.3 Metabolic dysregulation...................................................................................... 15 1.3.4 Interactions with APCs ....................................................................................... 17 1.4 Synopsis of research questions ............................................................................... 19 2. ADENOVIRAL-TRANSDUCED DENDRITIC CELLS ARE SUSCEPTIBLE TO SUPPRESSION BY T REGULATORY CELLS AND PROMOTE INTERLEUKIN 17 PRODUCTION ................................ 21 2.1 Introduction ............................................................................................................. 21 2.2 Materials and methods ............................................................................................ 23 2.3 RAdV-transduced DCs remain susceptible to suppression by Tregs ..................... 27 2.4 Th17-induced maturation of RAdV-transduced DCs is suppressed by Tregs ........ 29 2.5 RAdV-exposed DCs induce IL-17 production and upregulate IL-23R expression on Tregs.. ................................................................................................................................. 32 2.6 Tregs exposed to RAdV-transduced DCs have enhanced suppressive capacity .... 34 2.7 RAdV exposure stimulates mature DCs to secrete Th17-promoting cytokines ..... 36 2.8 Discussion ............................................................................................................... 37 3. FUNCTIONAL RELEVANCE OF CCL3 AND CCL4 PRODUCTION BY T REGULATORY CELLS ...................................................................................................................................... 40 3.1 Introduction ............................................................................................................. 40 3.2 Materials and methods ............................................................................................ 41  vii 3.3 Mouse CD4 + Foxp3 +  Tregs produce CCL3 and CCL4 ........................................... 44 3.4 Supernatants containing CCL3 and CCL4 from Tregs attract both CD4+ and CD8+ T cells. ................................................................................................................................. 47 3.5 T1D patients have a decreased proportion of CD4 + FOXP3 + CCL3 +  Tregs ............ 50 3.6 Discussion ............................................................................................................... 52 4. GENE-EXPRESSION PROFILE OF HUMAN T REGULATORY CELLS .................................. 57 4.1 Introduction ............................................................................................................. 57 4.2 Materials and methods ............................................................................................ 59 4.3 Isolation and phenotypic characterization of subsets of human Tregs defined by the expression of CD25 and CD45RA ...................................................................................... 63 4.4 Establishment of the molecular signature of human Tregs through microarray analysis ................................................................................................................................ 67 4.5 Quantitative PCR validation of the Treg gene signature ........................................ 72 4.6 Multiplex validation of the Treg gene signature using NanoString nCounter Analysis System .................................................................................................................. 82 4.7 Discussion ............................................................................................................... 89 5. CONCLUSIONS ................................................................................................................. 94 REFERENCES ......................................................................................................................... 104   viii LIST OF TABLES  Table 4.1 Summary of the samples in each subset of T cells for the microarray analysis. .... 71 Table 4.2 Summary of the four Treg datasets obtained from Gene Expression Omnibus and analyzed in GSEA. .................................................................................................................. 71 Table 4.3 Quantitative PCR validation of the Treg signature. ................................................ 75 Table 4.4 Description of genes in the Treg signature. ............................................................ 78 Table 4.5 Fold change of the NanoString nCounter validation. ............................................. 87   ix LIST OF FIGURES  Chapter 2 Figure 2.1 Treg-mediated down-regulation of CD80 and CD86 on mature untransduced and RAdV-transduced DCs. .......................................................................................................... 28 Figure 2.2 Mature DCs isolated from co-cultures with Tregs show decreased stimulatory capacity. .................................................................................................................................. 29 Figure 2.3 Treg-mediated suppression of Th17-induced maturation of untransduced and RAdV-transduced DCs. .......................................................................................................... 31 Figure 2.4 Tregs exposed to RAdV convert into IL-17 producing and IL-23R expressing cells. ........................................................................................................................................ 33 Figure 2.5 Exposure of Tregs to RAdV-transduced DCs enhances their suppressive capacity. ................................................................................................................................................. 35 Figure 2.6 Mature RAdV-exposed DCs produce Th17-polarizing cytokines. ....................... 36 Chapter 3 Figure 3.1 Mouse CD4 + Foxp3 +  Tregs secrete CCL3 and CCL4. ........................................... 46 Figure 3.2 FOXP3 regulates the promoters of CCL3 and CCL4. ........................................... 47 Figure 3.3 CCR5 is not expressed on ex vivo mouse CD4 +  and CD8 +  T cells. ...................... 49 Figure 3.4 The presence of CCL3 and CCL4 in supernatants from Tregs induces CD4 + and CD8 +  T cell migration. ............................................................................................................ 49 Figure 3.5 FOXP3 +  T cells after 24 h of stimulation are bona-fide Tregs. ............................. 51 Figure 3.6 T1D patients have a decline in the percentage of CD4 + FOXP3 + CCL3 +  Tregs. .... 52   x Chapter 4 Figure 4.1 Gating strategy for naive and memory Tregs and their respective ex vivo and stimulated FOXP3 expression................................................................................................. 65 Figure 4.2 Surface marker expressions of naïve and memory subsets of Tregs and Tconv cells. ........................................................................................................................................ 66 Figure 4.3 In vitro suppressive activity of naïve and memory Tregs. .................................... 66 Figure 4.4 Profiles of enrichment plot and positions of Treg gene set members on the ranked order list. ................................................................................................................................. 71 Figure 4.5 Gene profile plot of IKZF2 (Helios). ..................................................................... 72 Figure 4.6 Microarray gene expression profiling can discriminate between Treg and Tconv populations. ............................................................................................................................. 76 Figure 4.7 Principal component analysis of the 31 Treg-specific genes. ............................... 79 Figure 4.8 Treg gene signature can differentiate Treg and Tconv populations from both mouse and human. .................................................................................................................. 80 Figure 4.9 Treg gene signature can discriminate between health controls and patients with ulcerative colitis. ..................................................................................................................... 81 Figure 4.10 Reference genes selected for use with the NanoString nCounter Analysis System. ................................................................................................................................................. 85 Figure 4.11 Naïve Treg fold change comparison between the Affymetrix and NanoString platforms. ................................................................................................................................ 86 Figure 4.12 Summary of the NanoString nCounter validation. .............................................. 88  xi LIST OF ABBREVIATIONS    Acute myeloid leukemia APCs Antigen-presenting cells B2M Beta-2-microglobulin CCL CC chemokine ligand Tconv cells Conventional T cells cAMP Cyclic adenosine monophosphate CBA Cytometric bead assay CTLs Cytotoxic T lymphocytes CTLA-4 Cytotoxic T-lymphocyte antigen 4 DCs Dendritic cells Ebi3 Epstein-Barr virus induced gene 3 FDR False discovery rate FOXP3   Forkhead box P3 IPEX Immune dysregulation, polyendocrinopathy, enteropathy, and X-linked syndrome IDO Indoleamine 2,3-dioxygenase i Inducible IBD Inflammatory bowel disease IFN Interferon IL Interleukin LAP Latency-associated peptide LFA-1 Lymphocyte function-associated antigen 1 LAG-3 Lymphocyte-activation gene 3 mTOR Mammalian target of rapamycin m Memory MOI Multiplicity of infection n Naïve NK Natural killer NOD Non-obese diabetic PBMC Peripheral blood mononuclear cells PD-1 Programmed death-1 PD-L1 Programmed death ligand-1 qPCR Quantitative real time polymerase chain reaction RAdV Recombinant adenoviruses Tregs    Regulatory T cells RORα Retinoic acid-related orphan receptor alpha  xii RORγt Retinoic acid-related orphan receptor gamma t TCR T cell receptor Th cell T helper cell TGF Transforming growth factor TSDR Treg-specific  demethylated region TNF Tumour necrosis factor TAA Tumour-associated antigen T1D Type 1 diabetes Tr1 Type 1 T regulatory cells UD Undetectable WT Wildtype         xiii ACKNOWLEDGEMENTS  I am forever grateful to my amazing supervisor, Dr. Megan Levings, for her unconditional support and encouragement. She created a stimulating and vigorous experience for my PhD that involved not only academic endeavours but an exciting opportunity to work in Boston. She is truly an inspirational mentor. I am extremely lucky to be given this wonderful opportunity to work with and learn from Megan. I hope to follow her success in both career and family aspects. I thank Dr. Nick Haining for providing the opportunity and space to work in his lab in Boston; for instructing me with the microarray analysis; and for stimulating discussions. I thank Kathleen Yates and Sabrina Imam for their help and advice during my time in Boston. In addition, I thank Drs. Rusung Tan and Constadina Panagiotopoulos for providing me with samples from patients with type 1 diabetes. Furthermore, I want to thank my Supervisory Committee of Dr. Ken Harder, Dr. Brad Nelson and Dr. Kirk Schultz. It has been a pleasure to learn from this fantastic group of researchers who have brought helpful insights and discussions to my research. I am grateful to Drs. Natasha Crellin and Sarah Crome for the mentorship, friendship and thoughtful discussions. I would also like to acknowledge my fellow lab members for supporting me both intellectually and emotionally; I look forward to work every day because of them. I thank Dr. Raewyn Broady, Rosa Garcia, Jana Gillies, Jessica Huang and Lixin Xu for providing technical help to my research as well as their support and friendship. Moreover, I would like to thank Megan Himmel and Alicia McMurchy for countless motivating chats throughout my PhD career and a wonderful friendship that will last for years to come.  xiv  My research has been supported by grants from the Canadian Institutes of Health Research, Michael Smith Foundation for Health Research, Terry Fox Foundation and Genome BC. I thank these agencies for their tremendous support. In addition, throughout my studies, I have been personally supported by CIHR/UBC Strategic Training Program in Translation Research, CIHR Canada Graduate Award, CIHR/ MSFHR Foreign Study Supplement and CFRI Research Methodology Training Grant. I would further like to acknowledge the blood donors and patients who have made my research possible.  Finally, I offer my enduring gratitude to those who have been with me throughout my PhD journey. Special thanks are owed to my parents and sister, Tina, for their love and endless support. Their belief in me has pushed me along the path towards achievement. In addition, I would like to thank my friends, Grace Li, Leanna Loy, Jennifer Tam and Kayun Yu, for allowing me to put my research before friendship in these past 5 years. I am eternally grateful for their love, understanding and support.   xv DEDICATION    This work is dedicated to my parents, for their unconditional love and support; to my sister, who makes me laugh every day. Thanks for telling me that nothing is impossible if I put in the effort.  1 1. INTRODUCTION  Immune homeostasis is maintained by a balance of regulatory and effector mechanisms that elicit protection against extracellular pathogens while preventing autoimmunity. It is now well established that regulatory T cells (Tregs) are necessary for normal peripheral tolerance, and alteration of their function and/ or numbers have important consequences in a wide range of contexts, such as autoimmunity, transplantation and cancer (1). Since the characterization of Tregs in 1995 (2) and the identification of forkhead box P3 (FOXP3) as a critical transcription regulator for their suppressive activity (3, 4), multiple studies have investigated their mechanism in the maintenance of self-tolerance (5, 6). In mouse models, enhancing Treg function can cure autoimmunity and promote transplant tolerance, while blocking Treg activity can enhance anti-tumour immunity (7).  Therefore therapeutic approaches harnessing the potential of these cells are highly attractive. Further characterizations of Tregs in vitro and in vivo will hopefully translate the success we see in animal models to the clinic.  1.1 Phenotypic and functional characteristics of Tregs 1.1.1 Naturally-occurring Tregs in mice and humans  Naturally-occurring Foxp3 + Tregs were originally defined on the basis of constitutive expression of CD25 (also known as interleukin (IL)-2 receptor alpha chain) (2), and represent 5-10% of the total CD4 + T cell population in humans and mice (3, 8). These Tregs exit the thymus as a distinct and functionally mature T cell subpopulation with a diverse T cell receptor (TCR) repertoire that is poised to suppress responses to self-antigens (9). As  2 discussed in more detail below, since most tumour antigens are self-antigens, naturally- occurring Tregs represent a significant factor by which effective anti-tumour immunity is suppressed. Various suppressive mechanisms have been proposed and can be mediated through these 4 modes of action (10): inhibitory cytokines, cytolytic pathways, metabolic dysregulation and interactions with antigen-presenting cells (APCs). In humans, other proteins that are characteristically expressed by Tregs include LAP, IL-1R, CD39 and CD73, but not CD127, CD49d, IL-2 or IFN- (1, 7, 11). Markers of Tregs will be discussed in more detail below. The development, maintenance, and function of Tregs depends on the expression of the forkhead/winged-helix family transcription regulator, FOXP3 (4). A genetic deficiency of Foxp3 in mice results in hyper-activation of CD4 +  and CD8 +  T cells that causes severe organ-specific autoimmunity. Evidence that adoptive transfer of Tregs can reverse autoimmunity indicates that the pathology is a direct sequela of a deficiency in Tregs (4).  It is now clear that although there are many parallels between the phenotype and function of Tregs in mice and humans, there are several important differences. As in mice, mutation of FOXP3 in humans also causes autoimmunity, and is the underlying genetic defect in patients with immune dysregulation, polyendocrinopathy, enteropathy, and X- linked syndrome (IPEX); IPEX is a rare X-linked immunodeficiency disease that typically causes type 1 diabetes (T1D), inflammatory bowel disease (IBD), allergies, and hyper immunoglobulin E production (12, 13). The cellular changes in IPEX patients, however, are diverse and do not parallel those in Foxp3 -/-  mice. Specifically, the magnitude of the defect in Tregs in IPEX patients depends on the type of FOXP3 mutation: only patients with null mutations appear to have a complete block of the development and function of Tregs. In  3 addition, there is a parallel defect in cytokine production by conventional T (Tconv) cells, suggesting that in humans, FOXP3 may have a function outside the Treg subset (12). It was also shown that humans express two different splice variants of FOXP3, with the smaller FOXP3b form lacking exon 2 (14). It was recently shown that FOXP3b lacks part of the transcriptional repression domain and cannot interact with retinoic acid-related orphan receptor (ROR)-α or ROR-t, two transcription factors associated with T helper 17 (Th17) cell development (15, 16). Thus, because human T cells co-express both isoforms of FOXP3, cooperation with other transcription factors and transcriptional regulation is likely distinct from that in mice.  Another major difference between mouse and human cells is that FOXP3 is transiently expressed in human Tconv cells, but at levels that are insufficient to suppress cytokine production or confer suppressive function (17-19). This finding has significant implications for studies that have simply used analysis of FOXP3 expression (with or without CD25 assessment) to track and enumerate Tregs. In the absence of assays to determine whether the putative FOXP3 +  Tregs have repressed cytokines and/or suppressive capacity, accurate conclusions about changes in Treg numbers cannot be made from such analyses. The fact that transient expression of FOXP3 is not sufficient to confer a Treg phenotype suggests that co-expression of other transcription factors and stable epigenetic changes are required for Treg development. This notion is supported by the finding that only upon stable and long-term expression of FOXP3 can the Treg phenotype be recapitulated in naive or memory human CD4 +  T cells (20). Recent evidence indicating that a subset of FOXP3 +  cells secretes IL-17 and may or may not be suppressive underscores the need for better markers and standardized assays to identify Treg subsets (21-23).  4 1.1.2 Inducible Tregs  In addition to thymus-derived Tregs, there are also subsets of Tregs that develop in the periphery when they encounter antigen in a tolerogenic context. The best characterized types of induced (i) CD4 +  Tregs are those that are phenotypically similar to the naturally- occurring Tregs described above (i.e. FOXP3 + CD25 + ) or the IL-10 secreting type 1 T regulatory (Tr1) cells (24). Through their capacity to produce both IL-10 and TGF-, many tumour cells influence the development of iTregs via direct and indirect mechanisms (7). The main indirect mechanism is mediated by effects on APCs which when exposed to IL-10 and/or TGF- become tolerogenic and contribute to the development of different types of iTregs (25).  In mice, the development of Foxp3 +  iTregs (which were previously known as Th3 cells) can be stimulated by many different mechanisms. For example, exposure to TGF- in the absence or presence of retinoic acid, a vitamin A metabolite produced by CD103 +  dendritic cells (DCs) in the gut-associated lymphoid tissue, leads to the development of Foxp3 +  Tregs that can suppress autoimmunity in Foxp3-deficient mice (26). Activation of naive T cells by V8-integrin-expressing DCs (27) or DCs lacking in suppressor of cytokine signalling 3 (28) can contribute to the generation of peripheral Tregs. Blockade of the phosphatidylinositol 3-kinase/Akt/mammalian target of rapamycin (mTOR) pathway also stimulates iTreg development (29). This finding has had broad clinical implications since rapamycin (Sirolomus), a commonly-used immunosuppressive drug that inhibits the mTOR pathway, has the potential to specifically block Tconv cells and spare/promote naturally- occurring Tregs and iTregs (29-31). FOXP3 +  T cells can also be induced by TGF- in humans, but compared to their mouse counterparts, the resulting T cells have variable  5 suppressive capacity in vitro (19). This discrepancy indicates that more research is required to understand if the process of iTreg development is fundamentally different in mice and humans, or if current methods to detect suppression in vitro are inadequate to reveal their true suppressive potential.  1.1.3 Subsets of human natural Tregs CD45RA+FOXP3lo naive Tregs  Since human Tconv cells also express FOXP3 upon activation (32), several reports have demonstrated that human FOXP3 + T cells are not a homogeneous population and consist of suppressive and non-suppressive subsets (5). Naïve (n) Tregs, based on the surface expression of CD45RA and low level of FOXP3, are prevalent in cord blood (33), are the main subset of Tregs in fetal tissues (34), and possess potent suppressive capacity (35). The frequency of nTregs declines with age, mainly as a consequence of thymic involution (36). In support of their naïve phenotype, nTregs also express CD31, a cell surface marker indicative of recent thymic emigrants (37). These cells are usually in the resting stage as revealed by the absence of expression of Ki67, a nuclear proliferation marker (35). However, upon in vitro TCR stimulation, they readily proliferate and are highly resistant to apoptosis. Once activated, in vivo TCR repertoire analysis showed that nTregs upregulate FOXP3 and convert to CD45RO + CD25 hi FOXP3 hi  Tregs (35). These cells are also involved in autoimmunity, as multiple sclerosis (MS) patients have been shown to have decreased number and impaired function of nTregs (38), suggesting that targeting this population could be beneficial in therapeutic immune interventions. Supporting this theory, depletion of CD31 +  naïve cells from populations of Tregs diminishes the suppressive capacity of healthy  6 but not MS-derived Tregs and neutralizes the difference in inhibitory potencies between the two groups (37). Further investigation into this subset of Tregs is necessary to determine the implication of its role in other autoimmune diseases. CD45RA-FOXP3hi memory Tregs  In contrast to nTregs, memory (m) Tregs, based on the absence of surface expression of CD45RA and high level of FOXP3, are highly susceptible to apoptosis after activation in vitro and in vivo (22, 39). Despite this apoptotic susceptibility, mTregs still possess potent in vitro suppressive activity (22). In support of this observation, Tregs fixed with paraformaldehyde after activation remain suppressive (40), suggesting that once these cells undergo activation-induced phenotypical changes, they are able to mediate suppression regardless of viability. As expected with the memory phenotype, this subset of Tregs is more enriched in adults and elder people (35). Longitudinal studies in adults have demonstrated that nTregs convert to mTregs in humans, with T cell receptor (TCR) repertoire detected initially in nTreg subset found 18 months later in the mTregs, but not in the nTregs (35). Although human mTregs in peripheral blood are susceptible to apoptosis, murine tissue- resident mTregs are relatively stable (41). Hence, it remains to be seen whether human tissue-specific mTregs differ in apoptotic ability compared to the ones in the peripheral blood.  Several reports have also described populations of human mTregs that co-express FOXP3 and IL-17 (23, 42). Depending on the strength of stimulation, IL-17 + FOXP3 +  T cell clones retain their suppressive capacity (42). In contrast to these data, Miyara et al. found that mTregs did not produce cytokines and identified a population of non-suppressive, cytokine-producing CD45RA - CD25 + FOXP3 lo T cells that are usually sorted out alongside  7 with true CD25 hi FOXP3 hi mTregs (22). Therefore, more detailed analysis of these IL-17- producing FOXP3 +  cells is required to confirm that they are indeed bona fide Tregs. Given these conflicting findings, it is clear that more research on mTregs and their involvement in various diseases is needed. 1.1.4 Key surface markers on human Tregs Isolation of pure and viable populations of Tregs is essential for the understanding of their suppressive mechanism as well as for the development of approaches to manipulate their activities in vivo. In order to achieve these goals, it is critical to determine specific surface markers that will accurately identify Tregs. The use of CD25 as a surrogate marker of Tregs is not effective since it is also expressed on activated Tconv cells (43). The identification of FOXP3 has proven to be a better marker of Tregs, but its intracellular expression renders it useless for isolation of live cells. Recent studies have also shown the up-regulation of FOXP3 expression in activated Tconv cells, further revising the previous thought that FOXP3 is Treg-specific (32, 44-46). Unfortunately, all previous attempts to identify Treg-specific marker(s) have not been successful. In the sections below, I will give an overview of molecules commonly associated with Tregs, and their limitations in specifically identifying these cells. CTLA-4 Cytotoxic T-lymphocyte antigen 4 (CTLA-4) is constitutively expressed in FOXP3 +  Tregs (35) and delivers inhibitory signals to activated T cells by competing with CD28 for CD80 and C86 binding on APCs (47). Indeed, Tregs expressing the highest level of CTLA-4 have shown to possess most potent suppressive activity (48). Nonetheless, it should be noted that CTLA-4 is also expressed by Tconv cells upon activation with a similar expression level  8 to Tregs as determined by FACS (49). Therefore, the dual presence of CTLA-4 in both Tregs and Tconv cells precludes its use as a marker of Tregs. CD127  The combination of CD25 along with the lack of CD127 (IL-7Rα) expression on CD4 +  T cells has been used to isolate FOXP3 +  Tregs that are functionally suppressive (50) as well as accurately identify true Tregs in patients with autoimmune diseases (51). However, after activation, Tconv cells also down-regulate CD127 expression (52) thus hindering the discrimination of Tregs from activated Tconv cells. Additionally, the non-suppressive CD45RA - FOXP3 lo fraction has been shown to express low levels of CD127 (35). Therefore, additional markers are needed in parallel with CD25 hi and  CD127 lo/- to distinguish Tregs apart from recently activated Tconv cells. CD39 Tregs express CD39 to inactivate extracellular proinflammatory ATP released by damaged cells, and use this mechanism to prevent Tconv cell activation (53, 54). A recent study showed CD39 expression can differentiate between non-suppressive, cytokine- producing CD45RA - CD25 +  T cells and suppressive CD45RA - CD25 hi  mTregs (54). These CD39 + Tregs more effectively suppress IL-17 production compared to their CD39 - counterparts (55). Moreover, patients with multiple sclerosis or renal allograft rejection have significantly reduced numbers and function of CD39 +  Tregs in the blood (43, 56). Unfortunately, like all other Treg markers, CD39 is also upregulated in activated Tconv cells (45).    9 LAP/ IL-1R/ GARP Expression of latency-associated peptide (LAP) and IL-1R is not observed on resting or expanded FOXP3 +  Tregs and only appear on FOXP3 +  Tregs for a brief period after TCR stimulation (11). Isolation of activated FOXP3 + Tregs with these molecules can successfully distinguish these functional suppressors apart from FOXP3-positive and negative activated Tconv cells (11). This panel of markers seems promising to identify recently activated Treg from patients with inflammatory diseases. In mouse studies, the interaction of IL-1 with IL- 1R is critical for the early programming of Th17 cells (57), thus whether human IL-1R +  Tregs are more susceptible to Th17-inducing conditions remains to be seen. Moreover, glycoprotein A repetitions predominant, GARP (also known as LRRC32), which directly binds to LAP (58), can also selectively discriminates activated human FOXP3 +  Tregs from IL-17-producing activated Tconv cells (59). On the other hand, another group has found that CD4 + LAP +  Tregs lack FOXP3, express TGF-βRII and activation marker CD69, and suppress via TGF-β- and IL-10-depedent mechanism in vitro (60). Overall, more research is required on LAP, IL-1R and GARP to definitely determine their roles in Tregs.  1.2 Tregs in cancer and autoimmunity In order to manipulate Treg biology in clinical studies to restore immune homeostasis, it is critical to comprehend the influence of these cells on the overall immune response. Impairment in Treg numbers and/ or function can contribute to the development of autoimmunity and conversely, excessive Treg activity can dampen the effectiveness of anti- tumour immunity.  Understanding the connection between these two contrasting conditions may uncover the complexity of immune interactions and aid in achieving a balance between  10 tolerance and tumour immunity. In the sections below, I will describe the involvement of Tregs in these two diseases. For the autoimmunity section, I will focus on T1D since the experiments in Chapter 3 are carried out in this context. 