expansion and characterization of human double negative ... · immunosuppression of autologous t...
TRANSCRIPT
EXPANSION AND CHARACTERIZATION
OF HUMAN DOUBLE NEGATIVE
REGULATORY T CELLS
by
PAULINA ACHITA
A thesis submitted in conformity with the requirements
for the degree of
Master of Science
Institute of Medical Science
University of Toronto
© Copyright by Paulina Achita, 2018
ii
EXPANSION AND CHARACTERIZATION OF HUMAN DOUBLE
NEGATIVE REGULATORY T CELLS
Paulina Achita
Master of Science, 2018
Institute of Medical Science
University of Toronto
ABSTRACT
A subset of αβ-T cell receptor (TCR) positive, NK lineage marker negative,
CD4−CD8− double negative regulatory T cells (DN Tregs) comprise only ~1% of
peripheral blood T lymphocytes. Clinical applications of DN Tregs in humans are
limited by their scarce number and the lack of effective expansion method. Herein is
described a protocol for large-scale ex vivo expansion of functional and pure human
DN Tregs. In vitro, expanded DN Tregs induce a cell-contact dependent
immunosuppression of autologous T cells and B cells, and are cytotoxic towards
various lung cancer, and leukemic cells. In vivo, infusion of DN Tregs delays an
onset of xenogeneic graft-versus-host disease (GVHD) in humanized mouse model.
Furthermore, short treatment of expanded DN Tregs with rapamycin augments their
suppressive function. Taken together, these results indicate a dual function of ex vivo
expanded DN Tregs and suggest their therapeutic potential in suppression of allograft
rejection and treatment of malignancies.
iii
ACKNOWLEDGMENTS
This project and adventure with science would not have been possible without
Dr. Li Zhang, my mentor and supervisor, who had given me an enormous opportunity
to exercise my curiosity and had continuously encouraged me to strive for excellence.
Furthermore, I would like to thank my advisory committee members Dr. Michele
Anderson and Dr. Reginald Gorczynski for providing all the insightful and important
remarks that drove this project forward, for their encouragement, understanding and
patience.
Special thanks go to all the past and present members of the Zhang lab. I feel
privileged to have met and worked with such brilliant people, many of whom became
my dear friends. Many thanks go to Dr. Dzana Dervovic who tirelessly spent
countless hours discussing scientific concepts with me, sharing her life knowledge
and being a huge supporter in everything I do; thanks to Dr. Dalam Ly for having
patience in answering all the questions I bombarded him with and for his calm,
grounding attitude during stormy days. Moreover, I would like to thank Jong-Bok
Lee for introducing me to the lab and helping me wet my feet; Dr. Cheryl D’Souza
and Betty Joe for always lending a helping hand and great advices; Tabea Haug who
transitioned from being my ‘pupil’ to being a great friend and travel partner; Dr.
Elena Streck for her humour and making Friday nights spent at the lab much more
bearable; and all the other current and former, graduate and summer students who had
made my days brighter: Branson Chen, Linda Junlin Yao, Heidi Kang, Aatif Quareshi,
Jonathan Musat and many others. I would also like to acknowledge other wonderful
people on the second floor of MaRS TMDT that had made this experience fun,
especially Dr. Michael Tang for being an incredible friend and ‘work husband,’ and
Ramzi Khattar for cheering me up in crises, and there were many. Another
recognition deserves Camilla Balgobin for her administrative assistance and members
of Gorczynski, Cattral and Cardella labs. Lastly, I would like to thank the UHN
Animal Facility and the SickKids-UHN Flow Cytometry Facility for playing an
iv
important part in execution of this project. Thank you everyone for all the laughs,
encouragement and believing in me. It would not be the same without you!
I would also like to thank my wonderful family who were very supportive and
encouraging during this challenging process. Even though they still have troubles
understanding what exactly I worked on, they are my biggest inspiration and
motivation to move forward in both science, and life. I am happy to show them that
all the inconvenient rides they were willing to give me, sometimes at very odd times,
were invaluable in completion of this project. My best friends deserve another
recognition; special thanks go to Caroline Lin for always being there for me through
good and bad, for being understanding and giving me the extra push when I needed it;
Susan Lee for the right dose of competitive edge, good laughs and best stress-
relieving activities. Last but not least, nothing would be possible without God who
had this all planned out for me.
v
CONTRIBUTIONS
Paulina Achita designed the experiments, acquired, analyzed and interpreted
the data under direct supervision from Dr. Li Zhang. Jong Bok Lee, Betty Joe and
Tabea Huag contributed to the acquisition of the data. Dr. Dzana Dervovic and Dr.