1.2.1 Treg-mediated suppression of tumour immunity Upon identification of CD25, and subsequently FOXP3, as proteins that could be used to track and enumerate Tregs in humans, many clinical studies have attempted to correlate changes in Tregs with outcomes in various types of cancer. As in mice, a higher frequency of putative Tregs is seen in patients with a wide variety of cancers [reviewed by Zou (25)]. For instance, in acute myeloid leukemia, accumulation of functionally suppressive Tregs is observed in the peripheral blood of patients and leads to a poor response to chemotherapy (61). This observation is further supported by a mouse model of acute myeloid leukemia which showed improved response to therapy upon depletion of Tregs (62). In addition, increased subsets of Tregs have been demonstrated in the draining lymph nodes in patients with cervical cancer (63). Following therapy, the number of Tregs declined in proportion to the degree of tumour regression (63). Due to the lack of specific markers, it has been difficult to define the relative contribution of naturally-occurring Tregs versus different subsets of iTregs, but it is widely assumed that both types of cells co-exist in the tumour microenvironment. Unfortunately, the results of many of these studies are difficult to interpret since both CD25 and FOXP3 are also expressed on activated Tconv cells. The recent finding that a subset of FOXP3 + cells produces IL-17 (23, 42) further underlines the need for more precise tools to monitor bona fide Tregs in human diseases. One strategy is to couple analysis of FOXP3 expression with cytokine production; for example, a recent study enumerated  11 FOXP3 + CD4 +  Treg cells in conjunction IL-2 and IFN-levels and found a significant increase in the proportion of Tregs in intra-tumoural and peri-tumoural sections of metastatic melanoma tumours, but not in peripheral blood (64). Another approach is to monitor the methylation status of the FOXP3 promoter because cells that only express FOXP3 transiently and at low levels do not have stable de-methylation of defined regions (65). Notably, Huehn and colleagues developed a quantitative real-time PCR-based methylation  assay to enable more precise identification of Treg numbers by measuring the degree of demethylation at the Treg-specific  demethylated region (TSDR) of the FOXP3 promoter. Using this method, they found that Treg numbers were elevated in the peripheral blood of patients with IL-2-treated melanoma and in tissue from patients with lung and colon carcinomas (66). More direct evidence for the role of Tregs in human cancers has come from elegant studies on patients with ovarian cancer. FOXP3 + Treg cells were found to be abundant in the ovarian tumour microenvironment, inhibit TAA-specific CD8 + T-cell cytotoxicity, and predict patient mortality (67). Evidence that the Tregs isolated from peripheral blood, ascites, or solid tumours of these patients were equivalent in their capacity to suppress T-cell activation in vitro (67) suggests that tumour-infiltrating Tregs are not more suppressive than those in peripheral blood and indicates that the decrease in anti-tumour immunity is related to their increased numbers rather than altered function at the tumour site (25). Preferential trafficking of Tregs to tumours may be mediated by CC-chemokine ligand (CCL) 22, which binds to CCR4 on Tregs (67). Conversely, another study of ovarian cancer showed that increased FOXP3 +  Treg numbers within tumour infiltrating lymphocytes was associated with improved survival (68). These conflicting findings once again reinforce the need for the  12 discovery of Treg-specific markers as analysis based on FOXP3 expression alone may include activated Tconv cells leading to an overestimation of Treg numbers. 1.2.2 Functional defects of Tregs in T1D T1D is an autoimmune disease that is caused by immune cell-mediated destruction of the insulin-producing beta cells in the pancreas. In individuals with inherited genetic risk for the disease, unknown environmental triggers initiate the process of autoimmunity and beta cell destruction. As mentioned previously, IPEX patients who lack functional Tregs because of a mutation in FOXP3 are born with T1D, indicating that one of the normal functions of Tregs is to stop islet-antigen directed immunity (13). This function of Tregs is recapitulated in the non-obese diabetic (NOD) mouse model of T1D, where deletion of Tregs accelerates diabetes (69) and therapeutic restoration of Tregs can prevent disease progression (70-72). Similarly, targeting molecules that are essential for the survival of Tregs, such as IL-2 or CD28, also exacerbates diabetes in NOD mice (73, 74). Remarkably, in humans the proportion of Tregs in the blood is apparently normal in T1D patients (75, 76), and these cells are able to suppress T cell proliferation in vitro, similar to those from healthy subjects (77). In parallel, no differences in FOXP3 expression as a marker of Tregs were observed between healthy controls and patients (78). Furthermore, the level of demethylation at the TSDR of FOXP3 was similar between the two groups (79). Interestingly, individuals with new-onset diabetes have a relative increase in the IL-17-producing CD45RA - FOXP3 low non- Treg population compared with controls, suggesting a defective number of functional Tregs in the early stage of the disease (80). Although the number of Tregs in the peripheral blood appears to be normal in patients with T1D, there are differences at sites of inflammation. Insufficient numbers of Tregs in the  13 islets of individuals with T1D is evidenced by a histological study of T1D islets that were isolated from patients immediately post-mortem; FOXP3 +  T cells were scarcely present (81). Impaired FOXP3 expression in human Tregs could be due to the deficiency in IL-2 since T1D Tregs do express normal levels of CD25 but have defects in IL-2-mediated signalling (78). Likewise, the loss of Foxp3 expression by Tregs infiltrating the islets in NOD mice is contributed to a decrease in the levels of available IL-2 (82). In summary, it is widely accepted that a change in Treg function likely underlies T1D, however better ways are required to gain access to Tregs at sites of inflammation to accurately determine the involvement of Tregs in T1D.  1.3 Suppressive mechanisms of Tregs  Tregs have a remarkable ability to suppress the proliferation and effector function of many different types of immune cells, including CD4 +  T cells, CD8 +  T cells, natural killer (NK) cells, NK-T cells, DCs, monocytes and B cells (1, 7). Knowledge of how Tregs regulate other immune cells is fundamental to the development of pharmacologic compounds that can be used to modulate Treg activity for therapeutic purposes. Next, I will review the major suppressive mechanisms that are thought to be employed by CD4 + FOXP3 +  Tregs that develop in either the thymus or periphery. 1.3.1 Inhibitory cytokines  IL-10 and TGF- are cytokines that are central mechanisms of suppression. Although FOXP3 +  Tregs were originally thought to operate via a cytokine-independent, cell-contact dependent mechanism, it is now clear that this paradigm needs to be revised based on recent data that, at least in mice, FOXP3 + Tregs express cell surface-bound TGF- and mediate  14 cytokine-dependent suppression and infectious tolerance, the induction of a suppressive state that can be transferred from one cell population to another (11, 83). TGF- affects many aspects of T cell function, most notably suppression of proliferation and cytokine production (84). IL-10  has similar anti-inflammatory and inhibitory effects on most haematopoietic cells (24) and is required for normal immune homeostasis. Similarly to TGF-, IL-10 acts via a positive feedback pathway to promote its own expression, thereby reinforcing immune suppression and expanding the immune-regulatory network (24, 83).  Another inhibitory cytokine, IL-35, a member of the IL-12 family and a heterodimer composed of Ebi3 and p35 subunits, is preferentially expressed by mouse Foxp3 + Tregs and contributes to their suppressive activity (85). Interestingly, Tregs purified from in vitro Treg suppression assays dramatically upregulated Ebi3 and p35 mRNA expression, indicating that Tregs in the process of suppressing Tconv cells could enhance IL-35 secretion (85). Moreover, Tregs that are deficient in IL-35 cannot suppress T cell proliferation and fail to rescue IBD in vivo. Some of the effects of IL-35 were found to rely on IL-10 production, and indeed both IL-10 and IL-35 are required for maximal Treg-mediated suppression in vitro (86). In contrast, human Tregs isolated ex vivo and FOXP3-transduced T cells do not express detectable amounts of IL-35 (20, 87). Nonetheless, a recent finding determined that IL-35 was detectable only after activating human Tregs for more than 3 days; the authors claimed that previous studies were looking at the expression of IL-35 too early (88). Therefore, IL-35 could potentially serve as an attractive therapeutic target to manipulate the activity of Tregs. However, more patient studies are necessary to address the involvement of this cytokine in the context of autoimmunity and cancer and to determine whether it has the ability to suppress other non- T cell populations.  15 1.3.2 Cytolytic pathways  CD8 +  cytotoxic T lymphocytes (CTLs) and NK cells are thought to be the primary mediators of cytolysis via perforin and granzymes. More recently, it has become evident that Tregs can also express perforin and/or granzymes and thereby directly kill their targets of suppression. For example, activated human Tregs express granzyme A and perforin, and can eliminate T cells and APCs through this pathway (89). In addition, the induction of apoptosis in Tconv cells by mouse Tregs is dependent on granzyme B expression (90): Tregs from granzyme-B-deficient mice have decreased suppressive capacity in vitro. Tregs can also kill B cells through a granzyme B- and perforin-dependent manner (91). In addition to perforin and granzyme, Tregs also express Fas and FasL (92), and FasL-induced apoptosis of target cells is another possible mechanism of suppression. In support of this observation, human Tregs were found to induce Fas-mediated apoptosis of autologous CD8 +  T cells (93), and this process is amplified in patients with cancer. Since Tregs interfere with tumour cell removal by granzyme B and perforin-mediated killing of NK cells and CTLs (94), these data suggest that Treg-mediated cytolysis is relevant to anti-tumour immunity. 1.3.3 Metabolic dysregulation  Tregs constitutively express the high affinity IL-2 receptor, which is a heterotrimeric receptor composed of CD25 (-chain), CD122 (-chain) and CD132 (-chain). The trimeric complex has a 100-fold higher affinity for IL-2 than the dimeric (CD122 and CD132) form (16). Therefore, it has been speculated that Tregs might starve Tconv cells by competitive consumption of IL-2 (95). Although there is no experimental evidence to support this mechanism in humans (8), mouse Tregs have been shown to consume IL-2 and induce cytokine-deprivation apoptosis of Tconv cells in vitro (96). On the other hand, another study  16 suggests that IL-2 deprivation alone is not sufficient to suppress cytokine production from Tconv cells (97). Interestingly, Leveque et al. (98) suggested that IL-2 could contribute to anti-tumour immunity by interfering with Treg suppression; upon culture with IL-2, ovarian cancer-associated Treg cells were converted into pro-inflammatory Th17 cells and lost their suppressive ability. Therefore, more work is required to better understand if IL-2 consumption is truly a significant mechanism of suppression.  Another potential way that Tregs can inhibit the metabolic environment is by expression of CD39 and CD73, enzymes that catalyze the conversion of cyclic adenosine monophosphate (cAMP), a potent inhibitor of proliferation and IL-2 synthesis in T cells (99), to adenosine (56, 100). Tregs can also suppress Tconv cells by cell contact-dependent transfer of cAMP using membrane gap junctions (99). Adenosine is an immunosuppressive molecule that has long been known to inhibit Tconv cell proliferation via activation of the adenosine receptor 2A. Notably adenosine can also enhance the development of iTregs by promoting the production of TGF- (101). The accumulation of extracellular adenosine has been demonstrated in the tumour microenvironment (102), but it remains to be determined what proportion is derived from Tregs (103). Reduced numbers of CD39 +  Tregs have been observed in the peripheral blood of patients with multiple sclerosis (56). Moreover, preliminary experiments in a murine transplantation model showed that CD39 + Tregs can determine allograft outcome by degrading ATP released during tissue injury, which would otherwise trigger inflammation (104). Studies to investigate whether Tregs use these metabolic pathways for their activity in various diseases are necessary for the development of treatment therapies.   17 1.3.4 Interactions with APCs  Increasing evidence indicates that Tregs communicate with APCs, in particular with DCs and monocytes, to modulate their maturation and function (105-107). One molecule that is critical to this process is CLTA-4. Although CTLA-4 has long been known to be associated with the immunosuppressive function of Tregs (108), it has been difficult to establish the mechanistic basis for its effects. Onishi et al. elegantly provided evidence that the role of CTLA-4 in Treg-mediated suppression is mediated via DCs, not via direct effects on T cells (106). Specifically, Tregs were found to aggregate around DCs and cause down- regulation of CD80/86 expression by a mechanism that required both lymphocyte function- associated antigen-1 (LFA-1) and CTLA-4. Indeed, a Treg-specific deficiency in CTLA-4 impairs the suppressive capacity of Tregs in vitro and in vivo, and eliminates Treg-mediated down-regulation of CD80/86 on DCs (109). CTLA4 +  Tregs can also stimulate the production of indoleamine 2,3-dioxygenase (IDO) by DCs. IDO is an enzyme that degrades the essential amino acid tryptophan (110), representing yet another mechanism of Treg-mediated immunosuppression. Notably, mice with CTLA-4 -/-  Tregs have enhanced anti-tumour immunity (109), highlighting the importance of Treg-DC interactions in cancer. Since Tregs found in human tumours express CTLA-4 (67), and IDO +  APCs are found both in tumours and their draining lymph nodes (110), CTLA-4-mediated suppression of APCs is likely a major mechanism that counteracts anti-tumour immunity. Another cell surface antigen that may play a role in Treg suppression of DC function is lymphocyte-activation gene 3 (LAG- 3), which binds MHC class II molecules with very high affinity. Binding of LAG-3 to MHC class II molecules expressed by immature DCs induces an inhibitory signal that suppresses DC maturation and immunostimulatory capacity (111). Since activated human T cells can  18 express MHC class II, the interaction between Tregs and Tconv cells through LAG-3 might also result in suppression. Indeed, LAG-3 + FOXP3 +  Treg subset was found with enhanced frequency in PBMCs of patients with cancer and at tumour sites (112).  Another mechanism by which Tregs can suppress APCs is via induction of B7-H4, a member of the B7 family of T cell co-stimulatory molecules (113, 114). Ovarian-tumour- associated macrophages, but not normal macrophages, express B7-H4 and contribute to the suppression of tumour-associated antigen (TAA)-specific T-cell responses (115). Moreover, treatment of mice with B7-H4-immunoglobulin fusion protein decreased the incidence of diabetes, which was associated with transient increase of FOXP3 +  Tregs (116). In addition, emerging evidence suggests that another member of the B7 family, programmed death ligand-1 (PD-L1), which is broadly  and constitutively expressed by B cells, DCs, macrophages and T cells (117), is important for the maintenance of self-tolerance (118) and has a major role in regulation of tumour immunity (119). Its ligand, programmed death-1 (PD-1) is upregulated on activated T cells and upon engagement reverses the activation pathway (118). Mechanistically, stimulation of PD-1 may directly impact Tregs since ligation by PD-L1 promotes sustained expression of FOXP3 (120). Interestingly, PD-1 blockade in melanoma patients appeared to interfere with the suppressive capacity of Tregs and enhanced the generation  of melanoma antigen-specific CTLs (119). In parallel, PD-L1 inhibition of Treg proliferation by interference with STAT5 phosphorylation was demonstrated in patients chronically infected with hepatitis C virus (121).  In summary, the relationship between B7 family members and Tregs need to be more clearly defined in order to develop therapeutic approaches to modulate Treg function in vivo.   19 1.4 Synopsis of research questions  It is now well established that CD4 + FOXP3 +  Tregs are critical for the maintenance of immunological self-tolerance and immune homeostasis. However the mechanisms by which Tregs interact with other immune cells still require further understanding in order to effectively manipulate these cells for therapeutic benefits. Therefore the goals of my project were to explore the potential effects of Tregs on dendritic cell based cancer vaccines, to establish a novel suppressive mechanism of action by Tregs, to develop a Treg gene signature that can be used as a screening platform for compounds to modulate their activity in vivo. It is hypothesized that the efficiency of DC vaccines is often impeded by the presence of Tregs. Recombinant adenoviruses (RAdV) have been used successfully to engineer tumour antigen expression in DCs for use in adoptive immunotherapy, but the impact of virus transduction on susceptibility of these DCs to suppression by Tregs has not been well established. In Chapter 1 the functional consequences of exposure to adenovirus on the interactions between human monocyte-derived DCs and Tregs were determined. Since the development of Tregs is linked to that of pro-inflammatory Th17 cells, the role of Th17 cells and IL-17-producing Tregs in the context of DC-based immunotherapies was also investigated. The unexpected observation that Tregs have the ability to produce pro-inflammatory chemokines prompted the research in Chapter 2. The aim was to determine the functional relevance CCL3 and CCL4 production by Tregs and investigate whether chemokine secretion is involved in Treg-mediated suppressive mechanism.  20 Animal models have demonstrated the power of manipulating Tregs to treat immune- mediated diseases, but adequate tools to proficiently provide similar Treg-based therapies in humans do not exist. To build the capacity to tackle this problem, the goal of Chapter 3 was to use integrative genomic approaches to define the molecular phenotype of human Tregs and develop a Treg gene signature that will pave the way for future genetic and chemical screens to modulate the Treg differentiation state.   21 2. ADENOVIRAL-TRANSDUCED DENDRITIC CELLS ARE SUSCEPTIBLE TO SUPPRESSION BY T REGULATORY CELLS AND PROMOTE INTERLEUKIN 17 PRODUCTION 2.1 Introduction  Extensive pre-clinical work has established that RAdV vectors are efficient delivery platforms for gene therapies and vaccine applications (122). Because RAdV infects various cell types, including APCs and incorporates and expresses large transgenes at high levels (123), there is much interest in using them as tools to express TAAs in APCs and thus boost tumour immunity in cancer patients. Indeed virally-transduced DCs are superior to DCs transfected by other methods (124), and compared to peptide-pulsed DCs, RAdV-transduced DCs have improved migratory capacity (125). Although RAdV-based genetic vaccinations have proven both clinically safe and feasible in melanoma patients (122), RAdV-transduced DC-based regimens have yet to result in consistent tumour regression (126, 127).  Unfortunately, there has not been consistent tumour regression in patients who have participated in DC-based regimens, including one randomized phase III clinical trial (128, 129). It is hypothesized that a major obstacle preventing the efficacy of DC-based cancer vaccines is the presence of Tregs in the tumour microenvironment (130)  For example, co- culture of DCs with human lung carcinoma cells in order to load them with TAAs results in increased TGF-β production that in turn generates FOXP3+ Treg cells (131). Similarly, although DCs fused with breast carcinomas stimulate the expansion of Tconv cells, the Tregs that expand in parallel ultimately suppress T-cell responses (132). In summary, many tumours evade immunity by promoting the development and expansion of Tregs by secreting immunosuppressive cytokines such as IL-10 and TGF-, and by inhibiting Ag presentation  22 (25). Indeed Treg-mediated suppression of immune responses to tumour antigens represents a significant hurdle to successful cancer immunotherapy (25, 130, 133, 134). In addition to T cells, Tregs also modulate the maturation and function of APCs, including DCs and monocytes (1). For example, Tregs regulate the stimulatory capacity of both human and murine DCs by reducing expression of CD80 and CD86, two co-stimulatory molecules known to be critical for optimal T cell priming (106, 135). Additionally, the binding of LAG- 3 on Tregs to MHC class II molecules expressed on DCs delivers an inhibitory signal that interferes with DC maturation and decreases antigen presentation (135, 136). Moreover, Tregs induce PD-L1 expression on DCs hence reducing the ability of DCs to stimulate Tconv cell responses (137). Whether or not any of these Treg-mediated suppressive mechanisms impact the success of immunotherapy based on RAdV-transduced DCs remains an important outstanding question.  In addition to classical Th1- and Th2-mediated responses, increasing evidence suggests that the activity of pro-inflammatory Th17 cell responses also has a critical role in determining the outcome of anti-tumour immunity. In some cases Th17 cells appear to bolster anti-tumour responses by enhancing cytotoxic T cell activity (138), but on the other hand they may promote angiogenesis and tumour growth (139). In mice, the development of Tregs is linked to that of Th17 cells, with the local cytokine milieu influencing lineage commitment (140). Furthermore, human IL-17 +  Tregs can be isolated ex vivo or converted from IL-17 -  Tregs in vitro (23, 141). Thus knowledge of how genetically-modified DCs influence the cytokine phenotype and/or suppression function of Tregs is crucial to understanding how DC-based immunotherapies could impact the relative balance between suppression and inflammation.  23  Here I investigated how interactions between ex vivo Tregs, Th1 and/or Th17 cells affect the phenotype and function of DCs exposed to RAdV. I found that despite their mature phenotype, RAdV-transduced DCs remained susceptible to Treg-mediated suppression. Surprisingly, RAdV-exposed DCs promoted Treg plasticity, stimulated them to produce IL- 17 and increased their suppressive function. These data support the hypothesis that the efficacy of RAdV-transduced DCs in cancer immunotherapy is limited by pre-existing Tregs and indicate that strategies to block stimulation of Tregs should be incorporated in future RAdV-transduced DC- based vaccine strategies.  2.2 Materials and methods Differentiation, maturation and transduction of DC. Peripheral blood was obtained from healthy volunteers following approval by the University of British Columbia Clinical Research Ethics Board and after obtaining written informed consent. PBMCs were isolated by centrifugation over Ficoll. CD14 +  monocytes were purified from PBMCs by positive selection (StemCell Technologies) and immature DCs (iDCs) were generated by culturing monocytes for 5 days in DC medium [RPMI 1640 containing 10% fetal calf serum (FCS) (Invitrogen), penicillin/ streptomycin (Invitrogen), non-essential amino acids (0.1mM, StemCell Technologies), HEPES (10mM, StemCell Technologies), sodium pyruvate (1mM, Invitrogen) and 2 mercaptoethanol (50M, Bio-Rad)]. DC medium along with rhIL-4 (50ng/mL, kind gift from Ulf Korthaeuer, Novartis) and rhGM-CSF (50ng/mL, StemCell Technologies) was replenished every 2 days. Immature DCs were CD11c + HLA- DR int CD80 lo CD86 lo CD83 -  (data not shown). To mature DCs, an inflammatory cytokine cocktail (IL-1, 10ng/mL, Sigma-Aldrich; TNF- 10ng/mL, eBioscience; IL-6, 5ng/mL,  24 eBioscience; and prostaglandin-2 [PGE-2], 1g/mL, Sigma-Aldrich) was added to the iDCs for 24 hours at day 5. Mature DCs (mDCs) were CD11c + HLA-DR hi CD80 hi CD86 hi CD83 +  (data not shown).  For DC transduction, cells were infected with RAdV, at a multiplicity of infection (MOI) of 60, either alone (for iDC) or with the inflammatory cytokine cocktail (for mDC) for 24 hours. The RAdV type 5 vectors contained deletions of E1 and E3 regions (142). A GFP expression cassette was inserted into the E1 region under the control of the murine CMV promoter and the SV40 polyadenylation sequence. The virus was propagated using 293 T cells and purified using CsCl gradient centrifugation as described previously (143). The transduction efficiency was ~50% based on expression of GFP (data not shown). On day 6, DCs were washed extensively before coculture with T cells.  Isolation of T cell subsets and co-culture with DC. CD4 +  T cells were enriched from PBMCs by negative selection (StemCell Technologies, Vancouver, Canada). Tregs were enriched from CD4 +  T cells by positive selection for CD25 expression (Miltenyi Biotec, Auburn, CA) over two columns to ensure 81-90% purity based on expression of CD25 (BD Pharmingen) and FOXP3 (eBioscience). To isolate Th17 (CD4 + CXCR3 - CCR4 + CCR6 + ) and Th1 (CD4 + CXCR3 - ) cells, CD4 + CD25 -  were purified by incubation with CD25 beads (Miltenyi Biotec) and passed over a LS depletion column. The resultant CD4 + CD25 -  T cells were labelled with antibodies to CD4 (eBioscience), CXCR3 (BD Pharmingen), CCR4 (BD Pharmingen) and CCR6 (eBioscience), and sorted using FACS Aria (BD Biosciences) to a purity of >96% as previously described [24]. The cytokine-polarized phenotype of the Th1 and Th17 cells was confirmed by measuring levels of secreted IFN- and IL-17 (data not  25 shown). For T cell co-cultures, allogeneic DCs were incubated at a ratio of 5:1 (T cells to DCs) with Tregs, Th1, and/or Th17 cells as indicated. When DCs were co-cultured with Th1 or Th17 cells and Tregs, the T cell subsets were at a 1:1 ratio. Cells were cultured for 96 hours in DC medium with recombinant human IL-2 (rh-IL-2) (50U/ mL, Chiron). I initially performed the culture experiment with the inclusion of α-CD3. However, α-CD3 alone was contributing to the upregulation of CD80 and CD86 on DCs, perhaps due to their expression of Fc receptors (144). I did not measure the human leukocyte antigen type of the cells so the amount of T cell activation due to the degree of alloreactivity was varied between experiments. However I consistently observed significant downregulation CD80 and CD86 on mDCs upon co-culture with Tregs in multiple replicates.  To test the stimulatory capacity of DCs exposed to Tregs, DCs were incubated with either CD4 + CD25 - Tconv cells or CD4 + CD25 + Tregs for 72 hours at a 1:1 ratio (DCs to T cells). T cells were depleted using positive selection for CD3 +  cells (StemCell Technologies), and the remaining DCs (10,000 cells/well) were tested for their stimulatory capacity in an MLR with 50,000 allogeneic CD4 + T cells/well. Proliferation of T cells was assessed after 96 hours by [ 3 H]thymidine incorporation (1Ci/well, Amersham Biosciences), added for the final 16 hours of culture.  Flow cytometric analyses. DCs were monitored for cell surface expression of CD11c (eBioscience), CD80 and CD86 (both BD Pharmingen). T cells were stained for CD4 (eBioscience) and IL-23R (R&D Systems). For analysis of intracellular cytokine production, T cells were activated with 10 ng/mL PMA and 500 ng/mL Ca 2+ ionophore (both Sigma- Aldrich) for 6 hours, with brefeldin A (10 g/mL, Sigma-Aldrich) added half-way through  26 activation. Following surface staining, cells were fixed in 2% formaldehyde and permeabilized with 0.5% saponin. Intracellular cytokine staining was performed with antibodies against IL-17 (eBioscience) and IFN-(BD Pharmingen). A minimum of 20,000 live cell events were acquired on a BD FACSCanto and analyzed with FCS Express Pro Software Version 3 (De Novo Software, Thornhill, Canada).  Determination of cytokine concentration. Supernatants from untransduced and RAdV- transduced DCs (500,000 cells/mL/well) were collected after 24 hours and frozen at -80 C until analysis. IL-1IL-6, IL-10, IL-12p70 and tumour necrosis factor-alpha (TNF-) were measured by Human Inflammatory Cytokines Cytometric Bead Arrays (BD Biosciences). IL- 23 was measured by ELISA (eBioscience).  Suppression assays. Tregs (125,000 cells/well) were co-cultured with mDCs or RAdV mDCs (25, 000 cells/well; ratio of 5:1 T cells to DCs) in RPMI complete medium for 7 days. Tregs were then re-isolated from co-cultures by positive selection for CD4 (StemCell Technologies). To test for suppressive capacity, autologous PBMCs, which were frozen down when the Tregs were obtained, were labelled with 2.5 mM 5- (6-) CFSE (Molecular Probes) and stimulated at 100, 000 cells/ well with CD3 (1 g/mL) in the presence or absence of various numbers of Tregs corresponding to the indicated ratios. As a control, various numbers of CD4 + CD25 -  T cells were titrated in, and as expected had no detectable suppressive capacity (data not shown). Suppression was assessed after 72 hours by staining the samples for CD8 expression (BD Pharmingen) and analysis of CFSE dilutions in the CD8 +  T cells using flow cytometry.  27  Statistical analysis. All analyses for statistically significant differences were performed with 1-tailed paired Student’s t test. P values of less than 0.05 were considered significant. All error bars represent mean / standard errors.  