Dalam Ly contributed to the analysis and the interpretation of the data. Canadian
Institute of Health Research has funded this study, and Paulina Achita was a recipient
of Ontario Graduate Scholarship.
vi
TABLE OF CONTENTS
ACKNOWLEDGMENTS ......................................................................................... iii
CONTRIBUTIONS ..................................................................................................... v
TABLE OF CONTENTS ........................................................................................... vi
LIST OF ABBREVIATIONS .................................................................................... xi
LIST OF FIGURES .................................................................................................. xiv
LIST OF TABLES .................................................................................................... xvi
CHAPTER 1. INTRODUCTION ............................................................................... 1
1.1. Overview of the Immune System ................................................................. 2
1.2. Tolerance Mechanisms ................................................................................. 4
1.2.1. Central Tolerance .................................................................................. 4
1.2.2. Peripheral Tolerance .............................................................................. 5
1.3. Overview of Tregs ........................................................................................ 7
1.3.1. nTregs .................................................................................................... 8
1.3.1.1. Development of nTregs .................................................................. 9
1.3.1.2. Mechanisms of nTreg Suppression ................................................ 9
1.3.2. CD4+ iTregs ......................................................................................... 14
1.3.2.1. Th3 Cells ...................................................................................... 15
1.3.2.2. Tr1 Cells ....................................................................................... 15
vii
1.3.3. CD8+ Tregs .......................................................................................... 17
1.3.4. DN Tregs ............................................................................................. 19
1.3.4.1. Phenotype of DN Tregs ................................................................ 19
1.3.4.2. Overview of DN Treg Function ................................................... 20
1.3.4.3. Role of DN Tregs in GVHD ......................................................... 21
1.3.4.4. Role of DN Tregs in Cancer ......................................................... 22
1.3.4.5. DN Tregs in Lymphoproliferative Syndrome .............................. 22
1.3.4.6. DN Tregs in Diabetes ................................................................... 23
1.3.4.7. Mechanisms of DN Treg-mediated Suppression .......................... 24
1.3.4.8. Development of DN Tregs ........................................................... 25
1.4. Immunotherapy with Tregs ........................................................................ 26
1.4.1. Treg Expansion Methods ..................................................................... 27
1.4.1.1. Role of IL-2, IL-7 and IL-15 Growth Factors in Treg
Maintenance and Proliferation ..................................................... 27
1.4.1.2. Role of Rapamycin in Treg Expansion and Suppressive
Function ....................................................................................... 29
1.4.2. Treatment of GVHD ............................................................................ 30
1.4.3. Treatment of Autoimmune Diseases ................................................... 32
1.5. Hypothesis and Specific Research Aims .................................................... 33
CHAPTER 2. MATERIALS AND METHODS ..................................................... 36
viii
2.1. Blood Samples ............................................................................................ 37
2.2. Cell Isolation and Magnetic Sorting ........................................................... 37
2.3. Freezing and Thawing of Cells ................................................................... 38
2.4. Expansion of DN Tregs .............................................................................. 38
2.5. Antibodies and Flow Cytometry ................................................................ 39
2.6. Detection of Cytokines and Chemokines Secreted by DN Tregs ............... 41
2.7. In Vitro T cell and B cell Suppression Assays ........................................... 41
2.8. In Vitro Suppression Assays with Rapamycin-treated DN Tregs .............. 42