2.3 RAdV-transduced DCs remain susceptible to suppression by Tregs  In addition to their well-known ability to suppress T cells, Tregs also interact with DCs and reduce their capacity to stimulate T cells via a variety of mechanisms (145, 146). We first investigated whether transduction with RAdV alters the ability of Tregs to suppress DCs. DCs were differentiated from CD14 + monocytes in the presence of IL-4 and GM-CSF for 5 days, then cultured for an additional 24 hours in the absence (for iDC) or presence of a maturation cocktail (IL-1, IL-6, TNF- and PGE-2) (for mDC), with or without the addition of RAdV-GFP. Expression of CD80 and CD86 on RAdV-transduced DCs was determined on gated CD11c + GFP + cells. Exposure to RAdV alone at an MOI of 60 did not mature the DC (data not shown), consistent with previous reports (147-149). As expected, maturation stimulated high levels expression of CD80 and CD86 compared to iDCs (data not shown), and RAdV transduction did not alter this phenotype (Fig. 2.1A). After maturation, in the absence or presence of RAdV, mDCs were co-cultured with allogeneic Tregs in order to provide the required TCR-dependent activation signal (106) and the phenotype of CD11c +  DCs was determined. After 96 hours of exposure of mDCs to Tregs, there was a significant reduction of CD80 and CD86 expression on both untransduced and transduced subsets (Fig. 2.1B) (106, 135). In comparison, mDCs co-cultured with Tconv cells maintained their state of maturation (Fig. 2.1B). Moreover, mDCs isolated from co-cultures with Tregs had a  28 significantly reduced ability to stimulate the proliferation of CD4 +  T cells compared with mDCs isolated from co-cultures with Tconv cells (Fig. 2.2). Notably, mDCs that were transduced with RAdV also remained fully susceptible to Treg-mediated suppression of CD80 and CD86 expression.   Figure 2.1 Treg-mediated down-regulation of CD80 and CD86 on mature untransduced and RAdV-transduced DCs. After 5 days of differentiation in IL-4 and GM-CSF, DCs were matured for 24 hours by addition of IL-1, IL-6, TNF- and PGE-2 in the absence or presence of RAdV. DCs were then cultured alone or co-cultured with either allogeneic Tregs or Tconv cells at a 1:5 ratio (DCs to T cells) for 96 hours and expression of CD80 and CD86 was determined on gated CD11c +  cells. RAdV-transduced DCs were further gated as GFP +  cells. (A) A representative experiments with percentages of positive cells, set according to fluorescent minus one (FMO) controls (not shown), are displayed in each quadrant. Mean fluorescence intensities (MFI) is shown below the percentages. (B) Average fold change in MFI in 4 independent experiments. * indicated p < 0.05 compared to DCs not exposed to Tregs.  29    Figure 2.2 Mature DCs isolated from co-cultures with Tregs show decreased stimulatory capacity. Immature or mature DCs were co-cultured with Tconv cells (DCTc) or Tregs (DCTr) at 1:1 ratio for 72 hours. DCs were isolated and used to stimulate allogeneic CD4+ T cells at a 1:5 ratio (DC to T cells). Proliferation was assessed by [3H]thymidine incorporation after additional 96 hours. Data represent the average calculated from two independent experiments with different donors. * indicated p < 0.05.  2.4 Th17-induced maturation of RAdV-transduced DCs is suppressed by Tregs  Since RAdV-transduced DCs remained susceptible to suppression by Tregs, we next asked whether this also held true in the presence of polarized Th1 and/or Th17 cells. Ex vivo Th1 cells were sorted as CXCR3 +  cells and Th17 cells were sorted as CXCR3 - CCR4 + CCR6 +  cells; ELISAs for IL-17 and IFN- confirmed the expected polarized cytokine phenotype of these cells (data not shown) (150). Exposure of iDCs (in the absence of cytokine maturation) to Th1 (Fig. 2.3A) or Th17 (Fig. 2.3B) cells resulted in significant up-regulation of CD80 and CD86 expression, consistent with previous reports (151, 152). Addition of Tregs at a 1:1 ratio to co-cultures with Th1 (Fig. 2.3A) or Th17 (Fig. 2.3B) cells reduced T cell-induced maturation of iDCs. When Th1 or Th17 cells were co-cultured with mDCs, which were either untransduced or transduced with RAdV (gated on CD11c + GFP + cells), there was a negligible  30 change in the expression of CD80 and CD86, likely because these DCs were already maximally matured (Figs 2.3A and B). Similar to the effect on iDCs, when Tregs were added to co-cultures with mDCs, they significantly decreased expression of CD80 and CD86 on both untransduced and transduced cells. Therefore, the presence of inflammatory Tconv cells, in this case Th1 or Th17 cells, are unable to protect RAdV-transduced DCs from Treg- mediated suppression.        31  Figure 2.3 Treg-mediated suppression of Th17-induced maturation of untransduced and RAdV-transduced DCs. The indicated type of DC was cultured alone, or co-cultured with ex vivo (A) Th1 (CXCR3 + ) or (B) Th17 (CCR4 + CCR6 + CXCR3 - ) cell at a 1:5 ratio, or co-cultured with Th1 or Th17 cells in the presence of a 1:1 ratio of Tregs for 96 hours. Expression of CD80 and CD86 on DCs,  32 gated as CD11c +  cells, was analyzed by flow cytometry. RAdV-transduced DCs were further gated as GFP +  cells. Left panels depict representative experiments, and right panels depict average fold change of MFI from 3 independent experiments. * indicated p < 0.05.  2.5 RAdV-exposed DCs induce IL-17 production and upregulate IL-23R expression on Tregs  Generation of potent immunity involves a bidirectional feedback between DCs and T cells. Therefore I examined the phenotype of CD4 +  T cell subsets stimulated with RAdV- exposed DCs. After 96 hours of co-culture, intracellular cytokine staining of T cells revealed that a significant percentage of Tregs stimulated with RAdV-exposed mDCs, expressed IL- 17 (Fig. 2.4A) (mean % IL-17 +  for Tregs cultured with RAdV-exposed mDCs was 9.6±2.2% versus 2.3±0.4% for Tregs cultured with untransduced mDCs). Similarly, stimulation of Th1 or Th17 cells with RAdV-exposed mDCs also caused upregulation of IL-17 (Fig. 2.4A). In contrast, the proportion of IFN- produced by all three CD4+ T cell subsets remained consistent upon stimulation with RAdV-exposed DCs (Fig. 2.4A). A characteristic phenotype of Th17 cells is expression of the IL-23 receptor (IL-23R) (153, 154), which confers responsiveness to the Th17-stabilizing properties of IL-23 (155). In addition to IL-17, we also found that RAdV-exposed DCs caused a significant upregulation of IL-23R on Tregs, Th17 and Th1 cells (Fig. 2.3B). Taken together, these data indicate that exposure of DCs to RAdV induces a Th17-polarizing program, and support the notion that the cytokine profile of polarized Th cell subsets can be re-directed by APCs.      33  Figure 2.4 Tregs exposed to RAdV convert into IL-17 producing and IL-23R expressing cells. Tregs, Th17 (CCR4 + CCR6 + CXCR3 - ) or Th1 (CXCR3 + ) cells were cultured with either untransduced or RAdV-exposed DCs at a 5:1 ratio (T cells to DCs). (A) After 96 hours of co- culture, cells were stimulated with PMA and ionomycin for 6 hours and expression of IL-17 and IFN- gated on CD4+ cells was analyzed by intracellular staining. Top panel depicts representative data and the bottom panel depicts the average % of IL-17 +  cells from 3 independent experiments. (B) IL-23R expression on CD4 +  T cells was determined by flow cytometric analysis after 96 hours of co-culture. Top panel depicts representative data and the bottom panel depicts the average % of IL23R +  cells from 3 independent experiments. * indicates p < 0.05.     34 2.6 Tregs exposed to RAdV-transduced DCs have enhanced suppressive capacity  I next investigated whether exposure to RAdV-transduced DCs alters the suppressive capacity of Tregs. Tregs were co-cultured with either mDCs or RAdV-exposed mDCs for 7 days. At the end of the co-culturing period, Tregs were re-isolated from the DC-T cell mixture by positive selection of CD4 + T  cells. Various amounts of Tregs (from mDC co- cultures) or RAdV Tregs (from RAdV mDC co-cultures) were incubated with autologous CFSE-labelled PBMCs for 72 hours. Suppression was assessed by analyzing the amount of CFSE dilution in gated CD8 + T cells. Interestingly, RAdV Tregs displayed a significantly enhanced suppressive capacity compared to Tregs (Fig. 2.5). Hence RAdV-exposed mDCs are not only susceptible to Treg-mediated suppression themselves, they further promote suppression by increasing the potency of Tregs.   35  Figure 2.5 Exposure of Tregs to RAdV-transduced DCs enhances their suppressive capacity. Tregs were cultured with either mDCs or RAdV mDCs for 7 days and then purified as CD4 +  T cells by positive selection. (a) Autologous CFSE-labeled PBMCs were stimulated with α- CD3 and cultured with Tregs (from mDC co-cultures; closed circle) or RAdV Tregs (from RAdV mDC co-cultures; open circle) at a 1:8 ratio (Tregs to PBMCs) for 72 hours. The percentage of CD8 + CFSE +  cells was determined by flow cytometry. (b) Averaged data from 3 independent experiments are expressed as percentage suppression, calculated by the formula: 1 - (proliferation in the presence of Tregs/proliferation of CD8 + cells alone) * 100. * indicates p < 0.05.       36 2.7 RAdV exposure stimulates mature DCs to secrete Th17-promoting cytokines  In order to better understand the mechanistic basis for why RAdV promotes Th17- polarizing DCs, I examined how RAdV infection impacts cytokine production from DCs. Although the nature of Th17-polarizing cytokines in humans is a subject of much debate, substantial evidence suggests that the combination of IL-1 and IL-6 is required for induction of IL-17 (156). Other cytokines that could also be involved in Th17 development include: IL-21, IL-23 and TGF- (154). In order to test the cytokine profile of DCs exposed to RAdV, supernatants were collected from mDCs and RAdV mDCs 24 hours after transduction. Consistent with their ability to induce IL-17 production, exposure to RAdV significantly enhanced production of IL-1 and IL-6, but not IL-10, IL-12, or IL-23 (Fig. 2.6). Similarly, exposure of DCs matured by CD40 stimulation to RAdV also resulted in a specific increase in IL-1 and IL-6 production (data not shown). Hence, RAdV confers DCs with the capacity to polarize T cells towards Th17 cells by modulating their cytokine production profile.   Figure 2.6 Mature RAdV-exposed DCs produce Th17-polarizing cytokines. After 5 days of differentiation, DCs were matured overnight with the maturation cocktail with or without the addition of RAdV. The next day, mDCs and RAdV mDC (500, 000  37 cells/mL) were collected and washed extensively and stimulated for 24 hours with IL-4 (50 ng/mL) and GM-CSF (50 ng/mL). Supernatants were collected and analyzed for IL-1β, IL-6, IL-10, IL-12, IL-23 and TNF-α. Data represent averages from three independent experiments. * indicates p < 0.05.   2.8 Discussion  Successful cancer immunotherapy must simultaneously stimulate effector immunity and break tumour-induced tolerance. A great deal of work has focused on the capacity of RAdV-transduced DCs to stimulate Tconv cell immunity, but little is known about how such cells impact Tregs. Here we show that DCs transduced with RAdV remain susceptible to suppression by Tregs and develop a Th17-polarizing cytokine profile. Although RAdV- exposed DCs stimulate IL-17 production in both regulatory and effector cells, these Tregs remain suppressive and are capable of blocking Th1- and Th17-induced maturation of DCs. Together these data indicate that to effectively stimulate T cell immunity, delivery of antigen by RAdV-transduced DCs must be coupled with strategies to remove or prevent Treg- mediated suppression.  RAdV are commonly used gene-delivery vectors in immunotherapy as they efficiently transduce DCs and induce T cell and antibody responses against the delivered transgene products (123). Previous reports have shown that exposure of DCs to Tregs suppresses the expression of costimulatory molecules (106, 135, 136); here I found that despite their mature phenotype RAdV-transduced DC are also susceptible to these effects. In addition to stimulation with inflammatory cytokines or pathogen-associated molecular patterns, DCs can also be matured by interaction with T cells, at least partly via CD40- dependent signals (151, 157). Notably, my data represent the first demonstration that human Th17 cells stimulate DC maturation and thus likely enhance their ability to prime Tconv  38 cells. I found that Tregs suppress DC maturation stimulated by ex vivo Th1 or Th17 cells, indicating that even in the presence of fully polarized T cells, Tregs manifest their inhibitory effects on DCs. Recent findings in mice have suggested that Tregs suppress DCs by outcompeting naive T cells in the formation of LFA-1-dependent aggregates around DCs in vitro (106). Further investigation will be required to define whether a similar mechanism underlies the ability of Tregs to suppress Th1 and Th17-driven maturation. Overall these data indicate the existence of Tregs in the tumour microenvironment could limit the effectiveness of RAdV-transduced DCs to present tumour antigens even if inflammatory T cells are present. Accumulating evidence suggests that Tregs are plastic and can differentiate into IL- 17-producing cells in the presence of IL-1, IL-2, IL-21, and IL-23 (23, 42). In support of Treg plasticity, I discovered that stimulation of Tregs to RAdV-exposed mDCs resulted in elevated proportions of IL-17 but not IFN- secreting cellsMoreover, an increased proportion of IL-17-producing cells was also observed in both Th1 and Th17 cells. I found that DCs exposed to RAdV produced high amounts of IL-1and IL-6, cytokines involved in the differentiation of human Th17 cells (154, 156). Induction of IL-1and IL-6 by RAdV is consistent with previous findings (158), and provides at least part of the mechanistic basis for why mDCs exposed to RAdV stimulate CD4 + T cells to produce IL-17. Elevated IL-17 correlated with increased expression of the IL-23R in Tregs, Th1 and Th17 cells, a molecule that is known to be regulated by RORC, human orthologue of RORγt (150) and is required for IL-23-dependent stabilization of the Th17 lineage (159).  Circulating IL-17 +  Tregs have been identified in humans and shown to be equally suppressive as IL-17 -  Tregs in vitro (23, 42). In contrast, I found that Tregs isolated from co-  39 cultures with RAdV mDCs displayed enhanced suppressive capacity in a dose dependent fashion compared to Tregs isolated from co-cultures with untransduced mDCsFurther investigation into the molecular basis for how RAdV-transduced DCs potentiate Treg suppression and whether it involves regulation of IL-17 production will be of considerable interest. It will also be of interest to determine whether similar mechanisms may be operational in the context of natural viral infections.  Although more potent Tregs would be predicted to be detrimental to anti-tumour immunity, the possibility that they may have a beneficial effect cannot be ruled out. Indeed the concept that inflammation precedes cancer is gaining increased support (160) and depending on the context IL-17-producing Tregs could act to control inflammation and therefore reduce cancer progression.  In conclusion, I determined that Tregs block the ability of RAdV-transduced DCs to stimulate immune responses by down-regulating the expression of costimulatory molecules, CD80 and CD86. These data indicate that pre-existing Tregs may limit the effectiveness of DC-based vaccines in vivo. Indeed, depletion of Tregs enhanced the efficacy of DC immunotherapy and protected vaccinated mice against glioma (161). Evidence that RAdV- exposed DCs promote IL-17 production from multiple T cell subsets indicates that further investigation into how IL-17 impacts immunity will be crucial to understanding and optimizing the therapeutic efficacy of this approach. Moreover, my data clearly indicate that for DC-based immunotherapies to be maximally effective, a combination therapy where RAdV-transduced DC-based vaccines are administered in parallel with Treg inhibitory agents may be required.   40 3. FUNCTIONAL RELEVANCE OF CCL3 AND CCL4 PRODUCTION BY T REGULATORY CELLS  3.1 Introduction  Immune regulation by Tregs prevents the development of autoimmunity, organ rejection and allergy, and also actively controls host responses to tumours and infections (6). Animal studies have shown that manipulating Treg function is an effective approach to prevent, and in some cases cure, immune-mediated diseases (7). Methods to specifically modulate the activity of Tregs are already being translated to humans for cellular therapy, yet we are still attempting to understand how Tregs achieve their potent immunosuppressive effects and whether their therapeutic function may change in response to different microenvironments in vivo. Accumulating evidence has demonstrated that Tregs maintain immune homeostasis and mediate their suppressive effects by migrating to both lymphoid and non-lymphoid tissues based on their distinct chemokine receptor and integrin expression profiles (162). Chemotaxis studies showed that CCR7-expressing nTregs migrate towards CCL19 in order to traffic to T-cell zones in lymph nodes (163).  In disease settings, CD62L +  nTregs are thought to be more suppressive in allograft rejection than their CD62L -  counterparts (164) while CCR4-expressing Tregs accumulate at tumour sites due to CCL22 secretion by ovarian cancer cells (165). Hence, altering Treg trafficking patterns to different sites by chemokines is a promising approach for therapeutic development (166).  On the other hand, the ability of Tregs themselves to produce chemokines and thus direct immune cell trafficking is not well characterized. A colleague reported previously that  41 human CD4 + FOXP3 +  Tregs produce CXCL8 and attract neutrophils in vitro (167). Aside from CXCL8, human Tregs were also shown to secrete a wide range of CC and CXC family chemokines, and in particular, high levels of CCL3 and CCL4. These data were surprising since these are chemokines typically thought to be involved in promoting, not suppressing, inflammation. Since chemokines typically function to promote inflammation by recruiting innate and adaptive immune cells and previous reports demonstrating that the broad and potent suppressive activities of FOXP3 + Tregs require intimate cell contact/proximity (10), I developed the hypothesis that the production of the pro-inflammatory chemokines may be involved in bringing immune cells closer to Tregs to be suppressed.  In this study, my goal was to confirm these data in humans and mice and begin exploring the biological relevance of chemokine production by Tregs. I also examined how chemokine production by Tregs changes in the context of T1D. The results described below suggest that indeed chemokine- mediated active recruitment of their targets of suppression may be a novel mechanism of action by Tregs.  3.2 Materials and methods Isolation and purification of human T cells. Peripheral blood was obtained from healthy volunteers following approval by the University of British Columbia Clinical Research Ethics Board, and after obtaining written informed consent. CD4 +  T cells were purified by RosetteSep (StemCell Technologies). Cryopreserved PBMCS from subjects with new-onset (<6 mo from diagnosis) T1D and from age-matched controls without T1D attending British Columbia Children’s Hospital were provided by Drs. Tan and Panagiotopoulos (80).   42 Isolation and purification of mouse T cells. C57BL/6 Foxp3-EGFP (wildtype, WT), CCR5 - /-  and CCL3 -/-  mice (The Jackson Laboratory) were maintained in specific pathogen-free conditions in accordance with ethics protocols approved by the University of British Columbia Animal Care Committee. CD4 +  T cells from spleen and lymph nodes were enriched to >90% purity using EasySep CD4 negative selection kit (StemCell Technologies) and sorted into CD4 +  Foxp3 + EGFP  +  (Treg) and CD4 + Foxp3 - EGFP –  (Tconv) to 98% purity on a FACSAria. CD4 + CD25 hi (Treg) and CD4 + CD25 - CD45RB hi  (Tconv) cells from CCR5 -/- and CCL3 -/-  mice were sorted using FACSAria to a purity of >98%. CD4-FITC (RM4-5) and CD25-APC (PC61) and CD45RB-PE (16A) were from BD Biosciences. CD8 + T cells from spleen and lymph nodes were enriched to >90% purity using EasySep CD8 negative selection kit (StemCell Technologies).  Determination of chemokine production. FACS-sorted mouse Tregs and Tconv cells (1 × 10 6  / mL) were stimulated with plate-bound α-CD3 (10 μg/mL; 2C11) and soluble α-CD28 (2.5 μg/mL; 37.51) in the presence of IL-2 (100 U/mL) for 48 h in RPMI complete media [RPMI 1640 containing 10% fetal calf serum (FCS) (Invitrogen), penicillin/streptomycin (Invitrogen), Glutamax (Invitrogen), HEPES  (10 mM,  StemCell  Technologies), and 2 mercaptoethanol (50 μM, Bio-Rad)]. Concentrations of CCL3, CCL4 and CCL5 in supernatants were determined using a Cytometric Bead Array Flex Set according to the manufacturer’s instructions (BD Biosciences).  Intracelluar staining. Cryopreserved PBMCs were thawed and activated overnight with αCD3/αCD28-coated beads at a 1:4 bead:cell ratio (Invitrogen) in XH complete media [X-  43 VIVO 15 (Cambrex Corp.) with 5% pooled AB human serum (Cambrex Corp.), and penicillin/streptomycin (Invitrogen), Glutamax (Invitrogen)]. Mouse CD4 +  T cells from Foxp3-EGFP mice were stimulated with plate-bound α-CD3 (10 μg/mL; 2C11) and soluble α-CD28 (2.5 μg/mL; 37.51) in the presence of IL-2 (100 U/mL)for 72 h in RPMI complete media. Cells were stimulated with 10 ng/mL PMA and 500 ng/mL Ca 2+  ionophore (both Sigma–Aldrich) for 6 h, with brefeldin A (10 μg/mL, Sigma–Aldrich) added 1-2 h through activation. PBMCs were stained with Fixable Viability Dye (eBioscience) along with cell- surface markers CD3 (eBioscience) and CD4 (eBioscience). Mouse cells were stained with CD4 (eBioscience). Following surface staining, cells were fixed in 4% formaldehyde and permeabilized with 0.5% saponin.  Intracellular cytokine staining was performed on PBMCs with antibodies against FOXP3 (BD Pharmingen), CCL3 (R&D), IL-17 (eBioscience) and IFN-γ (BD Pharmingen), and on mouse CD4+ cells with antibodies against Foxp3 (eBioscience), CCL3 (R&D) and IFN-γ (BD Pharmingen). A minimum of 20,000 live cell events were acquired on a BD FACSCanto or LSR II, and analyzed with FCS Express Pro Software Version 3 (De Novo Software, Thornhill, Canada).  CCL3 and CCL4 promoter luciferase assay. The Dialign TF software from Genomatix (168) was used for DNA sequence alignment and location of FOXP3 transcription factor binding sites in CCL3 and CCL4 gene promoters. CCL3 (region -1244 to +72; 1316bp) and CCL4 (region -1420 to +273; 1693 bp) were amplified from human genomic DNA and cloned into pGL3. Jurkat cells were transiently transfected as described (169) with pGL3, pGL3-CCL3 or pGL3-CCL4 and a renilla luciferase reporter vector (pRL-TK), in the presence or absence of FOXP3. After 24 h, cells were stimulated with PMA (10ng/mL) and  44 Ca 2+  ionophore (500ng/mL) for 6 h. Luciferase activity was measure using a luminometer (Molecular Devices) and a Dual Luciferase Reporter Assay System (Promega). All values were normalized to renilla luciferase activity and expressed relative to unstimulated controls.  In vitro migration assays. Supernatants (150μL) from FACS-sorted CD4+ Foxp3+EGFP + Tregs, CD4 + Foxp3 - EGFP –  Tconv cells, CCL3 -/-  CD4 + CD25 hi Tregs and CCL3 -/-  CD4 + CD25 - CD45RB hi  Tconv cells (cultured at 1 × 10 6 /mL for 48 h with plate-bound α-CD3 and soluble α-CD28 in the presence of IL-2 in RPMI complete media) were added to the lower chamber of a 3-μm membrane filter in 96-well plate (Millipore). 400,000 stimulated T cells were added to the upper chamber of the transwell plate. Following incubation for 2 h at 37 °C, the number of cells that had migrated into the lower chamber was determined using a hemocytometer in triplicate samples.  Statistical analysis. All analyses for statistically significant differences were performed with 1-tailed paired Student’s t test. P values of less than 0.05 were considered significant. All error bars represent standard deviation (SD) unless otherwise indicated.  3.3 Mouse CD4+Foxp3+ Tregs produce CCL3 and CCL4  It was previously reported that human Tregs secreted significant amounts of numerous chemokines, including those crucial for the inflammatory response, such as CCL2, CCL3, CCL4, CCL7 and CXCL10 (167). CCL3 and CCL4, ligands of the chemokine receptor CCR5, were of particular interest because Tregs secreted very high levels of these two chemokines. I first sought to confirm these findings with human cells in mice.  45 CD4 + Foxp3 - GFP -  Tconv cells and CD4 + Foxp3 + GFP +  Tregs were stimulated and supernatants were collected at 48 h to determine levels of CCL3 and CCL4. On average, mouse Tregs produced 2.2±0.3 ng/mL of CCL3 and 2.6±0.2 ng/mL of CCL4, levels which were comparable to those made by Tconv cells, which produced 1.7±0.4 ng/mL of CCL3 and 2.0±0.4 ng/mL of CCL4 (Fig. 3.1A). The production of CCL5, the third member of the CCR5 ligand family, was also measured. Similar to human Tregs, mouse Tregs secreted a low level of CCL5 (36.0±45.3 pg/mL, data not shown). To confirm these results, the capacity of mouse Tregs to produce CCL3 was analyzed by intracellular staining. Optimal detection of CCL3 from Tregs required a 72 h pre- stimulation with plate-bound α-CD3 and soluble α-CD28 in the presence of IL-2 as we could not detect CCL3 expression in ex vivo CD4 + T cells by intracellular staining (data not shown). After pre-stimulation, activated CD4 +  T cells were stimulated with PMA/ ionomycin for 6 h and CCL3-producing cells were detected in both the Foxp3 -  and Foxp3 +  populations. IFN-γ-producing cells were only observed in Foxp3- cells (Fig. 3.1B and C). On average, 18.0±2.5 % (average±SEM) of stimulated CD4 + Foxp3 - GFP -  T conv cells and 14.8±1.7 % of stimulated CD4 + Foxp3 + GFP +  T cells were positive for CCL3. To determine if FOXP3 is involved in the regulation of these chemokine genes, I used the Dialign TF software to search for FOXP3 binding sites within mouse and human DNA sequences for CCL3 and CCL4 (168). I identified a putative FOXP3 binding site located in the conserved region between mouse and human of both chemokine sequences. I next performed the reported luciferase assay to test whether FOXP3 can transactivate CCL3 and CCL4 genes. When FOXP3 was transfected into human Jurkat cells, both CCL3 (p=0.04) and CCL4 (p=0.03) promoters displayed increased luciferase activity compared to the control  46 vector (Fig. 3.2). In parallel, Jurkat cells were transfected with CXCL8-promoter reporter as the positive control since it was previously reported that FOXP3 regulates this chemokine (170). As expected, positive luciferase activity was observed when cells were transfected with both FOXP3 and CXCL8 (data not shown). Together, these data demonstrate that mouse Tregs produce significant levels of CCL3 and CCL4 and that FOXP3 has a role in the expression of these genes.   Figure 3.1 Mouse CD4 + Foxp3 +  Tregs secrete CCL3 and CCL4. (A) Mouse CD4 +  T cells were sorted into CD4 + Foxp3 + EGFP +  Tregs and CD4 + Foxp3 - EGFP -  Tconv cells, and stimulated for 48 h with plate-bound α-CD3 (10ug/mL) and soluble α-CD28 (2.5ug/mL) in the presence of IL-2 (100U/mL IL-2). Supernatants were analyzed for CCL3 and CCL4 using Cytometric Bead Array flex sets. Data represent averages + SD from three independent experiments. (B and C) Total mouse CD4 +  T cells were stimulated for 72 h with plate-bound α-CD3 and soluble α-CD28 in the presence of IL-2. Cells were then stimulated with PMA/ ionomycin and the proportion of CCL3-positive cells in Foxp3 + and Foxp3 - populations was analyzed by flow cytometry.    47   Figure 3.2 FOXP3 regulates the promoters of CCL3 and CCL4.  (A and B) Human Jurkat cells were transfected with a pGL3 CCL3- or CCL4-promoter reporter construct along with empty vector of a FOXP3-expressing vector. Luciferase activity was measured and the fold increase in activity in stimulated cells over unstimulated controls was calculated. Data represent averages + SD of three independent experiments.  3.4 Supernatants containing CCL3 and CCL4 from Tregs attract both CD4+ and CD8+ T cells  One possible explanation for why Tregs produce chemokines could be that they need to attract their target cells in order to mediate their suppressive control. I therefore asked whether the chemokines produced by Tregs are biologically active by performing in vitro migration assays. Supernatants from CD4 + Foxp3 + GFP +  Tregs and CD4 + Foxp3 - GFP -  Tconv cells that were activated with plate-bound α-CD3 and soluble α-CD28 in the presence of IL-2 for 48 h were added to the bottom of a transwell and assayed for their ability to recruit the T cells. The responder T cells were also stimulated for 48 h in the same condition since ex vivo T cells do not express CCR5 (171). Indeed, the absence of CCR5 staining was observed for both ex vivo WT and CCR5 -/-  CD4 +  and CD8 +  T cells (Fig. 3.3). Supernatants from both cell populations significantly stimulated the migration of Tconv (Fig. 3.4A) and CD8 +  T cells (Fig. 3.4B) compared to medium alone. These supernatants were determined to have high concentrations of CCL3 and CCL4 as indicated in Fig. 3.1A.  48  I next tested the ability of T cells from CCR5-deficient mice to traffic towards supernatants from Tregs. Both CD4 +  Tconv and CD8 +  T cells with a deficiency in CCR5 expression displayed impaired migration towards the supernatants from both Tregs and Tconv cells compared to their corresponding WT counterparts. Notably T cell recruitment was not completely blocked in the absence of CCR5, possibly due to the presence of other chemokines in the supernatants and/or the ability of CCL3 to also bind to CCR1 (172). Since Tregs also express CCR5 on their surface (173), a preliminary examination showed they can migrate towards supernatants from Tregs and Tconv cells to a similar degree (data not shown). To further demonstrate the direct involvement of CCL3 in the attraction of T cells, Tregs and Tconv cells were FACS-sorted from CCL3 -/-  mice and supernatants from these cells were collected after 48 h of stimulation. As expected, migration of WT CD8 +  T cells was significantly diminished towards supernatants from CCL3 -/-  T cells (Fig. 3.4C). Cytometric bead assay (CBA) confirmed the absence of CCL3 in these supernatants (data not shown). Interestingly, the migration of CCR5 -/-  CD8 +  T cells was reduced towards the supernatants of CCL3 -/- T cells than of WT T cells (Fig. 3.4C). This observation supports the fact that CCR1 on CCR5 -/-  T cells might be contributing to the partial migratory ability of these cells towards the supernatants of WT T cells (Fig. 3.4A and B); the absence of CCL3 in the supernatants of CCL3 -/-  T cells means that CCR1 cannot bind to this chemokine to enable chemotaxis. In summary, these data indicate that CCL3 and CCL4 produced by Tregs are functional and contribute to the recruitment of immune cells targeted by Tregs in vitro.   49  Figure 3.3 CCR5 is not expressed on ex vivo mouse CD4 +  and CD8 +  T cells. Ex vivo and activated wildtype and CCR5 -/- CD4 +  and CD8 +  T cells were analyzed for their surface CCR5 expression by flow cytometry. Activated CD4 +  and CD8 +  T cells were stimulated for 48 h with plate-bound α-CD3 and soluble α-CD28 in the presence of IL-2. Data are representative plots from two independent experiments.   Figure 3.4 The presence of CCL3 and CCL4 in supernatants from Tregs induces CD4 + and CD8 +  T cell migration. (A-C) FACS-sorted CD4 +  Tconv and magnetic bead-sorted CD8 +  T cells from WT or CCR5 - /-  mice were stimulated for 48 h with plate-bound α-CD3 and soluble α-CD28 in the presence of IL-2. In parallel, FACS-sorted Tregs and Tconv cells from WT or CCL3 -/- mice were  50 stimulated in the same condition and supernatants were collected after 48 h. The fold migration compared to that stimulated by medium alone was calculated.  Data in A and B are average + SD from three independent experiments. Data in C are average + SD from two independent experiments.   3.5 T1D patients have a decreased proportion of CD4+FOXP3+CCL3+ Tregs  Functional abnormalities in Tregs have been implicated in the progression of autoimmune diseases in animal models and humans (174). To determine whether chemokine expression in Tregs is associated with changes in suppressive function in the context of disease, I obtained PBMCs from recent-onset T1D patients and measured the percentage of CCL3-positive cells in CD4 + FOXP3 +  Treg population by intracellular staining. To detect CCL3 from Tregs, PBMCs were activated for both healthy controls and T1D patients with α- CD3/ α-CD28-coated beads for 24 h before stimulation with PMA/ ionomycin for an additional 6 h. Similar to mouse Treg intracellular staining, CCL3 expression was not detected on ex vivo cells (data not shown). A caveat when analysing activated human T cells is that FOXP3 can be transiently upregulated after 48 to 72 hours (32). In order to analyze CCL3 expression in pre-activated T cells I therefore first confirmed that 24 h of TCR stimulation in healthy individuals did not result in expression of activation-induced FOXP3, meaning that at this time point all FOXP3 +  cells are Tregs demonstrated by the lack of IFN-γ and IL-17 production from the FOXP3 +  population (Fig. 3.5). Next, I stimulated PBMCs from controls and patients with T1D for 24 h and determined an equivalent proportion of FOXP3 + Tregs in both cases (Fig. 3.6A), confirming previous reports (175). However, there was a significant decrease in the proportion of CD4 + FOXP3 + CCL3 +  Tregs observed in patients compared to healthy controls (p=0.03, Fig. 3.6B). A similar percentage of CD4 + FOXP3 - CCL3 +  Tconv cells was seen between the two groups (Fig. 3.6B). Beside the  51 detection of CCL3 from these samples, I also analyzed expression of IFN-γ since it has recently been reported that Tregs from T1D that are expanded in vitro have high expression of IFN-γ (176). There was no difference in the IFN-γ secretory ability between control and T1D Tregs (Fig. 3.6C). Overall, these data suggest that the production of CCL3 by Tregs could be potentially involved in the regulation of tolerance to maintain immune homeostasis.   Figure 3.5 FOXP3 +  T cells after 24 h of stimulation are bona-fide Tregs. Cryopreserved PBMCs from healthy individuals were thawed and activated with with α- CD3/α-CD28-coated for 24 h. Cells were then stimulated with PMA/ Ionomycin for 6 h. Live population was determined by the Fixable Viability Dye. PBMCs were gated on CD3 + CD4 +  and analyzed for FOXP3, IFN-γ and IL-17. Data are representative plots for 3 donors.     52 Figure 3.6 T1D patients have a decline in the percentage of CD4 + FOXP3 + CCL3 +  Tregs. Cryopreserved PBMCs from healthy controls and T1D patients were thawed and activated with α-CD3/α-CD28-coated for 24 h. Cells were then stimulated with PMA/ Ionomycin for 6 h and analyzed for FOXP3, CCL3 and IFN-γ. (A) The proportion of FOXP3-positive cells was determined in each group. (B) CCL3-positive in FOXP3 +  and FOXP3 - populations were analyzed. (C) IFN-γ-positive cells in CD4+FOXP3+ population were also displayed (Control, n = 10; T1D, n = 10).   3.6 Discussion  Multiple mechanisms have been demonstrated to explain how Tregs regulate different cell types (6). Here I propose a potentially novel mechanism of action of Tregs: chemokine- mediated recruitment of their targets of suppression. In this study, I confirmed that mouse Tregs, like human Tregs, produced CCL3 and CCL4. FOXP3 affected the activity of the promoters of these chemokines leading to their enhanced expression. Aside from CCL3 and CCL4, human Tregs can make other chemokines including CCL2, CCL7, CXCL10 and XCL1 (167, 177). Notably, the reduced expression of XCL1 (lymphotactin) was associated with decreased Treg activity in patients with allergic asthma (177), further supporting that chemokine production by Tregs can contribute to their suppressive function. In addition, in vitro migration assays showed that CCL3 and CCL4 in the Treg supernatants could be responsible for the attraction of CD4 +  and CD8 +  T cells, as evidenced by the decreased migratory ability of CCR5 -/- T cells. The presence of other chemokines in the supernatants could contribute to the partial T cell recruitment despite the absence of CCR5 expression. Although CCR5 is the only receptor for CCL4 (178), CCL3 can also bind to CCR1, which is expressed on both CD4 +  and CD8 +  T cells (172). Therefore, CCR5 pathway along with its ligands, CCL3 and CCL4, might be only one of the several chemokine pathways involved in Treg activity. The notion that CCL3-CCR1 axis could be involved in the migration of CCR5 - /-  T towards supernatants of WT T cells is justified by the additional decrease observed in the  53 chemotactic ability of these cells towards supernatants lacking CCL3. In combination, these data indicate that Tregs have the capacity to secrete CCL3 and CCL4, which in turn contribute to the recruitment of immune cells through CCR5 signalling.  CCL3 and CCL4 are commonly considered as pro-inflammatory chemokines since increased levels have been observed in the draining lymph nodes and infection sites upon infection with pathogens such as Leishmania (179) and Mycobateria (180). Emerging evidence, however, suggests these chemokines as having additional immune modulatory function. Human in vivo CD8 iTregs have been reported to inhibit T cell activation through the secretion of CCL4 (178). The accumulation of these CD8 + CCL4 + Tregs in granulomas of Mycobacterium-infected lymph nodes may indicate their regulatory role in persistent infections to hinder effective clearance of the pathogens. Moreover, another group showed that TGF-β iTregs suppress autoimmune gastritis in a mouse model by secreting CCL3 and CCL4 to enhance the migration and enrichment of iTregs in the inflamed gastric mucosa (181). The increased number of iTregs was effective in the suppression of pro-inflammatory cytokines and chemokines produced by the effector T cell population. In our study, we did not find increased migration of naturally-occurring Tregs towards supernatants of other nTregs perhaps due to a similar CCR5 expression between Tregs and Tconv cells (182). The discrepancy could indicate a difference in regulatory mechanisms between induced and natural Treg populations. CCR5-deficient mice, which have a defect in sensing CCL3 and CCL4, have been found to display no immunological abnormalities in the steady state (183). However, after Listeria infection, these mice displayed enhanced delayed-type sensitivity and humoral immunity to bacteria (183). In addition, in a dextran-induced colitis model, CCR5-deficient  54 mice had increased numbers of infiltrating CD4 +  T cells and Th2 cytokines (184).  A similar observation was also made in Mycobacterium-infected CCR5-deficient mice which had more dendritic cells and activated T lymphocytes in the lung-draining lymph nodes (185). Together, the heightened inflammatory response observed in CCR5-deficient mice after infection may be explained by my finding that CCL3 and CCL4 are important for the function of Tregs.  In humans, genes involved in IL-2R signalling have been implicated in the pathogenesis of T1D (80). This defect contributes to the unstable FOXP3 expression in Tregs of individuals with T1D (79). Since I have demonstrated that FOXP3 can interact with the promoters of CCL3 and CCL4 and chromatin immunoprecipitation array profiling shows CCL3 and CCL4 are FOXP3 target genes (186), diminished stability of FOXP3 expression could therefore lead to declined promoter activity of the chemokine genes. This rationale fits nicely with my observation that the proportion of CCL3 + Tregs from patients with T1D was reduced compared to controls. Therefore, aberrant FOXP3 and CCL3 levels may in turn impact establishment of tolerance. The concept that chemokine production by Tregs is biologically relevant for their activity is supported by the finding that XCL1 from human Tregs is involved in controlling allergic asthma (177). In a recent paper, McClymont et al. reported a significantly increased percentage of FOXP3 +IFNγ+ Tregs in patients with T1D and suggested this population might be involved in the pathogenesis of the disease (176). In my hands, an enrichment of IFN-γ-producing Tregs in T1D PBMCs was not detected. These conflicting results could be due to the fact that this group made their observations only after expanding T1D Tregs for 14 days whereas my studies measured IFN-γ expression after overnight stimulation. Since the loss of FOXP3 leads to acquisition of IFN-γ (187), the fact  55 that T1D Tregs gained the ability to produce this effector cytokine suggests dysregulated FOXP3 expression which could then contribute to a decrease in the CCL3 production by these cells. Hence, it would be interesting to expand control and T1D Tregs as illustrated by McClymont et al. (176) and quantitatively compare the CCL3 level between these two populations. The concept of attracting help through chemokines has been demonstrated elegantly by Castellino et al. where CD4 +  T cells efficiently provide help to CD8 +  T cells during a primary immune response (188). Interaction of antigen-specific CD4 +  T cells and DCs induces the production of CCL3 and CCL4 from these cell types. The chemokines then recruit CCR5-expressing CD8 +  T cells to sites of DC-CD4 +  T cell activation and contribute to the development of CD8 +  T cell memory. Since Tregs do not secrete IL-2 and are dependent on this cytokine produced by their target cells for survival and function (189, 190), I propose a mechanism where Tregs utilize the CCR5 pathway to attract their own help by producing CCL3 and CCL4, and in turn mediate suppressive capacity on the target cells. Indeed, once Tconv cells migrate into the lymph nodes and are activated by APCs, they produce IL-2 through a positive feedback loop which then turns on their full effector function (191). At the same time, Tregs use IL-2 to augment their growth and survival, and negatively inhibit the effector function of other cell types (191). Tregs also have been shown to reduce adhesion molecule and chemokine receptor expression on other immune cells to inhibit trafficking of pathogenic cells to the site of inflammation (192, 193). The ability of Tregs to regulate cell trafficking is further supported by Lund et al. who demonstrated Tregs influence on the migration of variety of cells in response to infection (194), but they did not investigate whether the migration was due to chemokines. Moreover, if chemokines are involved in  56 immune regulation by Tregs, the defect in the suppressive capacity of these cells would not detected by in vitro migration assay where the responder cells are already in proximity; this proposed chemokine-mediated mechanism of action could partially explain the contradicting in vitro suppression assay results of diabetic Tregs (175). In conclusion, my study described the ability of mouse Foxp3 +  Tregs to produce CCL3 and CCL4, and showed these chemokines can attract both CD4 +  and CD8 +  T cells in vitro. In addition, I observed a reduced proportion of FOXP3 + CCL3 +  in patients with T1D compared to controls. These results suggest a novel regulatory role for CCL3 and CCL4 in Treg-mediated suppressive mechanism. My preliminary examination found that WT and CCL3 -/-  Tregs have a similar suppressive function in vitro, suggesting Tregs do not need to rely on chemokines once target cells are in proximity. To understand the physiological relevance of chemokine production by Tregs, it is necessary to determine whether this process is operational in vivo. One way to test this notion is using a model of Treg therapy for islet allograft rejection. In this model, BALB/c islets are transplanted into C57BL/6 mice that were treated with streptozotocin, and tolerance can be induced if the recipient mice also receive a donor specific transfusion, a dose of cyclophosphamide, and infusion of Tregs (195). Infusion of WT or CCL3 -/- Tregs into the C57BL/6 mice in this model would reveal whether chemokine production was necessary for their therapeutic effect. If the ability of chemokine production is critical to the immune regulation by Tregs then CCL3 -/- Tregs would be unable to prevent the rejection due to their failure to recruit and suppress the responder cells. Understanding how the chemokine pathway mediates Treg function could generate potential therapeutic targets in the manipulation of these cells for autoimmunity and cancer.  57 4. GENE-EXPRESSION PROFILE OF HUMAN T REGULATORY CELLS  4.1 Introduction  Mouse models have demonstrated that cellular therapy with CD25 hi Tregs can block unwanted immune responses in autoimmunity and transplantation, and there is much need to translate similar strategies to the clinic (196). In humans, however, the use of CD25 for the isolation of Tregs is limited by the fact that the marker is also transiently expressed by activated Tconv cells (45). Since humans do not live in a sterile environment, a proportion of CD4 +  T cells will always be activated CD25 hi  Tconv cells, which also express several Treg- associated molecules, including CTLA-4, GITR, and OX40 (197). Other potential cell surface markers proposed recently such as neuropillin-1 (198, 199), folate receptor 4 (200, 201), or the lack of CD127 (5, 50), also failed to specifically mark Tregs. Therefore, although CD4 + CD25 hi  cells are suppressive based on in vitro biological assays, they are in fact a heterogeneous mixture of Tregs and Tconv cells (35). Presently, the most reliable Treg marker is FOXP3, but cells cannot be purified based on this protein since it is located intracellularly.  Furthermore, human activated Tconv cells express as much FOXP3 as resting Tregs (32, 44) contributing to the revision of the concept that FOXP3 is a Treg-specific protein. All these confounding factors hinder investigation of the role of Tregs in human diseases. An additional problem with using CD25 as a marker for Tregs is the inability to obtain pure populations of Tregs to be expanded for adoptive immunotherapy. There is a real risk of outgrowth of contaminating activated Tconv cells after the expansion phase. Hence, improvements for isolating and tracking these cells in therapeutic contexts are required.  58  It is becoming clear that classification of cells based on surface and intracellular markers alone cannot sufficiently resolve the considerable heterogeneity of cell states in a single lineage (202). In many cases, genome-wide expression profiling represents a more powerful approach to distinguish discrete cell states within otherwise uniform cell types. For instance, in cancer biology, expression profiling allows subtly different biological states within a set of phenotypically identical tumours to be distinguished (203). The power of this approach is driven by scale: not only can ~20,000 'independent' markers be quantified, but coordinated patterns of gene expression involving multiple genes can also be identified. Complex gene expression signatures can therefore serve as unique molecular phenotypes and surrogates for biological cell states. In this context, gene expression profiles can be utilized as a drug screening tool when molecular targets are unknown and read out as the indicator for a desired phenotype. This concept has been demonstrated in acute myeloid leukemia (AML) where screening for the molecular signature of compounds that force myeloid differentiation of the AML cells has identified potential reagents that would induce clinical remission (204) and led to the validation of candidate targets for drug therapy (205). Therefore monitoring changes towards a desired gene expression profile in disease settings is an effective approach for drug discovery. Several groups have used genome-based studies in mice in an attempt to identify novel Treg-specific genes and proteins and gain insight into their suppressive mechanism (206-208). Although these studies advanced our understanding of Treg biology, they failed to achieve the original goal: identification of specific marker gene(s) that can be used to faithfully isolate and study live Tregs. In humans, similar gene-expression profiling studies are limited and fraught with problems, largely related to the lack of specific lineage markers  59 and thus the inability to isolate a homogeneous population of Tregs (209, 210). Proteome- based studies have faced similar challenges (211, 212). In this study, I hypothesized that it is necessary to move beyond a single candidate protein approach in order to find genes or proteins which specifically identify human Tregs. In addition I reasoned that all previous gene-expression based studies with human Tregs were confounded by 2 problems: 1) populations of CD4 + CD25 hi  Tregs were insufficiently homogeneous to produce meaningful molecular data; 2) and since they were compared to naïve CD4 + CD25 -  T cell, most of the differential gene expression reflected differences in memory CD25 +  versus naïve CD25 -  T cells rather than suppressive versus non-suppressive cells (213). By dissecting human Tregs into naive and memory subsets using CD45RA and CD25 and comparing their gene signatures to naive and memory Tconv cells respectively, I hoped to avoid these pitfalls and generate an accurate gene-expression profile of Tregs.  4.2 Materials and methods Isolation of T cell subsets. Peripheral blood was obtained from healthy volunteers following approval by the University of British Columbia Clinical Research Ethics Board, after obtaining written informed consent. CD4 +  T cells were purified by negative selection by RosetteSep (StemCell Technologies). A pre-enrichment was performed by incubation with CD25 magnetic  beads and passing over an LS column (Miltenyi Biotech). CD25 + and CD25 -  cells were then stained for CD4 (eBioscience), CD25 (Miltenyi) and CD45RA (eBioscience), and sorted using a FACSAria (BD Biosciences) into 4 ex vivo populations: CD4 + CD25 ++ CD45RA +  (nTreg) and CD4 + CD25-CD45RA +  (naïve Tconv cell, nTconv), as well as CD4 + CD25 +++ CD45RA -  (mTreg) and CD4 + CD25 - CD45RA -  (mTconv). Purities of  60 >98% on the basis of CD25 and CD45RA expression was achieved. Stimulated populations were obtained by activating the cells for 40 h with αCD3/αCD28-coated beads at a 1:1 bead:cell ratio (Invitrogen) in XH complete media in the presence of IL-2 (100 U/mL, Chiron).  T cell culture and expansion. The 4 previously-mentioned FACS-sorted populations were stimulated with CD3/CD28-coated T cell expander beads (Invitrogen) at 1:1 bead:cell ratio in the presence of 1000 U/mL IL-2 for 10 days in OpTMizer T cell Expansion Serum Free Medium (Invitrogen).  Flow cytometric analysis. Cell-surface markers CCR7 (BD Pharmingen), CD62L (eBioscience), CD39 (eBioscience) and CD127 (eBioscience) were performed only on ex vivo populations. Separate FOXP3 (clone 239D, BD Biosciences) staining was performed with eBioscience Foxp3 buffer kit according to manufacturer’s instructions after surface staining of CD4, CD25 and CD45RA.  Suppression Assays. To test for suppressive capacity, cryopreserved CD4 +  T cells were thawed and plated at 8000 cells/ well (in 250 μL) in 96-well round bottom plates, and stimulated with CD3/CD28-coated T cell expander beads (Invitrogen) at a 1:8 bead:cell ratio in XH complete media for 6 days. Tregs were added in decreasing amount, starting at a ratio of 1:1. As a control, Tconv cells were added to replace Tregs in the assays. For all 7 suppression assays, I used responder CD4 +  T cells from the same donor. Suppression was  61 assessed by measuring the amount of [ 3 H]-thymidine (1 µCi per well; Amersham) incorporation in the final 16 hours of culture.  Microarray analysis. Ex vivo and 40 h stimulated Treg and Tconv populations were resuspended in TRIzol (Invitrogen). RNA was extracted with the RNAdvance Tissue Isolation kit (Agencourt). Concentrations of total RNA were determined with a Nanodrop spectrophotometer. RNA purity was determined by Bioanalyzer 2100 traces (Agilent Technologies). Total RNA was amplified with the WT-Ovation Pico RNA Amplification system (NuGEN) according to the manufacturer's instructions. RNA was converted to cDNA and went under fragmentation and biotinylation with FL-Ovation™ cDNA Biotin Module V2 (NuGen) according to the manufacturer’s instructions. Fragmented, labelled cDNA was then hybridized to Affymetrix human genome U133A 2.0 microarrays. Prior to analysis, microarray data were pre-processed and normalized using robust multichip averaging, as previously described (214). Genes differentially expressed between Tregs (ex vivo and stimulated nTregs and ex vivo mTregs) and Tconv cells (ex vivo and stimulated populations of naïve and memory Tconv cells) were ranked using the signal to noise metric (215) with the GENE-E software package (216). The statistical significance of differentially expressed genes was determined using the comparative marker selection module in GENE-E (216). Hierarchical clustering was performed using GENE-E (216). To identify Treg-specific genes common to both species, Gene Set Enrichment Analysis (GSEA)  was performed as described previously (217).   62 Quantitative RT-PCR. RNA was isolated from cells using E.Z.N.A.  Total RNA Miniprep Kit (Omega Bio-Tek) and cDNA was generated using qScript cDNA SuperMix (Quanta Biosciences) according to the manufacturer’s protocol. The Taqman Gene Expression Master Mix (Life Technologies) was used to quantify mRNA levels. Data presented are normalized to β-2-microglobulin (B2M) using the comparative Ct method (ΔΔCt)  NanoString analysis NanoString nCounter Analysis System (218) was used to validate gene expression previously determined by microarray analysis. mRNA was detected using a custom-made nCounter Reporter probe set of 37 transcripts  (31 genes of interest and 6 reference genes). 100 ng of total RNA or 2 μL of whole cell lysate (5000 cells/ μL of Qiagen RLT lysis buffer) were hybridized overnight at 65°C with the capture and reporter probes according to nCounter Gene Expression Assay Manual. Following hybridization, excess probes were washed away on the nCounter Prep Station. To quantify the number of RNA molecules, colour-coded barcodes on the reporter probes were read on nCounter Digital Analyzer. The nCounter data were normalized to the reference genes and then subtracted by the background using the NanoString nSolver Analysis Software.  Statistical analysis. All analyses for statistically significant differences were performed with 1-tailed paired Student’s t test. P values of less than 0.05 were considered significant. All error bars represent standard errors unless otherwise indicated.   63 4.3 Isolation and phenotypic characterization of subsets of human Tregs defined by the expression of CD25 and CD45RA  To isolate naïve and memory populations of Tregs and Tconv cells, I first pre-sorted the CD4 +  cells into CD25-positive and CD25-negative fractions with anti-CD25 magnetic beads (Fig. 4.1A). Afterwards, the cells were further sorted by FACSAria into their respective subsets based on expression of CD25 and CD45RA. To obtain pure populations of Tregs, a stringent gating strategy was used and only CD25 ++ cells were sorted. As previously reported by Miyara et al., memory CD25 +  cells contain not only mTregs but a contaminating non-suppressive and cytokine-secreting T cell subset that also expresses CD25 and intermediate levels of FOXP3 (35). Indeed, as evidenced by our CD25 +  flow cytometry plot, there is a population of CD45RA - CD25 ++  cells that expresses similar CD25 level as the naïve Tregs. Therefore to avoid collecting these contaminating cells, the mTregs were sorted from the CD25 +++ cells.  Assessment of Treg purity was performed by FOXP3 staining on ex vivo sorted cells (Fig. 4.1B). As expected, the majority of naïve (mean±SEM; 86.0±2.3%) and memory (91.7±3.8%) Tregs were expressing FOXP3, compared to the near absent level in naïve (3.3±0.6%) and memory (10.3±1.6%) Tconv cells. FOXP3 staining was performed on all populations once more after stimulation with CD3/CD28-coated T cell expander beads in the presence of 100 U/mL IL-2 for 40 h (Fig. 4.1C). 40 h time point was chosen because activation-induced FOXP3 expression in Tconv cells reach maximum at 72 h (32). Thus, this time point allows enough stimulation for all T cells while differential FOXP3 expression can still be observed between Treg and Tconv populations. 91.7±3.8% of stimulated nTregs expressed FOXP3. Compared to ex vivo cells, this percentage of FOXP3-expressing cells  64 increased slightly and may reflect the optimal suppressive capacity of Tregs after activation. On the other hand, 62.4±7.2 % of mTregs expressed FOXP3 after stimulation. As seen on the flow cytometry plot, quite a few of the cells are CD25-negative. These cells may be dying as mTregs have been described to display high susceptibility of apoptosis after activation (5). Similar to their ex vivo populations, a low proportion of naïve (5.8±0.7%) and memory (17.8±2.8%) Tconv cells expressed FOXP3.  To investigate the homogeneity of the purified Treg populations, I determined the expression of several surface markers associated with Tregs on the ex vivo T cell subsets (Fig. 4.2). CCR7 and CD62L are associated with the naïve phenotype since they guide the cells to lymph nodes (182). As expected, a high proportion of naïve Tregs (87.0±1.9% CCR7; 71.6±10.2% CD62L) and Tconv cells (97.2±0.7%; 72.3±9.8%) expressed the two markers. Approximately half of the memory Tregs (41.7±3.7%; 63.1±8.9%) and Tconv cells (64.6±5.2%; 47.0±4.4%) expressed CCR7 and CD62L; this observation was expected since these two markers are also expressed on central memory cells (219). CD39 has recently been indicated as a good marker to separate between the mTregs and the CD25 ++ FOXP3 +  subset (54). Low proportions of nTregs (12.8±5.8%), nTconv cells (3.0±1.6%) and mTconv cells (14.8±1.8%) expressed CD39. However, a high percentage of mTregs expressed CD39 (88.7±1.3%), indicating that the stringent gating strategy was effective at removing the contaminating cells. Lastly, CD127 level was determined, and Tregs (naïve: 19.4±1.9%; memory: 8.1±2.0%) expressed a low level of the protein while the opposite was observed on Tconv cells (naïve: 67.2±5.1%; memory: 73.1±1.3%).  To assess the in vitro suppressive capacity of each ex vivo fraction, I measured the amount of thymidine incorporated by CD4 +  T cells cocultured with these cells after 6 days  65 (Fig. 4.3). Due to the limited numbers of Tregs after sorting, I used a microsuppression assay requiring only 8000 Tregs/ well at the 1:1 Treg: CD4 responder ratio (220). For all of the suppression assays, I used frozen CD4 +  T cells from the same donor to avoid variability in the proliferation ability of the responder cells. As expected, strong suppression of CD4 +  T cells was observed by both naïve and memory Tregs at the highest ratio and declined proportionally as the number of Tregs decreased. Even though mTregs seem to die after activation, they can still suppress efficiently as previously reported (5). On the other hand, naïve and memory Tconv cells did not have any detectable suppressive ability.  Together, these data demonstrate that sorting on the basis of CD45RA and CD25 expression allows isolation of naïve and memory Tregs that possess potent suppressive capacity and are not contaminated with non-regulatory CD25 ++ FOXP3 +  cells. I was thus confident in using these samples for gene expression array analysis.  Figure 4.1 Gating strategy for naive and memory Tregs and their respective ex vivo and stimulated FOXP3 expression. (A) CD4 +  T cells were pre-sorted into CD25-positive and negative fractions by anti-CD25 magnetic beads. CD25 + population was further sorted by FACSAria into two subsets of Tregs defined by the expression of CD45RA and CD25: I. CD25 ++ CD45RA +  nTregs and II. CD25 +++ CD45RA - mTregs. Similarly, CD25-negative population was sorted into two subsets  66 of Tconv cells based on CD45RA: III. CD25 - CD45RA +  nTconv cells and IV. CD25 - CD45RA -  mTconv cells. (B) Expression of FOXP3 and CD25 in ex vivo subsets was measured by flow cytometric analysis. (C) Expression of FOXP3 and CD25 in stimulated subsets after 40 h stimulation with CD3/CD28-coated T cell expander beads in the presence of 100 U/mL IL-2 was measured by flow cytometric analysis.  Data in A to C are representative of 7 different donors.    Figure 4.2 Surface marker expressions of naïve and memory subsets of Tregs and Tconv cells. Expression of CCR7, CD62L, CD39 and CD127 on the four ex vivo subsets of CD4 +  T cells was determined by flow cytometric analysis. Data are the average of 7 different donors.    Figure 4.3 In vitro suppressive activity of naïve and memory Tregs. Proliferation of CD4 +  responders (8000 cells/ well) was assessed by stimulation with CD3/CD28-coated T cell expander beads for 6 days in the presence of the indicated ratios of the four sorted T cell subsets. [ 3H]thymidine was added to the cultures for the final 16 h before harvesting.   67 4.4 Establishment of the molecular signature of human Tregs through microarray analysis RNA was extracted from ex vivo naïve and memory subsets of Tregs and Tconv cells as well as their corresponding 40 h stimulated samples, leading to 8 samples from each donor. Samples from 7 independent healthy donors were collected and a total of 56 samples were sent to the Genome Quebec Facility. Due to my stringent gating strategy of both subsets of Tregs (3% of CD4+CD25 hi  cells), the amount of RNA obtained from these samples was insufficient for the Affymetrix microarray platform. Therefore the RNA underwent an extra amplification step in order to reach the recommended amount of 5 μg by Affymetrix. In order for all the samples to be treated equally, the Tconv cells also went through the same amplification step. All the amplified RNA passed the quality control guidelines set by Genome Quebec (A260/ A280 ratio: 1.9 – 2.1 and RNA Integrity Number > 8.0). Finally, the RNA was hybridized to the Affymetrix U133A 2.0 microarray chip suggested by Dr. Nick Haining, my collaborator on this project, since he has previously published microarray gene expression data on human CD4 +  T cells using the same platform (213). Prior to data analysis, assessment of the data quality for the microarray was performed using a statistical package called affyQCReport, available from (221). Four samples did not pass the quality control after the assessment. As a result, 52 samples were included in the analysis instead of 56 as originally planned. Using the “Maker Selection” module of the GENE-E microarray analysis software (available from, 13321 genes were differentially expressed between the Treg and Tconv populations (216). Based on the low FOXP3  68 expression on stimulated mTregs (Fig. 4.1C), this Treg subset was excluded from the overall Treg population used in the “Marker Selection” analysis (Table 4.1). Among the 13321 genes, 1175 genes were found to be significantly different between Tregs and Tconv cells based on the p-value ≤ 0.05 and false discovery rate (FDR) ≤ 0.25: 618 genes were upregulated in Tregs and 557 genes were down-regulated. FDR was calculated using the Benjamini and Hochberg procedure (222) and represents the estimated probability of false positives within a set of genes that is differentially expressed. For example, a FDR of 25% indicates that the result is likely to be valid 3 out of 4 times. According to the recommendation by the Broad Institute where GENE-E software was developed, selecting genes based on a FDR of less than 25% allows selection of promising targets for validation purposes while preventing the possible overlook of significant differences when more stringent cut-offs are used.  To further narrow down the list of Treg-specific genes, I compared my list of 1175 genes with other published Treg datasets (mouse and human) available on the NCBI Gene Expression Omnibus (GEO) repository by applying Gene Set Enrichment Analysis (GSEA) (217). Classically, gene expression analysis focuses on identifying individual genes that are differentially expressed between two classes. However the changes between cells from one state to another are sometimes difficult to capture in a single gene level but may be reflected in a group of genes (202). GSEA is a general analytic approach that compares a ranked- ordered profile of genes (e.g., genes differentially expressed between Tregs versus Tconv cells) with an independent set of Treg genes from a separate experiment. Genes that are upregulated in Tregs from the publicly available data should be found or “enriched” at the top of the ranked list generated from my data analysis. Similarly, genes that are down-  69 regulated should be enriched at the bottom of my ranked list while genes that are not differentially expressed in both populations should be scattered randomly in the middle. In addition, since GSEA is able to compare datasets across different species, genes that are fundamentally important to Treg development and function conserved between human and mouse can be identified. In order to perform GSEA, 2 human and 2 mouse Treg microarray datasets (Table 4.2) from published literature were obtained from GEO and extracted using GENE-E to create lists of genes differentially expressed between Treg and Tconv populations. Each dataset generated 2 lists, one containing genes upregulated and the other with genes down-regulated in Tregs. Since there were 4 datasets, 8 gene lists were created, which were combined into one excel file. Taking the microarray file of 1175 genes and the excel file containing 8 columns of genes from the 4 datasets, the two files were entered into the GSEA software ( The software goes through each individual column and determines whether the defined gene set occurs at the top (or bottom) of the rank-ordered set of genes differentially expressed between my Treg and Tconv samples. The degree of similarity is reflected in the enrichment score with a range of 1 (positive correlation) to -1 (negative correlation) (Table 4.2). The set of genes that contribute most to the enrichment score is called the leading edge subset. For each dataset, GSEA generated an enrichment plot graph and a table containing the genes in the leading edge subset (217). For example, the enrichment plots of the human Treg dataset from Mold et al. (223) were as expected; genes that were determined to be upregulated in Tregs from their dataset clustered at the top of my ranked gene list (Fig. 4.4A), while genes down- regulated in Tregs clustered at the bottom of the list (Fig. 4.4B). This observation suggests that the microarray data that I generated was valid and related to the Treg phenotype.  70  Before comparing the genes obtained from GSEA to my microarray data, I applied one more restriction to the list of 1175 genes in order to create a more manageable number of genes. To generate a robust Treg signature, the difference in gene expression values between Tregs and Tconv cells must be as large as possible. Therefore, using “Filter Rows” in GENE- E, genes were filtered out if they had an absolute difference value of less than 750 between these two populations. This resulted in 232 genes. Next, individual gene profile plots were examined to ensure that genes highly expressed in Tregs were not in Tconv cells and vice versa. For instance, Ikaros family zinc finger 2, IKZF2 (Helios), a gene preferentially expressed by naturally-occurring Tregs (224), had an ideal gene profile plot where Tregs, regardless of activation status, expressed this gene to a high level while low expression was observed in Tconv cells (Fig. 4.5). By combining information from the gene plots as well the list of genes from GSEA, I selected 24 genes that were upregulated and 9 genes that were down-regulated (total of 33 genes) in Tregs for validation by qPCR from new sets of samples. The genes upregulated in Tregs can be broadly separated into 3 categories: (1) Genes upregulated in all subsets of Tregs (i.e. IKZF2); (2) Genes upregulated in stimulated subsets of Tregs (i.e. FOXP3, TNFRSF1B, CTLA-4); and (3) Genes upregulated in mTreg subsets (i.e. CSF2RB, CCR8). On the other hand, genes down-regulated in Treg populations were uniformly decreased in all Treg subsets.            71 Cell state Subsets # of samples from unique donors for analysis Ex vivo Naïve Treg (CD45RA + CD25 ++ ) 7  Naive Tconv (CD45RA + CD25 - ) 6  Memory Treg (CD45RA - CD25 +++ ) 6  Memory Tconv (CD45RA - CD25 - ) 6 Stimulated Naïve Treg 7  Naïve Tconv 7  Memory Tconv 7  Table 4.1 Summary of the samples in each subset of T cells for the microarray analysis.  Comparison Enrichment score Species Reference Adult CD4 + CD25 bright  Tregs vs CD4 + CD25 - CD45RA + CCR7 +  naïve Tconv cells 0.48/ -0.61 Human (223) CD4 + CD25 +  Tregs vs CD4 + CD25 -  Tconv cells 0.34/ -0.48 Human (35) CD4 + CD25 +  Tregs vs CD4 + CD25 - Tconv cells 0.28/-0.31 Mouse (225) Foxp3 +  Tregs vs FP3-deficient T cells 0.19/ -0.30 Mouse (187)  Table 4.2 Summary of the four Treg datasets obtained from Gene Expression Omnibus and analyzed in GSEA.   Figure 4.4 Profiles of enrichment plot and positions of Treg gene set members on the ranked order list. (A-B) A ranked-ordered list of genes comparing my Treg and Tconv populations was generated by GSEA. The software then determined the position of each gene in the dataset from Mold et al. relative to the rank-ordered list. On the left is the enrichment plot that shows genes upregulated in Tregs is positively correlated with my Treg dataset. On the right is the  72 enrichment plot that demonstrates genes down-regulated in Treg is negatively correlated with my Treg dataset. The leading edge subset of genes for each enrichment plot is indicated.   Figure 4.5 Gene profile plot of IKZF2 (Helios). Absolute expression value for IKZF2 was plotted by GENE-E for different subsets of Treg and Tconv populations.  4.5 Quantitative PCR validation of the Treg gene signature  Due to the limited frequency of nTregs and mTregs after sorting, an expansion phase of the subsets was required in order to have enough RNA for the qPCR validation of 33 genes. Modifying my protocol after the Bluestone group (77), FACS-sorted cells from a new cohort of healthy donors were expanded with CD3/CD28-coated T cell expander beads at a 1:1 bead:cell ratio in the presence of 1000 U/mL IL-2 for 10 days. At the end of the culture period, a mean expansion of 50-fold for nTregs and 13-fold for mTregs was obtained (data not shown). The low expansion capacity of mTregs may be due to their susceptibility to activation-induced apoptosis (5). I focused the validation qPCR on nTregs since these cells were over 90% positive for FOXP3 after the expansion phase as opposed to the 70%  73 observed in mTregs (data not shown). Although a stringent gate was placed when I sorted out mTregs, a few contaminating IL-17-producing non-suppressive memory T cells could have been sorted out in parallel since there is currently no marker to differentiate between these two populations. During the expansion phase, these activated T cells would then proliferate and contribute to the reduced FOXP3 expression. Besides comparing the gene expression of expanded nTregs to expanded nTconv cells, I also compared this expanded subset to ex vivo nTconv cells. In the context of drug screening assay, compounds that can convert nTconv cells to Tregs and vice versa have great therapeutic potential in human diseases. In order to successfully distinguish between these two cell states (Treg versus Tconv), the difference in gene expression level between the subsequent and initial population should be large. I considered a gene that was determined to be upregulated in Tregs by microarray to be validated if the fold change in qPCR was greater than 3. On the other hand a gene down- regulated in Tregs was validated if the fold change was greater than -3. As expected, known genes associated with Tregs such as FOXP3, CTLA-4 and IL-7R (6) were among the validated genes. These genes were used as controls since their expression pattern in Tregs is well characterized. For the genes in the IL-1 family (IL-1R1, IL-1R2, IL-1RN), these transcripts were undetectable (UD) in the Tconv cells indicated in the appropriate columns. Therefore although the Ct values for the Tregs were determined, the fold changes of these genes were unable to be calculated. Nonetheless, this observation indicates that there is a large difference in the expression of IL-1-related genes between these two populations, suggesting they are ideal candidates to include in the Treg gene signature. Similarly, the Ct value for LRRC32 was measured in nTregs but remained undetectable in both the expanded nTconv cell line and ex vivo nTconv cells. On the other hand, STAM was detected in nTconv  74 cells but not in nTregs. In summary, out of the 33 genes chosen from the microarray data, 15 genes were validated with 12 genes upregulated and 3 genes down-regulated in Tregs (Table 4.3).  Since only 15 genes were validated using qPCR, I decided to return to the original microarray data to select new genes for re-validation. The initial analysis from the GENE-E software generated 232 genes after I had set the difference in absolute gene expression between Tregs and Tconv cells to be greater than 750.  This restriction might be too stringent and could have removed some of the potential Treg genes. Therefore I removed this criterion, which meant starting again from the 1175 differentially expressed genes between Treg and Tconv populations. Further enrichment analysis using the GSEA software with additional lists of Treg genes obtained from GEO (226-228) along with the 4 previous datasets was performed (Table 4.2). The new analysis resulted in 16 new Treg-specific candidates in addition to the 15 validated genes. Hierarchical clustering using this updated list of 31 genes correctly separated Treg and Tconv populations and further grouped them into their distinct subsets based on the activation and memory status (Fig. 4.6 and Table 4.4). To further confirm this observation, principal component analysis (PCA) was executed to determine whether this gene list can accurately cluster the related cells into previously defined fractions in a qualitative format (Fig. 4.7). Similar to the hierarchical clustering, the Treg signature separated the cells into the 8 individual clusters of Tregs and Tconv cells. Taking mouse (225) and human (223) Treg datasets from publicly available literature (Table 4.2), the Treg signature distinctively teased apart the Treg and Tconv populations (Fig. 4.8). Due to the mouse data being run on a different microarray platform, several gene names were not found in the list resulting in a shorter heat map (Fig. 4.8A). Moreover, the signature also  75 distinguished between healthy individuals and patients with ulcerative colitis (226), an autoimmune disease associated with dysfunctional Tregs in the colon (229) (Fig. 4.9). Interestingly, the microarray data were not obtained from Tregs of these participants but rather from the colonic biopsies, suggesting this Treg signature has the potential to be developed into a diagnostic tool. Together, these analyses indicate that a robust gene signature has been established for human Tregs. Fold change in relative expression Expanded nTreg/ Expanded nTconv  Expanded nTreg/ Ex vivo nTconv   Donor 1 Donor 2 Donor 3 Donor 4 FOXP3 46.8 41.9 179.8 506.2 CSF2RB 14.3 10.5 5.3 14.3 LRRC32 UD UD UD UD TNFRSF1B 3.6 3.5 8.8 123.0 IL1R1 UD 10.4 UD UD IL1R2 71.4 UD UD UD IL1RN 1.7 -1.7 UD UD IKZF2 14.3 48.3 2.1 11.37 CTLA4 3.2 10.5 -1.3 6.2 ZBTB38 3.2 4.0 0.21 -1.77 TNFRSF9 -2.0 -2.1 UD 37.4 IL7R UD -50.0 -265.5 -9.51 LPIN2 -2.9 -2.6 -6.2 -5.31 STAM UD UD -2.9 UD NELL2 -457.4 -424.7 -49.0 -18.1  Table 4.3 Quantitative PCR validation of the Treg signature. The fold change in expression of genes indicated was determined by q PCR. Values shown are the fold changes of genes calculated over either expanded nTconv cells or ex vivo nTconv cells. Relative expression is normalized to B2M expression. Genes are separated based on whether they are determined to be upregulated (top) or downregulated (bottom) in Tregs by the microarray analysis. UD indicates that the fold change is undetermined due to undetectable Ct value in either Tregs or Tconv cells.       76  Figure 4.6 Microarray gene expression profiling can discriminate between Treg and Tconv populations. Expression profiles of 31 genes differentially expressed between my Treg and Tconv populations. Hierarchical clustering was performed and the Treg gene signature distinguished between these two populations and further separated each population into distinct subsets. Each row represents genes and column represents individual donor.   77 Gene name Full name Description Higher expression in Tregs compared to Tconv cells IL-1R1 Interleukin 1 receptor, type I Receptor for IL-1Rα, IL-1Rβ and IL-1RN. Involved in many cytokine induced immune and inflammatory responses. IL-1R2 Interleukin 1 receptor, type II Serves as a decoy receptor for IL-1β and prevents its binding to IL1R1. The combination of LAP, IL-1R1and IL- 1R2 distinguishes activated Tregs from FOXP3 +  Tconv cells in expansion cultures. IL-1RN Interleukin 1 receptor, antagonist Inhibits the activity of IL-1 by binding to receptor IL-1R1. IKZF2 IKAROS family zinc finger 2 (Helios) Differentiates naturally-occurring from peripheral induced Tregs. CTLA-4 Cytotoxic T-lymphocyte antigen 4 Tregs constitutively express CTLA-4, which downregulates CD80 and CD86 on DCs. Activated Tconv cells also express it. FOXP3 Forkhead box P3 Transcription factor essential for functional Tregs. LRRC32 Leucine rich repeat containing 32 (GARP) Serves as a receptor for latent TGF- β. Identifies activated Tregs from IL-17-producing activated Tconv cells. TNFRSF1B Tumour necrosis factor receptor superfamily, member 1 (TNFR2) Mediates most of the metabolic effects of TNF-α. IL-1R1 and TNFRSF1B are preferentially expressed on resting ex vivo isolated Tregs. TNFRSF9 Tumour necrosis factor receptor superfamily, member 9 (4-1BB) Contributes to the clonal expansion, survival, and development of T cells. CSF2RB Colony stimulating factor 2 receptor, beta, low-affinity (granulocyte-macrophage) Common beta chain of the high affinity receptor for IL-3, IL-5 and CSF. TRIB1 Tribbles homolog 1 Interacts with MAPK kinases and regulates activation of MAP kinases. METTL7A Methyltransferase like 7A Stimulates de novo methylation. RBMS3 RNA binding motif, single stranded interacting protein Mediates DNA replication, gene transcription, cell cycle progression and apoptosis STAM Signal transducing adaptor molecule (SH3 domain and ITAM motif) 1 Involved in intracellular signal transduction mediated by cytokines and growth factors ZBTB38 Zinc finger and BTB domain containing 38 Acts as a transcriptional activator. VAV3 Vav 3 guanine nucleotide exchange factor Binds physically to the nucleotide-free states of  GTPases and plays an important role in angiogenesis. ZNF532 Zinc finger protein 532 May be involved in transcriptional regulation by contributing to the stability of DNA domains. ECOP Vesicular, overexpressed in cancer, prosurvival protein 1 Increases the transcriptional activity of NFKB1 by facilitating its nuclear translocation.  ICA1 Islet cell autoantigen 1, 69kDa Acts as an autoantigen in T1D. C8ORF70 Chromosome 8 open reading frame 70 Unknown function HPGD Hydroxyprostaglandin dehydrogenase 15-(NAD) Metabolizes prostaglandins, which function in a variety of physiologic and cellular processes such as inflammation. Lower expression in Tregs compared to Tconv cells TMEM23 Transmembrane protein 23 Prevents cell death in response to osmotic stress PTPRK Protein tyrosine phosphatase, receptor type, K Regulates processes involving cell contact and adhesion.  IL-7R Interleukin 7 receptor (CD127) Blocks apoptosis during differentiation and activation of T cells. Absence of CD127 and high expression of CD25 is widely used to define functional Tregs.  78 LPIN2 Lipin 2 Plays important roles in controlling the metabolism of fatty acids at different levels. ID2 Inhibitor of DNA binding 2 Inhibits the function of basic helix-loop-helix transcription factors and may negatively regulate cell differentiation. ANK3 Ankyrin 3 Regulates activities such as cell motility, activation, proliferation, contact, and the maintenance of specialized membrane domains. DACT1 Dapper, antagonist of beta- catenin, homolog 1 Enhances the transcriptional activation of target genes of the Wnt signalling pathway. ABCB1 ATP-binding cassette, sub-family B (MDR/TAP), member 1 Acts as energy-dependent efflux pump responsible for decreased drug accumulation in multidrug-resistant cells. NELL2 NEL-like 2 (chicken) Plays a role in neural cell growth and differentiation as well as in oncogenesis. HDGFRP3 Hepatoma-derived growth factor, related protein 3 Enhances DNA synthesis and may play a role in cell proliferation.  Table 4.4 Description of genes in the Treg signature.  79  Figure 4.7 Principal component analysis of the 31 Treg-specific genes. Relationship between different clusters was illustrated by PCA. This analysis shows that Tregs and Tconv cells can be separated into two categories. Additionally, the gene signature can further distinguish individual subsets with the each population.   80  Figure 4.8 Treg gene signature can differentiate Treg and Tconv populations from both mouse and human. Treg microarray datasets were obtained from NCBI GEO repository and described in detail in Table 4.2. By applying the Treg gene signature to (A) mouse and (B) human Treg datasets, differential gene expressions can be observed on the heat maps with the correct populations clustered together. Each row represents genes and column represents individual sample or donor.    81   Figure 4.9 Treg gene signature can discriminate between health controls and patients with ulcerative colitis. Genome-wide gene expression data covering the gene expression profile of colonic biopsies from health controls and patients with ulcerative colitis were obtained from NCBI GEO repository (reference series GSE9452). Hierarchical cluster of the data was performed by applying my Treg gene signature. Each row represents genes and column represents individual participants.    82 4.6 Multiplex validation of the Treg gene signature using NanoString nCounter Analysis System The primary validation assay was performed using real time qPCR since this technique is the gold standard for measuring gene expression. However, qPCR can only measure a limited number of transcripts and, depending on how many transcripts need to be quantified, requires a relatively large amount of starting material as evident by the need to expand the Tregs for the initial qPCR validation. Therefore, to perform the secondary validation screen, I measured gene expression using the NanoString nCounter Analysis System. This multiplex platform can measure expression of up to 800 transcripts at a time and since it does not require mRNA amplification, there is no bias introduced by enzymatic reactions (218). For each mRNA of interest, two 50-base probes complementary to the sequence are hybridized: the Reporter Probe, coupled to a colour-coded tag, is complementary to the 5’ portion of a target sequence and provides the detection signal; and the Capture Probe is complementary to the 3’ portion of a target sequence and allows the complex to be immobilized in the solid phase with a biotinylated affinity tag so that non-hybridized material can be washed away. Each transcript is digitally quantified by counting the number of times the corresponding string of coloured fluorophores is detected. Relevant to the restricted number of naïve and memory Tregs after sorting, sufficient sample can be obtained from as few as 10,000 cells and be directly loaded onto the platform as whole cell lysates (230). Whole cell lysate is the suspension of the FACS-sorted cells in Qiagen lysis buffer thereby avoiding the need to purify RNA and loss of the RNA through the isolation process. Therefore, unlike microarray technology, it is unnecessary to amplify the starting material.  83 In total, 48 assays were purchased from NanoString which means samples from 6 donors (8 samples/ donor) could be performed. Since I had no previous experience with the NanoString platform, I decided to first run samples from 3 donors. RNA from 2 donors that were analyzed on the original microarray were included as well as a new set of samples isolated as whole cell lysate. From this experiment, I hoped to see that the input material of either RNA or whole cell lysate would not make a difference regarding the trend of the gene expressions. In addition, to normalize the data, it is essential to choose genes that are stably expressed in all the cells across both ex vivo and stimulated conditions. Unfortunately, my microarray data showed that common housekeeping genes such GAPDH or B2M vary in expression as T cells get activated; in fact, there was a log difference between the B2M Ct value of ex vivo and stimulated Tconv cells (data not shown). Hence, the recommendation from the NanoString technical support was to select genes from the microarray data with low, medium and high expression levels. Therefore I chose 6 reference genes by setting a parameter in GENE-E to look for genes with as little variation as possible across the different samples (Fig. 4.10). Any variation in gene expression due to activation status should be flattened out when the average is generated from these 6 genes. Data were collected using the NanoString nCounter Analysis System to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. Before the analysis, quality control metrics implemented in the software were performed on the raw counts of each sample to assess the technical performance in the gene expression assay (218). All the samples passed the quality control. Next, I used the software to calculate the geometric mean of the 6 reference genes and normalized data of the 31 microarray-chosen genes was generated. Lastly, I determined the fold change of each gene by dividing the normalized counts of nTreg  84 over nTconv populations. In this case, each population included both ex vivo and stimulated cells. As mentioned previously, nTregs are more homogeneous than mTregs and they maintain a high expression of FOXP3 even after 40 h of stimulation. Therefore I decided to start the nCounter analysis with only the naïve subsets. First of all, I wanted to determine whether the data obtained from nCounter System is comparable to Affymetrix microarray. Hence, I obtained the fold change of the genes from the microarray analysis (Fig. 4.11). In parallel, I calculated the fold change from the nCounter normalized counts. For the genes that were determined to be upregulated in Tregs by microarray, the majority of them were validated by NanoString (Fig. 4.11A).  The exceptions were HGPD and TMEM23, which had a low fold change by microarray to begin with. All the genes that were down-regulated in Tregs measured by microarray were validated by NanoString (Fig. 4.11B). Notably the fold change in Fig. 4.11A from NanoString seemed greater whereas the opposite was true in Fig. 4.11B. The difference in the level of fold change generated by the two platforms could be due to the fact that microarray and nCounter were loaded with amplified and unamplified RNA respectively. Nonetheless, nearly all the genes selected from the microarray analysis were validated using the NanoString nCounter technology. Aside from loading RNA on the nCounter, the assay can also use whole cell lysate (230). Thus, I collected samples from a new donor and resuspended the 8 sorted populations in Qiagen lysis buffer. Following the recommendation of the NanoString technical support, 5000 cells were resuspended for every μL. Illustrated in Table 4.5 are the gene expression fold changes of nTregs over nTconv cells for each donor: the first column is the whole cell lysate and the last two columns are the RNA samples. Once again, majority of the genes measured from the whole cell lysate behaved as expectedly according to the microarray analysis. In this case, genes such  85 as C8ORF70, ECOP, HPGD, ICA1, TMEM23, HDGFRP3 and PTPRK were not validated. It is difficult to say currently whether these candidate genes need to be removed from the Treg gene signature since only one new donor was assayed. I am in the process of collecting samples from two more healthy donors. Once both sets of data are measured on the nCounter, I will have a better idea regarding which genes should be removed from the final Treg gene signature. In the meantime, I averaged the gene expression fold changes of the three sets of data and 5 genes (ECOP, ICA1, C8ORF70, HPGD and TMEM23) did not pass the nCounter validation study (Fig. 4.12). Overall, whole cell lysate can be run on the nCounter and the results are comparable to those obtained when purified RNA is used as a starting material.  Figure 4.10 Reference genes selected for use with the NanoString nCounter Analysis System. Absolute expression value in log2 scale for reference genes with high (RPL23A, HNRPA1), middle (EIF3S6, UFC1) and low (NSUN5B, PMS2L11) level was plotted by GENE-E for different subsets of Treg and Tconv populations.  86   Figure 4.11 Naïve Treg fold change comparison between the Affymetrix and NanoString platforms. (A+B) Fold change gene expression data comparing nTregs over nTconv cells from both Affymetrix and NanoString technologies were determined on the same two donors. Amplified RNA was used for the Affymetrix while unamplified RNA was used for the NanoString.     87 Treg upregulated microarray genes  Treg downregulated microarray genes Gene N1 A4 A11  Gene N1 A4 A11 C8ORF70 -1.6 2.8 1.8  ABCB1 -10.9 -26.4 -7.1 CSF2RB 2.4 11.1 17.2  ANK3 -6.1 -3.2 -3.2 CTLA4 7.1 25.1 20.0  DACT1 -11.0 -10.4 -9.2 ECOP 1.1 1.6 2.0  HDGFRP3 2.1 -8.0 -4.7 FOXP3 37.4 41.6 83.3  ID2 -5.5 -2.8 -3.6 HPGD 1.5 2.7 -1.2  IL7R -2.6 -1.5 -1.5 ICA1 1.1 1.2 2.3  LPIN2 -3.2 -2.4 -2.0 IKZF2 89.3 217.5 30.8  NELL2 -66.0 -13.4 -14.9 IL1R1 31.4 43.8 117.9  PTPRK 1.1 -1.4 -2.5 IL1R2 22.3 50.4 395.8 IL1RN 38.1 81.8 140.0 LRRC32 71.0 195.9 84.8 METTL7A 3.5 6.3 8.2 RBMS3 30.1 38.4 82.2 STAM 5.7 5.3 4.5 TMEM23 -1.9 1.1 -1.1 TNFRSF1B 4.7 6.1 6.9 TNFRSF9 12.5 16.6 3.0 TRIB1 6.6 9.7 5.7 VAV3 5.2 4.6 2.7 ZBTB38 2.1 4.2 6.9 ZNF532 2.8 3.5 6.1  N, NanoString (whole cell lysate); A, Affymetrix (RNA) Table 4.5 Fold change of the NanoString nCounter validation.  88  Figure 4.12 Summary of the NanoString nCounter validation. Expression changes in 31 genes that are differentially expressed between nTregs and nTconv cells. Data represent the average fold change of 3 donors.                      89 4.7 Discussion  In this study, I analyzed the transcriptional program of Tregs using DNA microarray technology by separating the Treg population based on CD25 and CD45RA into naïve and memory subsets. In parallel, Tconv cells were isolated in a similar manner and compared to their respective counterparts. Recently, significant biological differences between human naive and memory Tregs have been discovered. In humans, different splice isoforms of CD45 define subpopulations of CD4 +  T cells, with expression of CD45RA, but not CD45RO, being characteristic of naive T cells and CD45RO, but not CD45RA, of memory T cells (231). Notably, naive CD4 + CD45RA + CD25 hi  Tregs were found to be more homogeneous in expression of FOXP3 and CTLA-4 than the CD45RO +  subset (232). On the other hand, although memory CD45RO + CD4 + CD25 hi  Tregs are more likely to be contaminated with non- Tregs, they are more potently suppressive on a single cell basis, and have likely undergone changes in gene expression in vivo that stabilize and amplify their biological program (42, 233). A study by Miyara et al identified these contaminating non-Tregs as CD4 + CD45RA - Foxp3 low  cells, and demonstrated that they are not suppressive and secrete cytokines such as IL-2, IFN-γ and IL-17 (35). These non-suppressive memory T cells express CD25 but have a lower expression compared to the functional mTregs (35).  As expected, I observed differential CD25 expression between these two fractions in our pre-sort CD25 +  fraction. Therefore to avoid the contaminating cells, I was stringent with the gating strategy and selected mTregs only from CD25 +++ cells. In vivo suppression assays indicated these isolated mTregs potently suppressed the proliferation of CD4 +  responder cells, and the majority of these cells expressed CD39, an ectonucleotidase that is coexpressed with CD73 by a subset of mouse CD4 +  Tregs (234). In human CD45RA - CD25 +  T cells, Dwyer et al found this  90 marker distinguished between Tregs and activated T cells that produce pro-inflammatory cytokines (54). The observation that the mTregs that I sorted expressed high levels of CD39 further supports the purity of these cells. In summary, sorting Tregs and Tconv cells on the expression of CD25 and CD45RA resulted in sufficiently homogenous populations for microarray analysis as evidenced by the assessment of FOXP3 and surface marker expressions, as well as suppressive capacity.  To my knowledge, this is the first study to analyze the global transcriptome profile of carefully isolated pure populations of naïve and memory Tregs across multiple healthy donors. I also included parallel conditions of stimulated cells, as Tregs become suppressive only when activated through their TCR (235, 236). Therefore, functionally-relevant changes in gene expressions associated with Treg suppressive capacity could be teased out by the comparison between ex vivo and stimulated Tregs. To identify a Treg-specific signature, I utilized a powerful analytical approach, GSEA, for interpreting gene expression data (217). Straightforward comparison of microarray data between two classes (i.e. Tregs versus Tconv cells) yields hundreds or thousands of “potentially interesting” differentially expressed gene candidates (202) as indicated by my own result of 1175 statistically significant genes. Narrowing down this huge list of genes into a manageable size for extraction of meaningful information and validation purposes is difficult and subjected to a researcher’s own expertise bias. Moreover, the “fold-change”-based analyses of individual genes ignores important effects on pathways as the combinational expression changes of multiples genes of a single pathway could override a large increase observed in a single gene (217).  Therefore, the GSEA approach allows the comparison of my Treg data with multiple gene sets of human and mouse Tregs as well as disease datasets associated with Tregs. Genes of fundamental  91 importance to Treg development and function differentiation should be conserved between species evidenced by the gene signature of mouse and human memory T cells (237-239). Therefore, I refined the human Treg gene signature by comparing my data to high-quality publicly available data on mouse Tregs (187, 225) to further identify Treg-specific genes common to both species. In the end, 31 genes were found to be differentially expressed between the Treg and Tconv populations. The enrichment score generated by GSEA is an indication of how closely related one dataset is to another. Interestingly, a low enrichment score appeared when comparing my Treg dataset to one of the mouse Treg datasets which contained genes upregulated in Foxp3 +  Tregs versus Foxp3-deficient cells (187). Over-expression of FOXP3 in mouse and human CD4 +  Tconv cells is sufficient to recapitulate the Treg suppressor function, leading to the conclusion that this transcription factor is a “master regulator” in Treg development (9, 20). As a result, numerous studies have investigated how FOXP3 contributes to Treg suppressive function in the hope of finding therapeutic approaches to manipulate this population to regulate immune tolerance (18, 169, 240, 241). However, accumulating evidence suggests FOXP3 alone is not enough to drive Treg lineage commitment, rather there seems to be a higher-order organization upstream of this factor (241, 242). As demonstrated in these studies using Foxp3–green fluorescent protein (GFP) reporter mice in which the GFP insert renders the Foxp3 protein non- functional, Treg-like cells develop in the thymus and circulate in the periphery. These cells possess multiple Treg characteristics including a transcriptional active Foxp3 locus, high expression of the majority of Treg signature genes (including Il2ra, Ctla4 and Il10), and low expression of Il2, but lack suppressive activity (45). Moreover, when Foxp3 was expressed by direct transduction or by induced conversion with TGF-β, the  92 presence of Foxp3 could rescue at most about one third of the Treg signature transcripts (241). In summary, although Foxp3 is important for Treg effector function, it is neither necessary nor sufficient to induce the full Treg gene signature. Thus, these findings further support that the recapitulation of Treg identity does not depend on one single gene but rather involves multiple corresponding genes.  The ideal Treg gene candidate is one that has high expression in all Tregs and low in all Tconv cells regardless of the cellular activation status. Interestingly, out of 13321 genes on the microarray, only IKZF2 (Helios) displayed this characteristic. Unfortunately, at the protein level Helios is not Treg-specific and is also marker of activation (243). This observation reconfirms the need for a Treg signature comprised of multiple genes in order to accurately categorize these cells in various diseases. Although every gene in the signature may not have a large difference between Tregs and Tconv cells, the ability to separate these two populations depends on the input of all 31 genes as evidenced by the hierarchical clustering analysis. Notably, each population was correctly clustered into its appropriate subset. To demonstrate the robustness of the Treg signature, I applied it to human and mouse Treg datasets from other research groups (223, 225) and found that it was consistently able to differentiate between Tregs and Tconv cells. This outcome confirms the importance of the genes in the Treg function since they are conserved across different species. More studies are being currently performed to investigate the biology of these genes. Furthermore, the gene signature separated healthy controls and patients with ulcerative colitis into two distinct groups based on the RNA extracted from colon biopsies (226), confirming these genes are essential in the contribution of suppressive phenotype of Tregs and supporting a possibility of using the Treg signature as a diagnostic tool. Additional refinement of this gene signature  93 could perhaps distinguish between naïve and memory Tregs, which is supported by the fact that few genes in the signature were found only to be expressed on mTregs such as CSR2RB, TRIB1 and HPGD. The contributions of naive versus memory Tregs in regulating responses in vivo is still unclear as there are currently no good biomarkers that can distinguish these cells in states of health and disease. In parallel, the extrinsic signals that regulate the activation, function and homeostasis of different Treg subsets, and the gene expression profile underlying such responses also remain poorly defined.  Secondary validation of the Treg gene signature was performed on the NanoString nCounter Analysis System. Presently, I ran samples from three donors and confirmed that the data from RNA and whole cell lysate were comparable. The majority of the genes in the signature were validated using the NanoString nCounter platform. I am currently collecting whole cell lysates from additional donors in order to refine the Treg gene signature. In summary, I have established a human Treg signature which discriminates between Tregs and Tconv regardless of their states of activation.           94 5. CONCLUSIONS   The aim of this research was to investigate and further characterize the phenotypic properties of CD4 + FOXP3 +  Tregs. The functional consequences of exposure to adenovirus on interactions between human monocyte-derived DCs and Tregs were determined. I found that RAdV-transduced DCs were still susceptible to immunosuppression by Tregs. Therefore, the data suggest genetically engineered DC-based cancer vaccines need to be administered in parallel with Treg blocking agent to achieve their maximal efficiency. In an effort to explore novel approaches to attenuate Treg activity, I investigated the role of chemokine production from Tregs and found that its function can be potentially linked to the suppressive ability of Tregs. Moreover, in order to precisely differentiate Tregs from Tconv cells, I generated an accurate molecular signature of human Tregs using Affymetrix microarray technology. The resulting gene expression data can then be used to discover new aspects of the cellular and molecular biology of human Tregs as well as to conduct parallel genome-scale RNA interference and chemical screens to identify pathways and compounds that can be exploited specifically to accelerate or eliminate Treg differentiation in humans. DCs are the major regulators of T- and B-cell immunity, due to their superior ability to take up, process and present antigens compared to other APCs. As immature cells, DCs travel throughout the peripheral and secondary lymphoid tissue, sampling and ingesting foreign and self Ag. Presentation of Ag to T cells in this immature state results in tolerance either due to the deletion of Ag-specific T cells, T-cell anergy, or the induction of Tregs. On the other hand, Ag presentation by DCs that encountered inflammatory stimuli and matured, elicits T-cell activation and immunity. In Chapter 2, my results highlighted the importance of  95 depleting Tregs in order to elicit the full immunostimulatory potential of RAdV-transduced DCs. I proposed that one suppressive mechanism that Tregs exert on RAdV-transduced DCs is by downregulating their expression of CD80 and CD86. Inhibition of these costimulatory molecules may limit the ability of DCs to stimulate naïve T cells through CD28 leading to the dampening of the immune response. Therefore one approach to overcome the Treg- mediated suppression on these genetically engineered DCs is to abrogate the interaction between CTLA-4 and CD80/ 86. Indeed, co-cultures of human Tregs and DCs in the presence of anti-CTLA-4 monoclonal antibodies increased DC proliferation (244). An alternative method is to deplete the number of Tregs by interfering with the IL-2 pathway. The efficacy of this tactic is supported by a study that demonstrated anti-tumour vaccination with RNA-transfected DCs in combination with Treg depletion by IL-2-diphtheria toxin conjugate (Ontak) improved stimulation of tumour-specific T cells (245). Beyond the use of monoclonal Abs to specifically target Tregs, several other indirect strategies have also proven to be effective. For example, autologous DCs pulsed with lysate derived from three melanoma cell lines resulted in a reduced number of CD4 + TGF-+ Tregs and an enhanced anti-melanoma immune response (246). Another method is to use siRNA to knock-down expression of CCL22 and CCL7, chemokines which attract Tregs, in monocyte-derived DCs (247). This strategy diminishes Treg numbers and increases infiltrating CD8 + T cells in human tumour xenografts in athymic nude mice. Alternatively, radiofrequency thermal ablation in lung cancer patients can also reduce FOXP3 + Tregs and enhance IFN- production by CD4 + T effector cells (248). Aside from the influence of Tregs on DCs, the effectiveness of adoptively transferred TAA-specific T cells is also inhibited by Tregs. For example, in a phase I trial of melanoma patients, although transfer of TAA-specific T cells together with  96 fludarabine resulted in a 2.9-fold improvement in the life-span of T cells, there was a parallel increase in endogenous FOXP3 +  Tregs (249). Thus, in order to amplify the anti-tumour response elicited by cancer immunotherapy, it is necessary to control Treg proliferation while allowing the expansion of T effector cells. The field of DC-based vaccine was recently encouraged by the development of Sipuleucil-T, the first FDA-approved cellular vaccine, which provided a 4.1-month extension of prostate cancer patient survival (250).This vaccine comprises DCs pulsed with a fusion protein of prostate cancer-associated Ag and GM-CSF, supporting the hypothesis that combinational therapy of autologous DCs expressing tumour Ags in parallel with agents aimed to overcome the immunosuppressive tumour microenvironment may improve clinical outcomes. Overall, these studies demonstrate that there are many different ways to effectively deplete Tregs in humans. In the next five years, the most feasible and effective Treg depleting regimens will likely become routine when DC immunotherapy is administered as more strategies shown to work in animal models are translated into the clinic. Pro-inflammatory Th17 cells influence the outcome of anti-tumour immunity. However the precise role of these cells is still under debate. In patient studies, the presence of Th17 cells in the periphery correlates with reduced tumour mass and increased survival (251). On the other hand, in tumour-bearing mice, Th17 cells seem to contribute to potent anti-tumour responses due to the initiation of the inflammatory activity (251). Numerous studies have described populations of Tregs that co-express FOXP3 and IL-17 and retain their suppressive function in both mice and humans (42, 252, 253). From my data, RAdV- exposed mDCs induced IL-17 production from Tregs. These Tregs in turn had increased suppressive function compared to Tregs cocultured with regular mDCs. It would be  97 interesting to take Tregs from patients receiving DC-based cancer vaccines and determine whether those Tregs produce IL-17. In parallel, the clinical outcomes of the patients can also be correlated with the state of the Tregs in order to shed some light on the role of IL-17 in DC immunotherapy. To improve the efficacy of DC vaccines, the ex vivo manipulation of these cells is also critical. Monocyte-derived DCs have been used most commonly in clinical trials to date due to their abundance in blood (254). The cocktail that I used to mature DCs was comprised of TNF-α, IL-1β, IL-6 and PGE-2 (cytokine mDCs) and is the most widely used protocol (255). The inclusion of PGE-2 in the maturation regimen has been a debate in the field for some time. The rationale of using PGE-2 in the first place was to enable the DCs to migrate to the lymph nodes (256). However subsequent studies showed that in the tumour microenvironment, PGE-2 can polarize CD4 +  T cells into a Th2 phenotype and promote DCs to secrete IL-10 (257). In turn, it abrogates the ability of DCs to produce IL-12, an important licensing cytokine to mediate Th1 polarization and thus the induction of potent CTL responses (255). On the other hand, another study has demonstrated that although PGE-2- cultured DCs were not able to secrete IL-12, they regained the ability upon interaction with T cells in vivo (258). Therefore, the role of PGE-2 on DC polarization is still under debate. Nonetheless, my finding that the cytokine mDCs produced no IL-12 and a small amount of IL-10 is in line with the previous observation. A study by Sporri and Reis e Sousa showed that optimal activation of DCs required TLR signalling (259). Interestingly, in the presence of RAdV, which activates TLR2 and TLR9 (260), the cytokine mDCs still did not secrete IL- 12. Hence, it would be interesting to coculture these cytokine mDCs in the presence of T cells and see if they become IL-12 producers. Moreover, removing PGE-2 from the cytokine  98 maturation cocktail could be another option since DCs matured in the absence of PGE-2 can still migrate towards CCL21, one of the main chemokines that is constitutively expressed in the high endothelial venules of lymph nodes (261). Indeed, these mDCs were demonstrated to induce much higher numbers of functional CD8 +  T cells against leukemia cells (262). Investigating the susceptibility of DCs matured without PGE-2 to Treg-mediated suppression is of interest to future clinical trials. In summary, a consensus regarding the optimal maturation cocktail for human monocyte-derived DCs still needs to be established. In order to develop ways to target Tregs therapeutically and generate effective Treg- based therapies, it is critical to understand more about the factors that control their suppressive function. Surprisingly, while investigating the cytokine targets of Treg suppression, my colleague discovered their capacity to produce chemokines (170). Several studies have characterized the expression of chemokine receptors on Tregs that directs them to inflamed tissues and tumour sites to exert regulatory activity (166, 219, 263). However, research on the capacity of Tregs to produce chemokines and how this phenomenon relates to Treg functionality is limited. Emerging evidence has begun to unveil the role of chemokines in Treg immune regulation (170, 177, 178, 181). In Chapter 3, I found that both human and murine naturally-occurring Tregs secrete significant amounts of CCL3 and CCL4, indicating this is a conserved property. Since chemokines typically function to promote inflammation by recruiting innate and adaptive immune cells, these data prompt the question: why would Tregs attract pro-inflammatory immune cells? Due to the fact that Tregs require IL-2 produced by Tconv cells for their survival and function (264), I predict that Tregs use this mechanism to attract their own help and ensure their targets are in close proximity so that one of the many cell-contact-dependent mechanisms of suppression can be initiated. The  99 chemokine-mediated approach has been demonstrated in CD4 +  T cells where CCL3 and CCL4 secretion by these cells provide help for the generation of an optimal memory CD8 +  T- cell population (188). Therefore I next confirmed that these Treg-derived chemokines are indeed functional and can attract both CD4 +  and CD8 +  T cells in vitro, indicating Tregs also utilize the chemokine system for cell recruitment which may then influence the outcome of the immune system. Indeed, the ability to secrete chemokines seems to be related to their immune intervention since a lower proportion of CCL3 + cells were found in T1D Tregs compared to control counterpart. If Tregs attract their targets using chemokines, this feature would have implications for the development of Treg-based immunotherapies. In the context of cancer, siRNA have been used to knock-down expression of chemokines that attract Tregs to tumour sites (265). Hence, it would be interesting to investigate whether TAA-specific Tregs have an enhanced ability to produce CCL3 and CCL4. If that is the case, then designing siRNA to specifically interfere with the chemokine secretion by Tregs would be advantageous. On the other hand, it could be important to include chemokine production as one of the evaluation criteria before adoptively transferred Tregs are to be used to induce tolerance in autoimmune diseases. Other future directions will involve defining whether the various Treg subsets sorted on the basis of differential chemokine receptor expression will make altered panel of chemokines. Would CXCR3 +  Tregs, which suppress CXCR3 +  Th1 cells (266), make different chemokines compared to Tregs that express chemokine receptors representative of Th17 cells? Is it possible to alter the chemokine expression of Tregs in order to manipulate their suppressive ability? In parallel, it would be interesting to compare the chemokine profile from circulating versus tissue localized Tregs to gain insight into designing a targeted Treg therapy. Finally, investigation into whether chemokine secretion is  100 linked to other aspects of Treg suppressive mechanisms, i.e. immunosuppressive cytokines or cytolytic pathways, could reveal previous unknown Treg biology. Manipulation of Tregs using genetic or chemical means would have profound therapeutic implications for diseases such as multiple sclerosis, T1D, and allergy, whereas specific elimination of tumour Ag-specific Tregs could allow efficient immune-mediated elimination of malignancies. Upon identification of CD25, and subsequently FOXP3, as proteins that could be used to track and enumerate Treg in humans, many clinical studies have attempted to correlate changes in Tregs with outcomes in various types of autoimmunity and cancer (267, 268). Unfortunately, the results of many of these studies are difficult to interpret since both CD25 and FOXP3 are also expressed on activated human Tconv cells (45). Additionally, there are no other proteins known that are uniquely expressed by Tregs. Moreover, testing the function of Tregs using in vitro suppression assays is unlikely to accurately reflect the in vivo function of these cells, where the environmental milieu may decrease the susceptibility of other cells to being suppressed. At the same time, functional assays are cumbersome and difficult to perform in the clinical setting. To overcome these barriers, I have utilized integrative genomic approaches to define a molecular phenotype of human Tregs using 31 genes which can discriminate between Tregs and conventional naive or memory T cells regardless of their states of activation. The list includes established Treg-associated genes such as FOXP3, Helios, CTLA-4, CD127, several genes related to the IL-1 pathway, as well as genes previously unknown to be differentially expressed between Tregs and Tconv cells. Aside from segregating Tregs from Tconv cells, the gene signature can further cluster these two populations into respective naïve and memory subsets as well as ex vivo and stimulated fractions. This observation suggests that  101 each subset of Tregs in various activation states possess a unique set of genes. Therefore the attempt to apply the same surface markers for the isolation of all Tregs may skew the actual involvement of a particular type of Treg over another in a disease setting. Moreover, the gene list can distinguish Tregs from Tconv cells in datasets published from other groups, which indicates that the genes included in the signature are robust and reflect the Treg property. In the clinical context, biopsies of patients with ulcerative colitis was clearly separated from controls using the Treg gene signature, suggesting that the signature can be an indication of tolerance malfunction at least in autoimmune diseases. Further validation is necessary thus I have transferred the Treg gene signature into a multiplex platform suitable for clinical studies using the NanoString nCounter Analysis System. The nCounter System validated majority of the genes from my first set of samples with three healthy donors. Several genes translate into surface proteins including CSF2RB, IL-1-related proteins and TNFR family members. It would be interesting to confirm the surface expression of these genes on Tregs and determine if a different combination would better differentiate between ex vivo and activated fractions in naïve and memory Tregs. Indeed, the combination of LAP along with IL-1R1 and IL-1R2 has identified activated human FOXP3 +  Tregs after expansion from FOXP3 + non-Tregs (11). Moreover, in the Treg signature identified based on the microarray analysis, I excluded the activated mTreg fraction due to the reduced FOXP3 expression. In support of this finding, Hoffman et al. also observed a loss of FOXP3 in their mTregs after in vitro TCR stimulation (269). This group’s recent paper demonstrates that it is possible to isolate intact mRNA from fixed, permeablized, and FACS-sorted FOXP3-positive and -negative populations (270). I will follow the protocol to isolate mRNA from these two populations in order to run these  102 samples on the nCounter using the Treg gene signature. If the signature that I developed is highly specific to Tregs then a difference should be observed. The implication from this validation would mean that the gene signature can be utilized at the end of the expansion protocol to measure the purity of the Tregs before adoptive transfer therapy rather than relying on the cumbersome in vitro suppression assay. As illustrated by several studies, the correlation between in vitro and in vivo suppression activity is not perfect since Treg dysfunction that results in a severe spontaneous lymphoproliferative disease in vivo cannot be detected using in vitro suppression assays (271-273). Aside from collecting more samples from healthy donors to validate this gene signature, I will determine whether the signature can also be used to identify changes in Tregs from patients with T1D. Preliminary analysis of applying the gene signature to T1D versus healthy Treg dataset appears to discriminate between these two groups, although not perfectly, likely due to the poor purity of the cells used in this study (228). It is likely that the signature needs to be further refined to determine a core of T1D-related genes. I will be working with a biostatistician to identify a panel of genes that is most relevant in T1D. The establishment of the Treg gene signature will position my lab to carry out two long-term goals. First, to assay the signature in larger cohorts of T1D patients with different stages of disease and ask whether it can be used as a biomarker to identify patients at risk for developing T1D and track the “fitness” of Tregs throughout the course of disease in a clinically-applicable and non-invasive manner; and second, to conduct parallel genome-scale RNA interference and chemical screens to identify pathways and compounds that can be exploited to specifically accelerate or eliminate Treg differentiation in humans. Monitoring  103 the progression of T1D using the Treg gene signature could reflect the state of beta cells and may shed light on the optimal timing for therapeutic interventions. In summary, the research presented in this thesis has contributed to the advancement of the immunology field regarding Treg biology and provided new insights into the manipulation of these cells to alleviate human diseases such as cancer and autoimmunity. Three concepts have been explored: the impact of Tregs on RAdV-transduced DC-based cancer vaccine, the biological relevance of chemokines in Treg-mediated suppression, and the establishment of human Treg gene signature. Overall, the functional characteristics of Tregs have been examined through multiple approaches, which have led to an enhanced understanding of their relevance in clinical settings.   104 REFERENCES 1. Vignali, D.A., Collison, L.W., and Workman, C.J. 2008. How regulatory T cells work. Nat Rev Immunol 8:523-532. 2. Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M., and Toda, M. 1995. Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol 155:1151-1164. 3. Hori, S., Nomura, T., and Sakaguchi, S. 2003. Control of regulatory T cell development by the transcription factor Foxp3. Science 299:1057-1061. 4. Josefowicz, S.Z., and Rudensky, A. 2009. Control of regulatory T cell lineage commitment and maintenance. Immunity 30:616-625. 5. Sakaguchi, S., Miyara, M., Costantino, C.M., and Hafler, D.A. 2010. FOXP3+ regulatory T cells in the human immune system. Nature reviews. Immunology 10:490-500. 6. Josefowicz, S.Z., Lu, L.F., and Rudensky, A.Y. 2012. Regulatory T Cells: Mechanisms of Differentiation and Function. Annual review of immunology. 7. Allan, S.E., Broady, R., Gregori, S., Himmel, M.E., Locke, N., Roncarolo, M.G., Bacchetta, R., and Levings, M.K. 2008. CD4+ T-regulatory cells: toward therapy for human diseases. Immunol Rev 223:391-421. 8. Levings, M.K., Sangregorio, R., and Roncarolo, M.G. 2001. Human cd25(+)cd4(+) t regulatory cells suppress naive and memory T cell proliferation and can be expanded in vitro without loss of function. J Exp Med 193:1295-1302. 9. Fontenot, J.D., Dooley, J.L., Farr, A.G., and Rudensky, A.Y. 2005. Developmental regulation of Foxp3 expression during ontogeny. J Exp Med 202:901-906. 10. Shevach, E.M. 2009. Mechanisms of foxp3+ T regulatory cell-mediated suppression. Immunity 30:636-645. 11. Tran, D.Q., Andersson, J., Hardwick, D., Bebris, L., Illei, G.G., and Shevach, E.M. 2009. Selective expression of latency-associated peptide (LAP) and IL-1 receptor type I/II (CD121a/CD121b) on activated human FOXP3+ regulatory T cells allows for their purification from expansion cultures. Blood 113:5125-5133. 12. Bacchetta, R., Passerini, L., Gambineri, E., Dai, M., Allan, S.E., Perroni, L., Dagna- Bricarelli, F., Sartirana, C., Matthes-Martin, S., Lawitschka, A., et al. 2006. Defective regulatory and effector T cell functions in patients with FOXP3 mutations. J Clin Invest 116:1713-1722. 13. Gambineri, E., Perroni, L., Passerini, L., Bianchi, L., Doglioni, C., Meschi, F., Bonfanti, R., Sznajer, Y., Tommasini, A., Lawitschka, A., et al. 2008. Clinical and molecular profile of a new series of patients with immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome: inconsistent correlation between forkhead box protein 3 expression and disease severity. J Allergy Clin Immunol 122:1105-1112 e1101. 14. Allan, S.E., Passerini, L., Bacchetta, R., Crellin, N., Dai, M., Orban, P.C., Ziegler, S.F., Roncarolo, M.G., and Levings, M.K. 2005. The role of 2 FOXP3 isoforms in the generation of human CD4+ Tregs. J Clin Invest 115:3276-3284.  105 15. Lopes, J.E., Torgerson, T.R., Schubert, L.A., Anover, S.D., Ocheltree, E.L., Ochs, H.D., and Ziegler, S.F. 2006. Analysis of FOXP3 reveals multiple domains required for its function as a transcriptional repressor. J Immunol 177:3133-3142. 16. Zhou, L., Lopes, J.E., Chong, M.M., Ivanov, II, Min, R., Victora, G.D., Shen, Y., Du, J., Rubtsov, Y.P., Rudensky, A.Y., et al. 2008. TGF-beta-induced Foxp3 inhibits T(H)17 cell differentiation by antagonizing RORgammat function. Nature 453:236- 240. 17. Allan, S.E., Crome, S.Q., Crellin, N.K., Passerini, L., Steiner, T.S., Bacchetta, R., Roncarolo, M.G., and Levings, M.K. 2007. Activation-induced FOXP3 in human T effector cells does not suppress proliferation or cytokine production. Int Immunol 19:345-354. 18. Gavin, M.A., Torgerson, T.R., Houston, E., DeRoos, P., Ho, W.Y., Stray-Pedersen, A., Ocheltree, E.L., Greenberg, P.D., Ochs, H.D., and Rudensky, A.Y. 2006. Single- cell analysis of normal and FOXP3-mutant human T cells: FOXP3 expression without regulatory T cell development. Proc Natl Acad Sci U S A 103:6659-6664. 19. Tran, D.Q., Ramsey, H., and Shevach, E.M. 2007. Induction of FOXP3 expression in naive human CD4+FOXP3 T cells by T-cell receptor stimulation is transforming growth factor-beta dependent but does not confer a regulatory phenotype. Blood 110:2983-2990. 20. Allan, S.E., Alstad, A.N., Merindol, N., Crellin, N.K., Amendola, M., Bacchetta, R., Naldini, L., Roncarolo, M.G., Soudeyns, H., and Levings, M.K. 2008. Generation of potent and stable human CD4+ T regulatory cells by activation-independent expression of FOXP3. Mol Ther 16:194-202. 21. Elyaman, W., Bradshaw, E.M., Uyttenhove, C., Dardalhon, V., Awasthi, A., Imitola, J., Bettelli, E., Oukka, M., van Snick, J., Renauld, J.C., et al. 2009. IL-9 induces differentiation of TH17 cells and enhances function of FoxP3+ natural regulatory T cells. Proc Natl Acad Sci U S A. 22. Miyara, M., Yoshioka, Y., Kitoh, A., Shima, T., Wing, K., Niwa, A., Parizot, C., Taflin, C., Heike, T., Valeyre, D., et al. 2009. Functional Delineation and Differentiation Dynamics of Human CD4(+) T Cells Expressing the FoxP3 Transcription Factor. Immunity. 23. Voo, K.S., Wang, Y.H., Santori, F.R., Boggiano, C., Arima, K., Bover, L., Hanabuchi, S., Khalili, J., Marinova, E., Zheng, B., et al. 2009. Identification of IL- 17-producing FOXP3+ regulatory T cells in humans. Proc Natl Acad Sci U S A 106:4793-4798. 24. Roncarolo, M.G., Gregori, S., Battaglia, M., Bacchetta, R., Fleischhauer, K., and Levings, M.K. 2006. Interleukin-10-secreting type 1 regulatory T cells in rodents and humans. Immunol Rev 212:28-50. 25. Zou, W. 2006. Regulatory T cells, tumour immunity and immunotherapy. Nat Rev Immunol 6:295-307. 26. Mucida, D., Park, Y., Kim, G., Turovskaya, O., Scott, I., Kronenberg, M., and Cheroutre, H. 2007. Reciprocal TH17 and regulatory T cell differentiation mediated by retinoic acid. Science 317:256-260. 27. Travis, M.A., Reizis, B., Melton, A.C., Masteller, E., Tang, Q., Proctor, J.M., Wang, Y., Bernstein, X., Huang, X., Reichardt, L.F., et al. 2007. Loss of integrin  106 alpha(v)beta8 on dendritic cells causes autoimmunity and colitis in mice. Nature 449:361-365. 28. Matsumura, Y., Kobayashi, T., Ichiyama, K., Yoshida, R., Hashimoto, M., Takimoto, T., Tanaka, K., Chinen, T., Shichita, T., Wyss-Coray, T., et al. 2007. Selective expansion of foxp3-positive regulatory T cells and immunosuppression by suppressors of cytokine signaling 3-deficient dendritic cells. J Immunol 179:2170- 2179. 29. Haxhinasto, S., Mathis, D., and Benoist, C. 2008. The AKT-mTOR axis regulates de novo differentiation of CD4+Foxp3+ cells. J Exp Med 205:565-574. 30. Battaglia, M., Stabilini, A., Migliavacca, B., Horejs-Hoeck, J., Kaupper, T., and Roncarolo, M.G. 2006. Rapamycin promotes expansion of functional CD4+CD25+FOXP3+ regulatory T cells of both healthy subjects and type 1 diabetic patients. J Immunol 177:8338-8347. 31. Sauer, S., Bruno, L., Hertweck, A., Finlay, D., Leleu, M., Spivakov, M., Knight, Z.A., Cobb, B.S., Cantrell, D., O'Connor, E., et al. 2008. T cell receptor signaling controls Foxp3 expression via PI3K, Akt, and mTOR. Proc Natl Acad Sci U S A 105:7797-7802. 32. Allan, S.E., Crome, S.Q., Crellin, N.K., Passerini, L., Steiner, T.S., Bacchetta, R., Roncarolo, M.G., and Levings, M.K. 2007. Activation-induced FOXP3 in human T effector cells does not suppress proliferation or cytokine production. International immunology 19:345-354. 33. Fritzsching, B., Oberle, N., Pauly, E., Geffers, R., Buer, J., Poschl, J., Krammer, P., Linderkamp, O., and Suri-Payer, E. 2006. Naive regulatory T cells: a novel subpopulation defined by resistance toward CD95L-mediated cell death. Blood 108:3371-3378. 34. Darrasse-Jeze, G., Marodon, G., Salomon, B.L., Catala, M., and Klatzmann, D. 2005. Ontogeny of CD4+CD25+ regulatory/suppressor T cells in human fetuses. Blood 105:4715-4721. 35. Miyara, M., Yoshioka, Y., Kitoh, A., Shima, T., Wing, K., Niwa, A., Parizot, C., Taflin, C., Heike, T., Valeyre, D., et al. 2009. Functional delineation and differentiation dynamics of human CD4+ T cells expressing the FoxP3 transcription factor. Immunity 30:899-911. 36. Seddiki, N., Santner-Nanan, B., Tangye, S.G., Alexander, S.I., Solomon, M., Lee, S., Nanan, R., and Fazekas de Saint Groth, B. 2006. Persistence of naive CD45RA+ regulatory T cells in adult life. Blood 107:2830-2838. 37. Haas, J., Fritzsching, B., Trubswetter, P., Korporal, M., Milkova, L., Fritz, B., Vobis, D., Krammer, P.H., Suri-Payer, E., and Wildemann, B. 2007. Prevalence of newly generated naive regulatory T cells (Treg) is critical for Treg suppressive function and determines Treg dysfunction in multiple sclerosis. Journal of immunology 179:1322- 1330. 38. Venken, K., Hellings, N., Broekmans, T., Hensen, K., Rummens, J.L., and Stinissen, P. 2008. Natural naive CD4+CD25+CD127low regulatory T cell (Treg) development and function are disturbed in multiple sclerosis patients: recovery of memory Treg homeostasis during disease progression. Journal of immunology 180:6411-6420. 39. Vukmanovic-Stejic, M., Zhang, Y., Cook, J.E., Fletcher, J.M., McQuaid, A., Masters, J.E., Rustin, M.H., Taams, L.S., Beverley, P.C., Macallan, D.C., et al. 2006. Human  107 CD4+ CD25hi Foxp3+ regulatory T cells are derived by rapid turnover of memory populations in vivo. The Journal of clinical investigation 116:2423-2433. 40. Baecher-Allan, C., Viglietta, V., and Hafler, D.A. 2002. Inhibition of human CD4(+)CD25(+high) regulatory T cell function. Journal of immunology 169:6210- 6217. 41. Rosenblum, M.D., Gratz, I.K., Paw, J.S., Lee, K., Marshak-Rothstein, A., and Abbas, A.K. 2011. Response to self antigen imprints regulatory memory in tissues. Nature 480:538-542. 42. Beriou, G., Costantino, C.M., Ashley, C.W., Yang, L., Kuchroo, V.K., Baecher- Allan, C., and Hafler, D.A. 2009. IL-17-producing human peripheral regulatory T cells retain suppressive function. Blood 113:4240-4249. 43. Chen, X., and Oppenheim, J.J. 2011. Resolving the identity myth: key markers of functional CD4+FoxP3+ regulatory T cells. International immunopharmacology 11:1489-1496. 44. Ziegler, S.F. 2007. FOXP3: not just for regulatory T cells anymore. European journal of immunology 37:21-23. 45. Miyara, M., and Sakaguchi, S. 2011. Human FoxP3(+)CD4(+) regulatory T cells: their knowns and unknowns. Immunol Cell Biol 89:346-351. 46. Miyao, T., Floess, S., Setoguchi, R., Luche, H., Fehling, H.J., Waldmann, H., Huehn, J., and Hori, S. 2012. Plasticity of foxp3(+) T cells reflects promiscuous foxp3 expression in conventional T cells but not reprogramming of regulatory T cells. Immunity 36:262-275. 47. Onishi, Y., Fehervari, Z., Yamaguchi, T., and Sakaguchi, S. 2008. Foxp3+ natural regulatory T cells preferentially form aggregates on dendritic cells in vitro and actively inhibit their maturation. Proceedings of the National Academy of Sciences of the United States of America 105:10113-10118. 48. Jonuleit, H., Schmitt, E., Stassen, M., Tuettenberg, A., Knop, J., and Enk, A.H. 2001. Identification and functional characterization of human CD4(+)CD25(+) T cells with regulatory properties isolated from peripheral blood. The Journal of experimental medicine 193:1285-1294. 49. Solstad, T., Bains, S.J., Landskron, J., Aandahl, E.M., Thiede, B., Tasken, K., and Torgersen, K.M. 2011. CD147 (Basigin/Emmprin) identifies FoxP3+CD45RO+CTLA4+-activated human regulatory T cells. Blood 118:5141- 5151. 50. Liu, W., Putnam, A.L., Xu-Yu, Z., Szot, G.L., Lee, M.R., Zhu, S., Gottlieb, P.A., Kapranov, P., Gingeras, T.R., Fazekas de St Groth, B., et al. 2006. CD127 expression inversely correlates with FoxP3 and suppressive function of human CD4+ T reg cells. The Journal of experimental medicine 203:1701-1711. 51. Michel, L., Berthelot, L., Pettre, S., Wiertlewski, S., Lefrere, F., Braudeau, C., Brouard, S., Soulillou, J.P., and Laplaud, D.A. 2008. Patients with relapsing-remitting multiple sclerosis have normal Treg function when cells expressing IL-7 receptor alpha-chain are excluded from the analysis. The Journal of clinical investigation 118:3411-3419. 52. Mazzucchelli, R., and Durum, S.K. 2007. Interleukin-7 receptor expression: intelligent design. Nature reviews. Immunology 7:144-154.  108 53. Alam, M.S., Kurtz, C.C., Rowlett, R.M., Reuter, B.K., Wiznerowicz, E., Das, S., Linden, J., Crowe, S.E., and Ernst, P.B. 2009. CD73 is expressed by human regulatory T helper cells and suppresses proinflammatory cytokine production and Helicobacter felis-induced gastritis in mice. The Journal of infectious diseases 199:494-504. 54. Dwyer, K.M., Hanidziar, D., Putheti, P., Hill, P.A., Pommey, S., McRae, J.L., Winterhalter, A., Doherty, G., Deaglio, S., Koulmanda, M., et al. 2010. Expression of CD39 by human peripheral blood CD4+ CD25+ T cells denotes a regulatory memory phenotype. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 10:2410-2420. 55. Fletcher, J.M., Lonergan, R., Costelloe, L., Kinsella, K., Moran, B., O'Farrelly, C., Tubridy, N., and Mills, K.H. 2009. CD39+Foxp3+ regulatory T Cells suppress pathogenic Th17 cells and are impaired in multiple sclerosis. Journal of immunology 183:7602-7610. 56. Borsellino, G., Kleinewietfeld, M., Di Mitri, D., Sternjak, A., Diamantini, A., Giometto, R., Hopner, S., Centonze, D., Bernardi, G., Dell'Acqua, M.L., et al. 2007. Expression of ectonucleotidase CD39 by Foxp3+ Treg cells: hydrolysis of extracellular ATP and immune suppression. Blood 110:1225-1232. 57. Chung, Y., Chang, S.H., Martinez, G.J., Yang, X.O., Nurieva, R., Kang, H.S., Ma, L., Watowich, S.S., Jetten, A.M., Tian, Q., et al. 2009. Critical regulation of early Th17 cell differentiation by interleukin-1 signaling. Immunity 30:576-587. 58. Stockis, J., Colau, D., Coulie, P.G., and Lucas, S. 2009. Membrane protein GARP is a receptor for latent TGF-beta on the surface of activated human Treg. European journal of immunology 39:3315-3322. 59. Wang, R., Kozhaya, L., Mercer, F., Khaitan, A., Fujii, H., and Unutmaz, D. 2009. Expression of GARP selectively identifies activated human FOXP3+ regulatory T cells. Proceedings of the National Academy of Sciences of the United States of America 106:13439-13444. 60. Gandhi, R., Farez, M.F., Wang, Y., Kozoriz, D., Quintana, F.J., and Weiner, H.L. 2010. Cutting edge: human latency-associated peptide+ T cells: a novel regulatory T cell subset. Journal of immunology 184:4620-4624. 61. Szczepanski, M.J., Szajnik, M., Czystowska, M., Mandapathil, M., Strauss, L., Welsh, A., Foon, K.A., Whiteside, T.L., and Boyiadzis, M. 2009. Increased frequency and suppression by regulatory T cells in patients with acute myelogenous leukemia. Clin Cancer Res 15:3325-3332. 62. Zhou, Q., Bucher, C., Munger, M.E., Highfill, S.L., Tolar, J., Munn, D.H., Levine, B.L., Riddle, M., June, C.H., Vallera, D.A., et al. 2009. Depletion of endogenous tumor-associated regulatory T cells improves the efficacy of adoptive cytotoxic T-cell immunotherapy in murine acute myeloid leukemia. Blood 114:3793-3802. 63. Battaglia, A., Buzzonetti, A., Monego, G., Peri, L., Ferrandina, G., Fanfani, F., Scambia, G., and Fattorossi, A. 2008. Neuropilin-1 expression identifies a subset of regulatory T cells in human lymph nodes that is modulated by preoperative chemoradiation therapy in cervical cancer. Immunology 123:129-138. 64. Ahmadzadeh, M., Felipe-Silva, A., Heemskerk, B., Powell, D.J., Jr., Wunderlich, J.R., Merino, M.J., and Rosenberg, S.A. 2008. FOXP3 expression accurately defines  109 the population of intratumoral regulatory T cells that selectively accumulate in metastatic melanoma lesions. Blood 112:4953-4960. 65. Huehn, J., Polansky, J.K., and Hamann, A. 2009. Epigenetic control of FOXP3 expression: the key to a stable regulatory T-cell lineage? Nat Rev Immunol 9:83-89. 66. Wieczorek, G., Asemissen, A., Model, F., Turbachova, I., Floess, S., Liebenberg, V., Baron, U., Stauch, D., Kotsch, K., Pratschke, J., et al. 2009. Quantitative DNA methylation analysis of FOXP3 as a new method for counting regulatory T cells in peripheral blood and solid tissue. Cancer Res 69:599-608. 67. Curiel, T.J., Coukos, G., Zou, L., Alvarez, X., Cheng, P., Mottram, P., Evdemon- Hogan, M., Conejo-Garcia, J.R., Zhang, L., Burow, M., et al. 2004. Specific recruitment of regulatory T cells in ovarian carcinoma fosters immune privilege and predicts reduced survival. Nat Med 10:942-949. 68. Leffers, N., Gooden, M.J., de Jong, R.A., Hoogeboom, B.N., ten Hoor, K.A., Hollema, H., Boezen, H.M., van der Zee, A.G., Daemen, T., and Nijman, H.W. 2009. Prognostic significance of tumor-infiltrating T-lymphocytes in primary and metastatic lesions of advanced stage ovarian cancer. Cancer Immunol Immunother 58:449-459. 69. Bluestone, J.A., Tang, Q., and Sedwick, C.E. 2008. T regulatory cells in autoimmune diabetes: past challenges, future prospects. J Clin Immunol 28:677-684. 70. Tarbell, K.V., Yamazaki, S., Olson, K., Toy, P., and Steinman, R.M. 2004. CD25+ CD4+ T cells, expanded with dendritic cells presenting a single autoantigenic peptide, suppress autoimmune diabetes. J Exp Med 199:1467-1477. 71. Tang, Q., Henriksen, K.J., Bi, M., Finger, E.B., Szot, G., Ye, J., Masteller, E.L., McDevitt, H., Bonyhadi, M., and Bluestone, J.A. 2004. In vitro-expanded antigen- specific regulatory T cells suppress autoimmune diabetes. J Exp Med 199:1455-1465. 72. Battaglia, M., Stabilini, A., Draghici, E., Migliavacca, B., Gregori, S., Bonifacio, E., and Roncarolo, M.G. 2006. Induction of tolerance in type 1 diabetes via both CD4+CD25+ T regulatory cells and T regulatory type 1 cells. Diabetes 55:1571- 1580. 73. Setoguchi, R., Hori, S., Takahashi, T., and Sakaguchi, S. 2005. Homeostatic maintenance of natural Foxp3(+) CD25(+) CD4(+) regulatory T cells by interleukin (IL)-2 and induction of autoimmune disease by IL-2 neutralization. J Exp Med 201:723-735. 74. Meagher, C., Tang, Q., Fife, B.T., Bour-Jordan, H., Wu, J., Pardoux, C., Bi, M., Melli, K., and Bluestone, J.A. 2008. Spontaneous development of a pancreatic exocrine disease in CD28-deficient NOD mice. J Immunol 180:7793-7803. 75. Marwaha, A.K., Crome, S.Q., Panagiotopoulos, C., Berg, K.B., Qin, H., Ouyang, Q., Xu, L., Priatel, J.J., Levings, M.K., and Tan, R. 2010. Cutting edge: Increased IL-17- secreting T cells in children with new-onset type 1 diabetes. J Immunol 185:3814- 3818. 76. McClymont, S.A., Putnam, A.L., Lee, M.R., Esensten, J.H., Liu, W., Hulme, M.A., Hoffmuller, U., Baron, U., Olek, S., Bluestone, J.A., et al. 2011. Plasticity of human regulatory T cells in healthy subjects and patients with type 1 diabetes. J Immunol 186:3918-3926. 77. Putnam, A.L., Brusko, T.M., Lee, M.R., Liu, W., Szot, G.L., Ghosh, T., Atkinson, M.A., and Bluestone, J.A. 2009. Expansion of human regulatory T-cells from patients with type 1 diabetes. Diabetes 58:652-662.  110 78. Brusko, T., Wasserfall, C., McGrail, K., Schatz, R., Viener, H.L., Schatz, D., Haller, M., Rockell, J., Gottlieb, P., Clare-Salzler, M., et al. 2007. No alterations in the frequency of FOXP3+ regulatory T-cells in type 1 diabetes. Diabetes 56:604-612. 79. Long, S.A., Cerosaletti, K., Bollyky, P.L., Tatum, M., Shilling, H., Zhang, S., Zhang, Z.Y., Pihoker, C., Sanda, S., Greenbaum, C., et al. 2010. Defects in IL-2R signaling contribute to diminished maintenance of FOXP3 expression in CD4(+)CD25(+) regulatory T-cells of type 1 diabetic subjects. Diabetes 59:407-415. 80. Marwaha, A.K., Crome, S.Q., Panagiotopoulos, C., Berg, K.B., Qin, H., Ouyang, Q., Xu, L., Priatel, J.J., Levings, M.K., and Tan, R. 2010. Cutting edge: Increased IL-17- secreting T cells in children with new-onset type 1 diabetes. Journal of immunology 185:3814-3818. 81. Willcox, A., Richardson, S.J., Bone, A.J., Foulis, A.K., and Morgan, N.G. 2009. Analysis of islet inflammation in human type 1 diabetes. Clinical and experimental immunology 155:173-181. 82. Tang, Q., Adams, J.Y., Penaranda, C., Melli, K., Piaggio, E., Sgouroudis, E., Piccirillo, C.A., Salomon, B.L., and Bluestone, J.A. 2008. Central role of defective interleukin-2 production in the triggering of islet autoimmune destruction. Immunity 28:687-697. 83. Shevach, E.M., Davidson, T.S., Huter, E.N., Dipaolo, R.A., and Andersson, J. 2008. Role of TGF-Beta in the induction of Foxp3 expression and T regulatory cell function. J Clin Immunol 28:640-646. 84. Li, M.O., Wan, Y.Y., Sanjabi, S., Robertson, A.K., and Flavell, R.A. 2006. Transforming growth factor-beta regulation of immune responses. Annu Rev Immunol 24:99-146. 85. Collison, L.W., Workman, C.J., Kuo, T.T., Boyd, K., Wang, Y., Vignali, K.M., Cross, R., Sehy, D., Blumberg, R.S., and Vignali, D.A. 2007. The inhibitory cytokine IL-35 contributes to regulatory T-cell function. Nature 450:566-569. 86. Collison, L.W., Pillai, M.R., Chaturvedi, V., and Vignali, D.A. 2009. Regulatory T cell suppression is potentiated by target T cells in a cell contact, IL-35- and IL-10- dependent manner. J Immunol 182:6121-6128. 87. Bardel, E., Larousserie, F., Charlot-Rabiega, P., Coulomb-L'Hermine, A., and Devergne, O. 2008. Human CD4+ CD25+ Foxp3+ regulatory T cells do not constitutively express IL-35. J Immunol 181:6898-6905. 88. Chaturvedi, V., Collison, L.W., Guy, C.S., Workman, C.J., and Vignali, D.A. 2011. Cutting edge: Human regulatory T cells require IL-35 to mediate suppression and infectious tolerance. Journal of immunology 186:6661-6666. 89. Grossman, W.J., Verbsky, J.W., Barchet, W., Colonna, M., Atkinson, J.P., and Ley, T.J. 2004. Human T regulatory cells can use the perforin pathway to cause autologous target cell death. Immunity 21:589-601. 90. Gondek, D.C., Lu, L.F., Quezada, S.A., Sakaguchi, S., and Noelle, R.J. 2005. Cutting edge: contact-mediated suppression by CD4+CD25+ regulatory cells involves a granzyme B-dependent, perforin-independent mechanism. J Immunol 174:1783-1786. 91. Zhao, D.M., Thornton, A.M., DiPaolo, R.J., and Shevach, E.M. 2006. Activated CD4+CD25+ T cells selectively kill B lymphocytes. Blood 107:3925-3932. 92. Strauss, L., Bergmann, C., Szczepanski, M., Gooding, W., Johnson, J.T., and Whiteside, T.L. 2007. A unique subset of CD4+CD25highFoxp3+ T cells secreting  111 interleukin-10 and transforming growth factor-beta1 mediates suppression in the tumor microenvironment. Clin Cancer Res 13:4345-4354. 93. Strauss, L., Bergmann, C., and Whiteside, T.L. 2009. Human circulating CD4+CD25highFoxp3+ regulatory T cells kill autologous CD8+ but not CD4+ responder cells by Fas-mediated apoptosis. J Immunol 182:1469-1480. 94. Cao, X., Cai, S.F., Fehniger, T.A., Song, J., Collins, L.I., Piwnica-Worms, D.R., and Ley, T.J. 2007. Granzyme B and perforin are important for regulatory T cell-mediated suppression of tumor clearance. Immunity 27:635-646. 95. Thornton, A.M., and Shevach, E.M. 1998. CD4+CD25+ immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production. J Exp Med 188:287-296. 96. Pandiyan, P., Zheng, L., Ishihara, S., Reed, J., and Lenardo, M.J. 2007. CD4+CD25+Foxp3+ regulatory T cells induce cytokine deprivation-mediated apoptosis of effector CD4+ T cells. Nat Immunol 8:1353-1362. 97. Oberle, N., Eberhardt, N., Falk, C.S., Krammer, P.H., and Suri-Payer, E. 2007. Rapid suppression of cytokine transcription in human CD4+CD25 T cells by CD4+Foxp3+ regulatory T cells: independence of IL-2 consumption, TGF-beta, and various inhibitors of TCR signaling. J Immunol 179:3578-3587. 98. Leveque, L., Deknuydt, F., Bioley, G., Old, L.J., Matsuzaki, J., Odunsi, K., Ayyoub, M., and Valmori, D. 2009. Interleukin 2-mediated conversion of ovarian cancer- associated CD4+ regulatory T cells into proinflammatory interleukin 17-producing helper T cells. J Immunother 32:101-108. 99. Bopp, T., Becker, C., Klein, M., Klein-Hessling, S., Palmetshofer, A., Serfling, E., Heib, V., Becker, M., Kubach, J., Schmitt, S., et al. 2007. Cyclic adenosine monophosphate is a key component of regulatory T cell-mediated suppression. J Exp Med 204:1303-1310. 100. Deaglio, S., Dwyer, K.M., Gao, W., Friedman, D., Usheva, A., Erat, A., Chen, J.F., Enjyoji, K., Linden, J., Oukka, M., et al. 2007. Adenosine generation catalyzed by CD39 and CD73 expressed on regulatory T cells mediates immune suppression. J Exp Med 204:1257-1265. 101. Zarek, P.E., Huang, C.T., Lutz, E.R., Kowalski, J., Horton, M.R., Linden, J., Drake, C.G., and Powell, J.D. 2008. A2A receptor signaling promotes peripheral tolerance by inducing T-cell anergy and the generation of adaptive regulatory T cells. Blood 111:251-259. 102. Ohta, A., Gorelik, E., Prasad, S.J., Ronchese, F., Lukashev, D., Wong, M.K., Huang, X., Caldwell, S., Liu, K., Smith, P., et al. 2006. A2A adenosine receptor protects tumors from antitumor T cells. Proc Natl Acad Sci U S A 103:13132-13137. 103. Sitkovsky, M.V., Kjaergaard, J., Lukashev, D., and Ohta, A. 2008. Hypoxia- adenosinergic immunosuppression: tumor protection by T regulatory cells and cancerous tissue hypoxia. Clin Cancer Res 14:5947-5952. 104. Salcido-Ochoa, F., Tsang, J., Tam, P., Falk, K., and Rotzschke, O. 2010. Regulatory T cells in transplantation: does extracellular adenosine triphosphate metabolism through CD39 play a crucial role? Transplantation reviews 24:52-66. 105. Misra, N., Bayry, J., Lacroix-Desmazes, S., Kazatchkine, M.D., and Kaveri, S.V. 2004. Cutting edge: human CD4+CD25+ T cells restrain the maturation and antigen- presenting function of dendritic cells. J Immunol 172:4676-4680.  112 106. Onishi, Y., Fehervari, Z., Yamaguchi, T., and Sakaguchi, S. 2008. Foxp3+ natural regulatory T cells preferentially form aggregates on dendritic cells in vitro and actively inhibit their maturation. Proc Natl Acad Sci U S A 105:10113-10118. 107. Veldhoen, M., Moncrieffe, H., Hocking, R.J., Atkins, C.J., and Stockinger, B. 2006. Modulation of dendritic cell function by naive and regulatory CD4+ T cells. J Immunol 176:6202-6210. 108. Takahashi, T., Tagami, T., Yamazaki, S., Uede, T., Shimizu, J., Sakaguchi, N., Mak, T.W., and Sakaguchi, S. 2000. Immunologic self-tolerance maintained by CD25(+)CD4(+) regulatory T cells constitutively expressing cytotoxic T lymphocyte- associated antigen 4. J Exp Med 192:303-310. 109. Wing, K., Onishi, Y., Prieto-Martin, P., Yamaguchi, T., Miyara, M., Fehervari, Z., Nomura, T., and Sakaguchi, S. 2008. CTLA-4 control over Foxp3+ regulatory T cell function. Science 322:271-275. 110. Mellor, A.L., and Munn, D.H. 2004. IDO expression by dendritic cells: tolerance and tryptophan catabolism. Nat Rev Immunol 4:762-774. 111. Woo, S.R., Li, N., Bruno, T.C., Forbes, K., Brown, S., Workman, C., Drake, C.G., and Vignali, D.A. 2010. Differential subcellular localization of the regulatory T-cell protein LAG-3 and the coreceptor CD4. European journal of immunology 40:1768- 1777. 112. Camisaschi, C., Casati, C., Rini, F., Perego, M., De Filippo, A., Triebel, F., Parmiani, G., Belli, F., Rivoltini, L., and Castelli, C. 2010. LAG-3 expression defines a subset of CD4(+)CD25(high)Foxp3(+) regulatory T cells that are expanded at tumor sites. Journal of immunology 184:6545-6551. 113. Kryczek, I., Wei, S., Zhu, G., Myers, L., Mottram, P., Cheng, P., Chen, L., Coukos, G., and Zou, W. 2007. Relationship between B7-H4, regulatory T cells, and patient outcome in human ovarian carcinoma. Cancer Res 67:8900-8905. 114. Zang, X., Loke, P., Kim, J., Murphy, K., Waitz, R., and Allison, J.P. 2003. B7x: a widely expressed B7 family member that inhibits T cell activation. Proc Natl Acad Sci U S A 100:10388-10392. 115. Kryczek, I., Zou, L., Rodriguez, P., Zhu, G., Wei, S., Mottram, P., Brumlik, M., Cheng, P., Curiel, T., Myers, L., et al. 2006. B7-H4 expression identifies a novel suppressive macrophage population in human ovarian carcinoma. J Exp Med 203:871-881. 116. Wang, X., Hao, J., Metzger, D.L., Mui, A., Ao, Z., Akhoundsadegh, N., Langermann, S., Liu, L., Chen, L., Ou, D., et al. 2011. Early treatment of NOD mice with B7-H4 reduces the incidence of autoimmune diabetes. Diabetes 60:3246-3255. 117. Chen, L. 2004. Co-inhibitory molecules of the B7-CD28 family in the control of T- cell immunity. Nat Rev Immunol 4:336-347. 118. Keir, M.E., Francisco, L.M., and Sharpe, A.H. 2007. PD-1 and its ligands in T-cell immunity. Curr Opin Immunol 19:309-314. 119. Wang, W., Lau, R., Yu, D., Zhu, W., Korman, A., and Weber, J. 2009. PD1 blockade reverses the suppression of melanoma antigen-specific CTL by CD4+ CD25(Hi) regulatory T cells. Int Immunol 21:1065-1077. 120. Francisco, L.M., Salinas, V.H., Brown, K.E., Vanguri, V.K., Freeman, G.J., Kuchroo, V.K., and Sharpe, A.H. 2009. PD-L1 regulates the development, maintenance, and function of induced regulatory T cells. J Exp Med 206:3015-3029.  113 121. Franceschini, D., Paroli, M., Francavilla, V., Videtta, M., Morrone, S., Labbadia, G., Cerino, A., Mondelli, M.U., and Barnaba, V. 2009. PD-L1 negatively regulates CD4+CD25+Foxp3+ Tregs by limiting STAT-5 phosphorylation in patients chronically infected with HCV. The Journal of clinical investigation 119:551-564. 122. Scheid, E., Ricci, M, and Foley, M. 2010. Dendritic Cell Based Cancer Vaccines: Practical Considerations. In Experimental and Applied Immunotherapy J. Medin, and Fowler, D, editor. 123. Nayak, S., and Herzog, R.W. 2009. Progress and prospects: immune responses to viral vectors. Gene Ther. 124. Fontana, R., Bregni, M., Cipponi, A., Raccosta, L., Rainelli, C., Maggioni, D., Lunghi, F., Ciceri, F., Mukenge, S., Doglioni, C., et al. 2009. Peripheral blood lymphocytes genetically modified to express the self/tumor antigen MAGE-A3 induce antitumor immune responses in cancer patients. Blood 113:1651-1660. 125. Nakamura, M., Iwahashi, M., Nakamori, M., Ueda, K., Ojima, T., Naka, T., Ishida, K., and Yamaue, H. 2005. Dendritic cells transduced with tumor-associated antigen gene elicit potent therapeutic antitumor immunity: comparison with immunodominant peptide-pulsed DCs. Oncology 68:163-170. 126. Butterfield, L.H., Comin-Anduix, B., Vujanovic, L., Lee, Y., Dissette, V.B., Yang, J.Q., Vu, H.T., Seja, E., Oseguera, D.K., Potter, D.M., et al. 2008. Adenovirus MART-1-engineered autologous dendritic cell vaccine for metastatic melanoma. J Immunother 31:294-309. 127. Tsao, H., Millman, P., Linette, G.P., Hodi, F.S., Sober, A.J., Goldberg, M.A., and Haluska, F.G. 2002. Hypopigmentation associated with an adenovirus-mediated gp100/MART-1-transduced dendritic cell vaccine for metastatic melanoma. Arch Dermatol 138:799-802. 128. Rosenberg, S.A., Yang, J.C., Schwartzentruber, D.J., Hwu, P., Marincola, F.M., Topalian, S.L., Restifo, N.P., Dudley, M.E., Schwarz, S.L., Spiess, P.J., et al. 1998. Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma. Nat Med 4:321-327. 129. Schadendorf, D., Ugurel, S., Schuler-Thurner, B., Nestle, F.O., Enk, A., Brocker, E.B., Grabbe, S., Rittgen, W., Edler, L., Sucker, A., et al. 2006. Dacarbazine (DTIC) versus vaccination with autologous peptide-pulsed dendritic cells (DC) in first-line treatment of patients with metastatic melanoma: a randomized phase III trial of the DC study group of the DeCOG. Ann Oncol 17:563-570. 130. Whiteside, T.L. Inhibiting the inhibitors: evaluating agents targeting cancer immunosuppression. Expert Opin Biol Ther. 131. Dumitriu, I.E., Dunbar, D.R., Howie, S.E., Sethi, T., and Gregory, C.D. 2009. Human dendritic cells produce TGF-beta 1 under the influence of lung carcinoma cells and prime the differentiation of CD4+CD25+Foxp3+ regulatory T cells. J Immunol 182:2795-2807. 132. Vasir, B., Wu, Z., Crawford, K., Rosenblatt, J., Zarwan, C., Bissonnette, A., Kufe, D., and Avigan, D. 2008. Fusions of dendritic cells with breast carcinoma stimulate the expansion of regulatory T cells while concomitant exposure to IL-12, CpG oligodeoxynucleotides, and anti-CD3/CD28 promotes the expansion of activated tumor reactive cells. J Immunol 181:808-821.  114 133. Dunn, G.P., Old, L.J., and Schreiber, R.D. 2004. The immunobiology of cancer immunosurveillance and immunoediting. Immunity 21:137-148. 134. Sakaguchi, S. 2005. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in immunological tolerance to self and non-self. Nat Immunol 6:345-352. 135. Bayry, J., Triebel, F., Kaveri, S.V., and Tough, D.F. 2007. Human dendritic cells acquire a semimature phenotype and lymph node homing potential through interaction with CD4+CD25+ regulatory T cells. J Immunol 178:4184-4193. 136. Liang, B., Workman, C., Lee, J., Chew, C., Dale, B.M., Colonna, L., Flores, M., Li, N., Schweighoffer, E., Greenberg, S., et al. 2008. Regulatory T cells inhibit dendritic cells by lymphocyte activation gene-3 engagement of MHC class II. J Immunol 180:5916-5926. 137. Amarnath, S., Costanzo, C.M., Mariotti, J., Ullman, J.L., Telford, W.G., Kapoor, V., Riley, J.L., Levine, B.L., June, C.H., Fong, T., et al. Regulatory T cells and human myeloid dendritic cells promote tolerance via programmed death ligand-1. PLoS Biol 8:e1000302. 138. Benchetrit, F., Ciree, A., Vives, V., Warnier, G., Gey, A., Sautes-Fridman, C., Fossiez, F., Haicheur, N., Fridman, W.H., and Tartour, E. 2002. Interleukin-17 inhibits tumor cell growth by means of a T-cell-dependent mechanism. Blood 99:2114-2121. 139. Murugaiyan, G., and Saha, B. 2009. Protumor vs antitumor functions of IL-17. J Immunol 183:4169-4175. 140. Weaver, C.T., and Hatton, R.D. 2009. Interplay between the TH17 and TReg cell lineages: a (co-)evolutionary perspective. Nat Rev Immunol 9:883-889. 141. Koenen, H.J., Smeets, R.L., Vink, P.M., van Rijssen, E., Boots, A.M., and Joosten, I. 2008. Human CD25highFoxp3pos regulatory T cells differentiate into IL-17- producing cells. Blood 112:2340-2352. 142. Ng, P., Parks, R.J., Cummings, D.T., Evelegh, C.M., and Graham, F.L. 2000. An enhanced system for construction of adenoviral vectors by the two-plasmid rescue method. Hum Gene Ther 11:693-699. 143. Yang, T.C., Millar, J., Groves, T., Grinshtein, N., Parsons, R., Takenaka, S., Wan, Y., and Bramson, J.L. 2006. The CD8+ T cell population elicited by recombinant adenovirus displays a novel partially exhausted phenotype associated with prolonged antigen presentation that nonetheless provides long-term immunity. J Immunol 176:200-210. 144. Boruchov, A.M., Heller, G., Veri, M.C., Bonvini, E., Ravetch, J.V., and Young, J.W. 2005. Activating and inhibitory IgG Fc receptors on human DCs mediate opposing functions. The Journal of clinical investigation 115:2914-2923. 145. Miyara, M., and Sakaguchi, S. 2007. Natural regulatory T cells: mechanisms of suppression. Trends Mol Med 13:108-116. 146. Adema, G.J. 2009. Dendritic cells from bench to bedside and back. Immunol Lett 122:128-130. 147. Zhong, L., Granelli-Piperno, A., Choi, Y., and Steinman, R.M. 1999. Recombinant adenovirus is an efficient and non-perturbing genetic vector for human dendritic cells. Eur J Immunol 29:964-972. 148. Diao, J., Smythe, J.A., Smyth, C., Rowe, P.B., and Alexander, I.E. 1999. Human PBMC-derived dendritic cells transduced with an adenovirus vectorinduce cytotoxic  115 T-lymphocyte responses against a vector-encoded antigen in vitro. Gene Ther 6:845- 853. 149. Jenne, L., Schuler, G., and Steinkasserer, A. 2001. Viral vectors for dendritic cell- based immunotherapy. Trends Immunol 22:102-107. 150. Crome, S.Q., Wang, A.Y., Kang, C.Y., and Levings, M.K. 2009. The role of retinoic acid-related orphan receptor variant 2 and IL-17 in the development and function of human CD4+ T cells. Eur J Immunol 39:1480-1493. 151. Antonysamy, M.A., Fanslow, W.C., Fu, F., Li, W., Qian, S., Troutt, A.B., and Thomson, A.W. 1999. Evidence for a role of IL-17 in organ allograft rejection: IL-17 promotes the functional differentiation of dendritic cell progenitors. J Immunol 162:577-584. 152. Pan, J., Zhang, M., Wang, J., Wang, Q., Xia, D., Sun, W., Zhang, L., Yu, H., Liu, Y., and Cao, X. 2004. Interferon-gamma is an autocrine mediator for dendritic cell maturation. Immunol Lett 94:141-151. 153. Wilson, N.J., Boniface, K., Chan, J.R., McKenzie, B.S., Blumenschein, W.M., Mattson, J.D., Basham, B., Smith, K., Chen, T., Morel, F., et al. 2007. Development, cytokine profile and function of human interleukin 17-producing helper T cells. Nat Immunol 8:950-957. 154. Annunziato, F., Cosmi, L., Santarlasci, V., Maggi, L., Liotta, F., Mazzinghi, B., Parente, E., Fili, L., Ferri, S., Frosali, F., et al. 2007. Phenotypic and functional features of human Th17 cells. J Exp Med 204:1849-1861. 155. Awasthi, A., Riol-Blanco, L., Jager, A., Korn, T., Pot, C., Galileos, G., Bettelli, E., Kuchroo, V.K., and Oukka, M. 2009. Cutting edge: IL-23 receptor gfp reporter mice reveal distinct populations of IL-17-producing cells. J Immunol 182:5904-5908. 156. Acosta-Rodriguez, E.V., Napolitani, G., Lanzavecchia, A., and Sallusto, F. 2007. Interleukins 1beta and 6 but not transforming growth factor-beta are essential for the differentiation of interleukin 17-producing human T helper cells. Nat Immunol 8:942- 949. 157. Kelleher, P., and Knight, S.C. 1998. IL-12 increases CD80 expression and the stimulatory capacity of bone marrow-derived dendritic cells. Int Immunol 10:749- 755. 158. Lyakh, L.A., Koski, G.K., Young, H.A., Spence, S.E., Cohen, P.A., and Rice, N.R. 2002. Adenovirus type 5 vectors induce dendritic cell differentiation in human CD14(+) monocytes cultured under serum-free conditions. Blood 99:600-608. 159. Yang, X.O., Panopoulos, A.D., Nurieva, R., Chang, S.H., Wang, D., Watowich, S.S., and Dong, C. 2007. STAT3 regulates cytokine-mediated generation of inflammatory helper T cells. J Biol Chem 282:9358-9363. 160. de Martel, C., Plummer, M., Parsonnet, J., van Doorn, L.J., and Franceschi, S. 2009. Helicobacter species in cancers of the gallbladder and extrahepatic biliary tract. Br J Cancer 100:194-199. 161. Maes, W., Rosas, G.G., Verbinnen, B., Boon, L., De Vleeschouwer, S., Ceuppens, J.L., and Van Gool, S.W. 2009. DC vaccination with anti-CD25 treatment leads to long-term immunity against experimental glioma. Neuro Oncol 11:529-542. 162. Ding, Y., Xu, J., and Bromberg, J.S. 2012. Regulatory T cell migration during an immune response. Trends Immunol 33:174-180.  116 163. Lim, H.W., Hillsamer, P., and Kim, C.H. 2004. Regulatory T cells can migrate to follicles upon T cell activation and suppress GC-Th cells and GC-Th cell-driven B cell responses. The Journal of clinical investigation 114:1640-1649. 164. Lu, S.Y., Liu, K.Y., Liu, D.H., Xu, L.P., and Huang, X.J. 2011. High frequencies of CD62L naive regulatory T cells in allografts are associated with a low risk of acute graft-versus-host disease following unmanipulated allogeneic haematopoietic stem cell transplantation. Clinical and experimental immunology 165:264-277. 165. Gobert, M., Treilleux, I., Bendriss-Vermare, N., Bachelot, T., Goddard-Leon, S., Arfi, V., Biota, C., Doffin, A.C., Durand, I., Olive, D., et al. 2009. Regulatory T cells recruited through CCL22/CCR4 are selectively activated in lymphoid infiltrates surrounding primary breast tumors and lead to an adverse clinical outcome. Cancer research 69:2000-2009. 166. Montane, J., Bischoff, L., Soukhatcheva, G., Dai, D.L., Hardenberg, G., Levings, M.K., Orban, P.C., Kieffer, T.J., Tan, R., and Verchere, C.B. 2011. Prevention of murine autoimmune diabetes by CCL22-mediated Treg recruitment to the pancreatic islets. The Journal of clinical investigation 121:3024-3028. 167. Himmel, M.E., Crome, S.Q., Ivison, S., Piccirillo, C., Steiner, T.S., and Levings, M.K. 2010. Human CD4(+)FOXP3(+) regulatory T cells produce CXCL8 and recruit neutrophils. European journal of immunology. 168. Cartharius, K., Frech, K., Grote, K., Klocke, B., Haltmeier, M., Klingenhoff, A., Frisch, M., Bayerlein, M., and Werner, T. 2005. MatInspector and beyond: promoter analysis based on transcription factor binding sites. Bioinformatics 21:2933-2942. 169. Bacchetta, R., Passerini, L., Gambineri, E., Dai, M., Allan, S.E., Perroni, L., Dagna- Bricarelli, F., Sartirana, C., Matthes-Martin, S., Lawitschka, A., et al. 2006. Defective regulatory and effector T cell functions in patients with FOXP3 mutations. The Journal of clinical investigation 116:1713-1722. 170. Himmel, M.E., Crome, S.Q., Ivison, S., Piccirillo, C., Steiner, T.S., and Levings, M.K. 2011. Human CD4(+) FOXP3(+) regulatory T cells produce CXCL8 and recruit neutrophils. Eur J Immunol 41:306-312. 171. Palmer, L.A., Sale, G.E., Balogun, J.I., Li, D., Jones, D., Molldrem, J.J., Storb, R.F., and Ma, Q. 2010. Chemokine receptor CCR5 mediates alloimmune responses in graft-versus-host disease. Biology of blood and marrow transplantation : journal of the American Society for Blood and Marrow Transplantation 16:311-319. 172. Lichterfeld, M., Leifeld, L., Nischalke, H.D., Rockstroh, J.K., Hess, L., Sauerbruch, T., and Spengler, U. 2002. Reduced CC chemokine receptor (CCR) 1 and CCR5 surface expression on peripheral blood T lymphocytes from patients with chronic hepatitis C infection. The Journal of infectious diseases 185:1803-1807. 173. Tan, M.C., Goedegebuure, P.S., Belt, B.A., Flaherty, B., Sankpal, N., Gillanders, W.E., Eberlein, T.J., Hsieh, C.S., and Linehan, D.C. 2009. Disruption of CCR5- dependent homing of regulatory T cells inhibits tumor growth in a murine model of pancreatic cancer. Journal of immunology 182:1746-1755. 174. Mellanby, R.J., Thomas, D., Phillips, J.M., and Cooke, A. 2007. Diabetes in non- obese diabetic mice is not associated with quantitative changes in CD4+ CD25+ Foxp3+ regulatory T cells. Immunology 121:15-28.  117 175. Buckner, J.H. 2010. Mechanisms of impaired regulation by CD4(+)CD25(+)FOXP3(+) regulatory T cells in human autoimmune diseases. Nature reviews. Immunology 10:849-859. 176. McClymont, S.A., Putnam, A.L., Lee, M.R., Esensten, J.H., Liu, W., Hulme, M.A., Hoffmuller, U., Baron, U., Olek, S., Bluestone, J.A., et al. 2011. Plasticity of human regulatory T cells in healthy subjects and patients with type 1 diabetes. Journal of immunology 186:3918-3926. 177. Nguyen, K.D., Fohner, A., Booker, J.D., Dong, C., Krensky, A.M., and Nadeau, K.C. 2008. XCL1 enhances regulatory activities of CD4+ CD25(high) CD127(low/-) T cells in human allergic asthma. Journal of immunology 181:5386-5395. 178. Joosten, S.A., van Meijgaarden, K.E., Savage, N.D., de Boer, T., Triebel, F., van der Wal, A., de Heer, E., Klein, M.R., Geluk, A., and Ottenhoff, T.H. 2007. Identification of a human CD8+ regulatory T cell subset that mediates suppression through the chemokine CC chemokine ligand 4. Proceedings of the National Academy of Sciences of the United States of America 104:8029-8034. 179. Antoniazi, S., Price, H.P., Kropf, P., Freudenberg, M.A., Galanos, C., Smith, D.F., and Muller, I. 2004. Chemokine gene expression in toll-like receptor-competent and - deficient mice infected with Leishmania major. Infection and immunity 72:5168- 5174. 180. Lande, R., Giacomini, E., Grassi, T., Remoli, M.E., Iona, E., Miettinen, M., Julkunen, I., and Coccia, E.M. 2003. IFN-alpha beta released by Mycobacterium tuberculosis- infected human dendritic cells induces the expression of CXCL10: selective recruitment of NK and activated T cells. Journal of immunology 170:1174-1182. 181. Nguyen, T.L., Sullivan, N.L., Ebel, M., Teague, R.M., and DiPaolo, R.J. 2011. Antigen-specific TGF-beta-induced regulatory T cells secrete chemokines, regulate T cell trafficking, and suppress ongoing autoimmunity. Journal of immunology 187:1745-1753. 182. Mailloux, A.W., and Young, M.R. 2010. Regulatory T-cell trafficking: from thymic development to tumor-induced immune suppression. Critical reviews in immunology 30:435-447. 183. Zhou, Y., Kurihara, T., Ryseck, R.P., Yang, Y., Ryan, C., Loy, J., Warr, G., and Bravo, R. 1998. Impaired macrophage function and enhanced T cell-dependent immune response in mice lacking CCR5, the mouse homologue of the major HIV-1 coreceptor. Journal of immunology 160:4018-4025. 184. Andres, P.G., Beck, P.L., Mizoguchi, E., Mizoguchi, A., Bhan, A.K., Dawson, T., Kuziel, W.A., Maeda, N., MacDermott, R.P., Podolsky, D.K., et al. 2000. Mice with a selective deletion of the CC chemokine receptors 5 or 2 are protected from dextran sodium sulfate-mediated colitis: lack of CC chemokine receptor 5 expression results in a NK1.1+ lymphocyte-associated Th2-type immune response in the intestine. Journal of immunology 164:6303-6312. 185. Algood, H.M., and Flynn, J.L. 2004. CCR5-deficient mice control Mycobacterium tuberculosis infection despite increased pulmonary lymphocytic infiltration. Journal of immunology 173:3287-3296. 186. Sadlon, T.J., Wilkinson, B.G., Pederson, S., Brown, C.Y., Bresatz, S., Gargett, T., Melville, E.L., Peng, K., D'Andrea, R.J., Glonek, G.G., et al. 2010. Genome-wide  118 identification of human FOXP3 target genes in natural regulatory T cells. Journal of immunology 185:1071-1081. 187. Williams, L.M., and Rudensky, A.Y. 2007. Maintenance of the Foxp3-dependent developmental program in mature regulatory T cells requires continued expression of Foxp3. Nature immunology 8:277-284. 188. Castellino, F., Huang, A.Y., Altan-Bonnet, G., Stoll, S., Scheinecker, C., and Germain, R.N. 2006. Chemokines enhance immunity by guiding naive CD8+ T cells to sites of CD4+ T cell-dendritic cell interaction. Nature 440:890-895. 189. Fontenot, J.D., Rasmussen, J.P., Gavin, M.A., and Rudensky, A.Y. 2005. A function for interleukin 2 in Foxp3-expressing regulatory T cells. Nature immunology 6:1142- 1151. 190. D'Cruz, L.M., and Klein, L. 2005. Development and function of agonist-induced CD25+Foxp3+ regulatory T cells in the absence of interleukin 2 signaling. Nature immunology 6:1152-1159. 191. Scheffold, A., Huhn, J., and Hofer, T. 2005. Regulation of CD4+CD25+ regulatory T cell activity: it takes (IL-)two to tango. European journal of immunology 35:1336- 1341. 192. Sakaguchi, S., Wing, K., Onishi, Y., Prieto-Martin, P., and Yamaguchi, T. 2009. Regulatory T cells: how do they suppress immune responses? International immunology 21:1105-1111. 193. Tran, D.Q., Glass, D.D., Uzel, G., Darnell, D.A., Spalding, C., Holland, S.M., and Shevach, E.M. 2009. Analysis of adhesion molecules, target cells, and role of IL-2 in human FOXP3+ regulatory T cell suppressor function. Journal of immunology 182:2929-2938. 194. Lund, J.M., Hsing, L., Pham, T.T., and Rudensky, A.Y. 2008. Coordination of early protective immunity to viral infection by regulatory T cells. Science 320:1220-1224. 195. Tang, Q., Bluestone, J.A., and Kang, S.M. 2012. CD4(+)Foxp3(+) regulatory T cell therapy in transplantation. J Mol Cell Biol 4:11-21. 196. Bluestone, J.A. 2005. Regulatory T-cell therapy: is it ready for the clinic? Nature reviews. Immunology 5:343-349. 197. Sakaguchi, S. 2004. Naturally arising CD4+ regulatory t cells for immunologic self- tolerance and negative control of immune responses. Annual review of immunology 22:531-562. 198. Bruder, D., Probst-Kepper, M., Westendorf, A.M., Geffers, R., Beissert, S., Loser, K., von Boehmer, H., Buer, J., and Hansen, W. 2004. Neuropilin-1: a surface marker of regulatory T cells. European journal of immunology 34:623-630. 199. Milpied, P., Renand, A., Bruneau, J., Mendes-da-Cruz, D.A., Jacquelin, S., Asnafi, V., Rubio, M.T., MacIntyre, E., Lepelletier, Y., and Hermine, O. 2009. Neuropilin-1 is not a marker of human Foxp3+ Treg. European journal of immunology 39:1466- 1471. 200. Yamaguchi, T., Hirota, K., Nagahama, K., Ohkawa, K., Takahashi, T., Nomura, T., and Sakaguchi, S. 2007. Control of immune responses by antigen-specific regulatory T cells expressing the folate receptor. Immunity 27:145-159. 201. Walker, L.S. 2007. Regulatory T cells: Folate receptor 4: a new handle on regulation and memory? Immunology and cell biology 85:506-507.  119 202. Haining, W.N., and Wherry, E.J. 2010. Integrating genomic signatures for immunologic discovery. Immunity 32:152-161. 203. Nevins, J.R., and Potti, A. 2007. Mining gene expression profiles: expression signatures as cancer phenotypes. Nature reviews. Genetics 8:601-609. 204. Stegmaier, K., Ross, K.N., Colavito, S.A., O'Malley, S., Stockwell, B.R., and Golub, T.R. 2004. Gene expression-based high-throughput screening(GE-HTS) and application to leukemia differentiation. Nat Genet 36:257-263. 205. Banerji, V., Frumm, S.M., Ross, K.N., Li, L.S., Schinzel, A.C., Hahn, C.K., Kakoza, R.M., Chow, K.T., Ross, L., Alexe, G., et al. 2012. The intersection of genetic and chemical genomic screens identifies GSK-3alpha as a target in human acute myeloid leukemia. The Journal of clinical investigation 122:935-947. 206. Sugimoto, N., Oida, T., Hirota, K., Nakamura, K., Nomura, T., Uchiyama, T., and Sakaguchi, S. 2006. Foxp3-dependent and -independent molecules specific for CD25+CD4+ natural regulatory T cells revealed by DNA microarray analysis. International immunology 18:1197-1209. 207. Zheng, Y., Josefowicz, S.Z., Kas, A., Chu, T.T., Gavin, M.A., and Rudensky, A.Y. 2007. Genome-wide analysis of Foxp3 target genes in developing and mature regulatory T cells. Nature 445:936-940. 208. Hill, J.A., Feuerer, M., Tash, K., Haxhinasto, S., Perez, J., Melamed, R., Mathis, D., and Benoist, C. 2007. Foxp3 transcription-factor-dependent and -independent regulation of the regulatory T cell transcriptional signature. Immunity 27:786-800. 209. McHugh, R.S., Whitters, M.J., Piccirillo, C.A., Young, D.A., Shevach, E.M., Collins, M., and Byrne, M.C. 2002. CD4(+)CD25(+) immunoregulatory T cells: gene expression analysis reveals a functional role for the glucocorticoid-induced TNF receptor. Immunity 16:311-323. 210. Pfoertner, S., Jeron, A., Probst-Kepper, M., Guzman, C.A., Hansen, W., Westendorf, A.M., Toepfer, T., Schrader, A.J., Franzke, A., Buer, J., et al. 2006. Signatures of human regulatory T cells: an encounter with old friends and new players. Genome biology 7:R54. 211. Garin, M.I., Chu, C.C., Golshayan, D., Cernuda-Morollon, E., Wait, R., and Lechler, R.I. 2007. Galectin-1: a key effector of regulation mediated by CD4+CD25+ T cells. Blood 109:2058-2065. 212. Kubach, J., Lutter, P., Bopp, T., Stoll, S., Becker, C., Huter, E., Richter, C., Weingarten, P., Warger, T., Knop, J., et al. 2007. Human CD4+CD25+ regulatory T cells: proteome analysis identifies galectin-10 as a novel marker essential for their anergy and suppressive function. Blood 110:1550-1558. 213. Haining, W.N., Angelosanto, J., Brosnahan, K., Ross, K., Hahn, C., Russell, K., Drury, L., Norton, S., Nadler, L., and Stegmaier, K. 2008. High-throughput gene expression profiling of memory differentiation in primary human T cells. BMC immunology 9:44. 214. Haining, W.N., Ebert, B.L., Subrmanian, A., Wherry, E.J., Eichbaum, Q., Evans, J.W., Mak, R., Rivoli, S., Pretz, J., Angelosanto, J., et al. 2008. Identification of an evolutionarily conserved transcriptional signature of CD8 memory differentiation that is shared by T and B cells. Journal of immunology 181:1859-1868. 215. Golub, T.R., Slonim, D.K., Tamayo, P., Huard, C., Gaasenbeek, M., Mesirov, J.P., Coller, H., Loh, M.L., Downing, J.R., Caligiuri, M.A., et al. 1999. Molecular  120 classification of cancer: class discovery and class prediction by gene expression monitoring. Science 286:531-537. 216. Gould, J., Getz, G., Monti, S., Reich, M., and Mesirov, J.P. 2006. Comparative gene marker selection suite. Bioinformatics 22:1924-1925. 217. Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S., et al. 2005. Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proceedings of the National Academy of Sciences of the United States of America 102:15545-15550. 218. Kulkarni, M.M. 2011. Digital multiplexed gene expression analysis using the NanoString nCounter system. Current protocols in molecular biology / edited by Frederick M. Ausubel ... [et al.] Chapter 25:Unit25B 10. 219. Wei, S., Kryczek, I., and Zou, W. 2006. Regulatory T-cell compartmentalization and trafficking. Blood 108:426-431. 220. McMurchy, A.N., and Levings, M.K. 2012. Suppression assays with human T regulatory cells: a technical guide. European journal of immunology 42:27-34. 221. Parman, C., Halling, C., and Gentleman, R. affyQCReport: QC Report Generation for affyBatch objects. 222. Green, G.H., and Diggle, P.J. 2007. On the operational characteristics of the Benjamini and Hochberg False Discovery Rate procedure. Statistical applications in genetics and molecular biology 6:Article27. 223. Mold, J.E., Venkatasubrahmanyam, S., Burt, T.D., Michaelsson, J., Rivera, J.M., Galkina, S.A., Weinberg, K., Stoddart, C.A., and McCune, J.M. 2010. Fetal and adult hematopoietic stem cells give rise to distinct T cell lineages in humans. Science 330:1695-1699. 224. Thornton, A.M., Korty, P.E., Tran, D.Q., Wohlfert, E.A., Murray, P.E., Belkaid, Y., and Shevach, E.M. 2010. Expression of Helios, an Ikaros transcription factor family member, differentiates thymic-derived from peripherally induced Foxp3+ T regulatory cells. Journal of immunology 184:3433-3441. 225. D'Alise, A.M., Auyeung, V., Feuerer, M., Nishio, J., Fontenot, J., Benoist, C., and Mathis, D. 2008. The defect in T-cell regulation in NOD mice is an effect on the T- cell effectors. Proceedings of the National Academy of Sciences of the United States of America 105:19857-19862. 226. Olsen, J., Gerds, T.A., Seidelin, J.B., Csillag, C., Bjerrum, J.T., Troelsen, J.T., and Nielsen, O.H. 2009. Diagnosis of ulcerative colitis before onset of inflammation by multivariate modeling of genome-wide gene expression data. Inflammatory bowel diseases 15:1032-1038. 227. Martinez-Llordella, M., Puig-Pey, I., Orlando, G., Ramoni, M., Tisone, G., Rimola, A., Lerut, J., Latinne, D., Margarit, C., Bilbao, I., et al. 2007. Multiparameter immune profiling of operational tolerance in liver transplantation. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons 7:309-319. 228. Jailwala, P., Waukau, J., Glisic, S., Jana, S., Ehlenbach, S., Hessner, M., Alemzadeh, R., Matsuyama, S., Laud, P., Wang, X., et al. 2009. Apoptosis of CD4+ CD25(high) T cells in type 1 diabetes may be partially mediated by IL-2 deprivation. PloS one 4:e6527.  121 229. Himmel, M.E., Yao, Y., Orban, P.C., Steiner, T.S., and Levings, M.K. 2012. Regulatory T cell therapy for Inflammatory Bowel Disease: more questions than answers. Immunology. 230. Geiss, G.K., Bumgarner, R.E., Birditt, B., Dahl, T., Dowidar, N., Dunaway, D.L., Fell, H.P., Ferree, S., George, R.D., Grogan, T., et al. 2008. Direct multiplexed measurement of gene expression with color-coded probe pairs. Nat Biotechnol 26:317-325. 231. Gray, D. 1993. Immunological memory. Annual review of immunology 11:49-77. 232. Hoffmann, P., Eder, R., Boeld, T.J., Doser, K., Piseshka, B., Andreesen, R., and Edinger, M. 2006. Only the CD45RA+ subpopulation of CD4+CD25high T cells gives rise to homogeneous regulatory T-cell lines upon in vitro expansion. Blood 108:4260-4267. 233. Baecher-Allan, C., Wolf, E., and Hafler, D.A. 2006. MHC class II expression identifies functionally distinct human regulatory T cells. J Immunol 176:4622-4631. 234. Deaglio, S., Dwyer, K.M., Gao, W., Friedman, D., Usheva, A., Erat, A., Chen, J.F., Enjyoji, K., Linden, J., Oukka, M., et al. 2007. Adenosine generation catalyzed by CD39 and CD73 expressed on regulatory T cells mediates immune suppression. The Journal of experimental medicine 204:1257-1265. 235. Thornton, A.M., and Shevach, E.M. 2000. Suppressor effector function of CD4+CD25+ immunoregulatory T cells is antigen nonspecific. Journal of immunology 164:183-190. 236. Godfrey, W.R., Ge, Y.G., Spoden, D.J., Levine, B.L., June, C.H., Blazar, B.R., and Porter, S.B. 2004. In vitro-expanded human CD4(+)CD25(+) T-regulatory cells can markedly inhibit allogeneic dendritic cell-stimulated MLR cultures. Blood 104:453- 461. 237. Haining, W.N., Ebert, B.L., Subrmanian, A., Wherry, E.J., Eichbaum, Q., Evans, J.W., Mak, R., Rivoli, S., Pretz, J., Angelosanto, J., et al. 2008. Identification of an evolutionarily conserved transcriptional signature of CD8 memory differentiation that is shared by T and B cells. J Immunol 181:1859-1868. 238. Wherry, E.J., Ha, S.J., Kaech, S.M., Haining, W.N., Sarkar, S., Kalia, V., Subramaniam, S., Blattman, J.N., Barber, D.L., and Ahmed, R. 2007. Molecular signature of CD8+ T cell exhaustion during chronic viral infection. Immunity 27:670- 684. 239. Sarkar, S., Kalia, V., Haining, W.N., Konieczny, B.T., Subramaniam, S., and Ahmed, R. 2008. Functional and genomic profiling of effector CD8 T cell subsets with distinct memory fates. J Exp Med 205:625-640. 240. Campbell, D.J., and Koch, M.A. 2011. Phenotypical and functional specialization of FOXP3+ regulatory T cells. Nature reviews. Immunology 11:119-130. 241. Gavin, M.A., Rasmussen, J.P., Fontenot, J.D., Vasta, V., Manganiello, V.C., Beavo, J.A., and Rudensky, A.Y. 2007. Foxp3-dependent programme of regulatory T-cell differentiation. Nature 445:771-775. 242. Lin, W., Haribhai, D., Relland, L.M., Truong, N., Carlson, M.R., Williams, C.B., and Chatila, T.A. 2007. Regulatory T cell development in the absence of functional Foxp3. Nature immunology 8:359-368. 243. Akimova, T., Beier, U.H., Wang, L., Levine, M.H., and Hancock, W.W. 2011. Helios expression is a marker of T cell activation and proliferation. PloS one 6:e24226.  122 244. Suarez, N., Alfaro, C., Dubrot, J., Palazon, A., Bolanos, E., Erro, L., Hervas-Stubbs, S., Martinez-Forero, I., Morales-Kastresana, A., Martin-Algarra, S., et al. 2011. Synergistic effects of CTLA-4 blockade with tremelimumab and elimination of regulatory T lymphocytes in vitro and in vivo. Int J Cancer 129:374-386. 245. Dannull, J., Su, Z., Rizzieri, D., Yang, B.K., Coleman, D., Yancey, D., Zhang, A., Dahm, P., Chao, N., Gilboa, E., et al. 2005. Enhancement of vaccine-mediated antitumor immunity in cancer patients after depletion of regulatory T cells. J Clin Invest 115:3623-3633. 246. Lopez, M.N., Pereda, C., Segal, G., Munoz, L., Aguilera, R., Gonzalez, F.E., Escobar, A., Ginesta, A., Reyes, D., Gonzalez, R., et al. 2009. Prolonged survival of dendritic cell-vaccinated melanoma patients correlates with tumor-specific delayed type IV hypersensitivity response and reduction of tumor growth factor beta-expressing T cells. J Clin Oncol 27:945-952. 247. Kang, S., Xie, J., Ma, S., Liao, W., Zhang, J., and Luo, R. 2009. Targeted knock down of CCL22 and CCL17 by siRNA during DC differentiation and maturation affects the recruitment of T subsets. Immunobiology. 248. Fietta, A.M., Morosini, M., Passadore, I., Cascina, A., Draghi, P., Dore, R., Rossi, S., Pozzi, E., and Meloni, F. 2009. Systemic inflammatory response and downmodulation of peripheral CD25(+)Foxp3(+) T-regulatory cells in patients undergoing radiofrequency thermal ablation for lung cancer. Hum Immunol. 249. Wallen, H., Thompson, J.A., Reilly, J.Z., Rodmyre, R.M., Cao, J., and Yee, C. 2009. Fludarabine modulates immune response and extends in vivo survival of adoptively transferred CD8 T cells in patients with metastatic melanoma. PLoS One 4:e4749. 250. Tanimoto, T., Hori, A., and Kami, M. 2010. Sipuleucel-T immunotherapy for castration-resistant prostate cancer. N Engl J Med 363:1966; author reply 1967-1968. 251. Wilke, C.M., Kryczek, I., Wei, S., Zhao, E., Wu, K., Wang, G., and Zou, W. 2011. Th17 cells in cancer: help or hindrance? Carcinogenesis 32:643-649. 252. Ayyoub, M., Deknuydt, F., Raimbaud, I., Dousset, C., Leveque, L., Bioley, G., and Valmori, D. 2009. Human memory FOXP3+ Tregs secrete IL-17 ex vivo and constitutively express the T(H)17 lineage-specific transcription factor RORgamma t. Proceedings of the National Academy of Sciences of the United States of America 106:8635-8640. 253. Voo, K.S., Wang, Y.H., Santori, F.R., Boggiano, C., Arima, K., Bover, L., Hanabuchi, S., Khalili, J., Marinova, E., Zheng, B., et al. 2009. Identification of IL- 17-producing FOXP3+ regulatory T cells in humans. Proceedings of the National Academy of Sciences of the United States of America 106:4793-4798. 254. Vulink, A., Radford, K.J., Melief, C., and Hart, D.N. 2008. Dendritic cells in cancer immunotherapy. Adv Cancer Res 99:363-407. 255. Gilboa, E. 2007. DC-based cancer vaccines. The Journal of clinical investigation 117:1195-1203. 256. Scandella, E., Men, Y., Gillessen, S., Forster, R., and Groettrup, M. 2002. Prostaglandin E2 is a key factor for CCR7 surface expression and migration of monocyte-derived dendritic cells. Blood 100:1354-1361. 257. Morelli, A.E., and Thomson, A.W. 2003. Dendritic cells under the spell of prostaglandins. Trends Immunol 24:108-111.  123 258. Boullart, A.C., Aarntzen, E.H., Verdijk, P., Jacobs, J.F., Schuurhuis, D.H., Benitez- Ribas, D., Schreibelt, G., van de Rakt, M.W., Scharenborg, N.M., de Boer, A., et al. 2008. Maturation of monocyte-derived dendritic cells with Toll-like receptor 3 and 7/8 ligands combined with prostaglandin E2 results in high interleukin-12 production and cell migration. Cancer Immunol Immunother 57:1589-1597. 259. Sporri, R., and Reis e Sousa, C. 2005. Inflammatory mediators are insufficient for full dendritic cell activation and promote expansion of CD4+ T cell populations lacking helper function. Nature immunology 6:163-170. 260. Lasaro, M.O., and Ertl, H.C. 2009. New insights on adenovirus as vaccine vectors. Mol Ther 17:1333-1339. 261. Mailliard, R.B., Wankowicz-Kalinska, A., Cai, Q., Wesa, A., Hilkens, C.M., Kapsenberg, M.L., Kirkwood, J.M., Storkus, W.J., and Kalinski, P. 2004. alpha-type- 1 polarized dendritic cells: a novel immunization tool with optimized CTL-inducing activity. Cancer research 64:5934-5937. 262. Lee, J.J., Foon, K.A., Mailliard, R.B., Muthuswamy, R., and Kalinski, P. 2008. Type 1-polarized dendritic cells loaded with autologous tumor are a potent immunogen against chronic lymphocytic leukemia. J Leukoc Biol 84:319-325. 263. Matsushima, H., and Takashima, A. 2010. Bidirectional homing of Tregs between the skin and lymph nodes. The Journal of clinical investigation 120:653-656. 264. Thornton, A.M., Donovan, E.E., Piccirillo, C.A., and Shevach, E.M. 2004. Cutting edge: IL-2 is critically required for the in vitro activation of CD4+CD25+ T cell suppressor function. J Immunol 172:6519-6523. 265. Kang, S., Xie, J., Ma, S., Liao, W., Zhang, J., and Luo, R. 2010. Targeted knock down of CCL22 and CCL17 by siRNA during DC differentiation and maturation affects the recruitment of T subsets. Immunobiology 215:153-162. 266. Koch, M.A., Tucker-Heard, G., Perdue, N.R., Killebrew, J.R., Urdahl, K.B., and Campbell, D.J. 2009. The transcription factor T-bet controls regulatory T cell homeostasis and function during type 1 inflammation. Nat Immunol 10:595-602. 267. Miyara, M., Gorochov, G., Ehrenstein, M., Musset, L., Sakaguchi, S., and Amoura, Z. 2011. Human FoxP3+ regulatory T cells in systemic autoimmune diseases. Autoimmun Rev 10:744-755. 268. Curiel, T.J. 2008. Regulatory T cells and treatment of cancer. Current opinion in immunology 20:241-246. 269. Hoffmann, P., Boeld, T.J., Eder, R., Huehn, J., Floess, S., Wieczorek, G., Olek, S., Dietmaier, W., Andreesen, R., and Edinger, M. 2009. Loss of FOXP3 expression in natural human CD4+CD25+ regulatory T cells upon repetitive in vitro stimulation. European journal of immunology 39:1088-1097. 270. Hansmann, L., Schmidl, C., Kett, J., Steger, L., Andreesen, R., Hoffmann, P., Rehli, M., and Edinger, M. 2012. Dominant Th2 differentiation of human regulatory T cells upon loss of FOXP3 expression. Journal of immunology 188:1275-1282. 271. Liston, A., Lu, L.F., O'Carroll, D., Tarakhovsky, A., and Rudensky, A.Y. 2008. Dicer-dependent microRNA pathway safeguards regulatory T cell function. J Exp Med 205:1993-2004. 272. Zhou, X., Jeker, L.T., Fife, B.T., Zhu, S., Anderson, M.S., McManus, M.T., and Bluestone, J.A. 2008. Selective miRNA disruption in T reg cells leads to uncontrolled autoimmunity. J Exp Med 205:1983-1991.  124 273. Lu, L.F., Boldin, M.P., Chaudhry, A., Lin, L.L., Taganov, K.D., Hanada, T., Yoshimura, A., Baltimore, D., and Rudensky, A.Y. 2010. Function of miR-146a in controlling Treg cell-mediated regulation of Th1 responses. Cell 142:914-929.   


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items