2.9. Transwell® Experiments ............................................................................ 42
2.10. Lymphocyte Cytotoxicity Assay .............................................................. 43
2.11. Cancer Cells Cytotoxicity Assay .............................................................. 43
2.12. Mice and Xenogeneic GVHD Model ....................................................... 44
2.13. Monitoring of Lymphocyte Migration and Proliferation In Vivo ............. 44
2.14. Data Analysis ............................................................................................ 45
CHAPTER 3. RESULTS .......................................................................................... 46
3.1. Frequencies and Phenotypic Analysis of TCRαβ+ CD3+ CD4− CD8−
DN Tregs in the Peripheral Blood of Healthy Adults ............................... 47
3.2. Human DN Tregs Can Be Expanded Ex Vivo ............................................ 51
3.3. Supplementation of IL-7 During Expansion Enhance Proliferation and
Suppressive Function of DN Tregs ........................................................... 54
ix
3.4. Ex Vivo Expanded DN Tregs are Potent Suppressors In Vitro ................... 56
3.5. Phenotype of Ex Vivo Expanded DN Tregs ............................................... 58
3.6. Cytokine Profile of Ex Vivo Expanded DN Tregs ...................................... 60
3.7. DN Treg-mediated Suppression Is Not Facilitated by IFN-γ or IL-10
Cytokines and Requires Cell-to-Cell Contact ........................................... 62
3.8. DN Tregs Do Not Suppress by Killing Responder Cells ........................... 69
3.9. Ex Vivo Expanded DN Tregs Kill Human Cancer Cells ............................ 71
3.10. Spatial and Temporal Dynamics of Human DN Tregs In Vivo ................ 73
3.11. Ex Vivo Expanded DN Tregs Delayed Onset of Xenogeneic GVHD in
NSG Mice .................................................................................................. 75
3.12. Rapamycin Augmented Immunosuppressive Function of DN Tregs In
Vitro and In Vivo ....................................................................................... 78
CHAPTER 4. DISCUSSION .................................................................................... 82
4.1. General Discussion ..................................................................................... 83
4.2. Future Directions ........................................................................................ 93
4.2.1. To Identify Markers That Are Critical for DN Treg Function ............ 94
4.2.2. To Determine the Effects of DN Treg-Immunosuppressive Agents
Combination Therapy on the Treatment of Xenogeneic GVHD ....... 95
4.2.3. To Determine Whether Human DN Tregs Suppress DCs ................... 96
4.2.4. To Determine Whether Trogocytosis is Critical for Human DN
Treg Function ..................................................................................... 97
x
4.2.5. To Determine Signalling Pathways That Govern DN Tregs ............... 98
4.3. Conclusion .................................................................................................. 99
REFERENCES ........................................................................................................ 101
xi
LIST OF ABBREVIATIONS
7-AAD aAPC Ab ACK ACT
7-Aminoactinomycin D Artificial antigen presenting cell Antibody Ammonium-chloride-potassium Adoptive cellular therapy
Ag AICD
Antigen Activation-induced cell death
AML Acute myeloid leukemia Allo Allogeneic ALPS Autoimmune lymphoproliferative syndrome APC Bcl-2 Bcl-xL BCR
Antigen presenting cell or Allophycocyanin B cell lymphoma 2 B cell lymphoma extra-large B cell receptor
BM BMT BSA
Bone marrow Bone marrow transplant Bovine serum albumin
CD Cluster of differentiation CFSE 5,6-carboxyfluorescein diacetate succinimidyl ester CsA Cyclosporine A CTL Cytotoxic T lymphocyte CTLA-4 DAPI DC DMSO DN
Cytotoxic T lymphocyte antigen 4 4',6-diamidino-2-phenylindole Dendritic cell Dimethyl sulfoxide Double negative
ELISA Enzyme-linked immunosorbent assay FACS FasL FBS
Fluorescence-activated cell sorting Fas ligand Fetal bovine serum
FITC Fgl-2
Fluorescein isothiocyanate Fibrinogen-like protein 2
FMO Fluorescence-minus-one Foxp3 G-CSF GALT GM-CSF
Forkhead box P3 Granulocyte colony-stimulating factor Gut-associated lymphoid tissue Granulocyte-macrophage colony-stimulating factor
GVHD Graft-versus-host disease GVL Graft-versus-leukemia HLA Human leukocyte antigen
xii
HSCT Hematopoietic stem cell transplant ICOS Inducible co-stimulator IDO Indoleamine 2,3-dioxygenase IFN Interferon Ig Immunoglobulin IL Interleukin IPEX iTreg
Immunodysregulation, polyendocrinopathy, enteropathy, X-linked Inducible regulatory T cells
i.v. Intravenous JAK Janus kinase LAG-3 LAP
Lymphocyte-activation gene 3 Latency-associated peptide
LN Lymph node LPR Lymphoproliferation mAb MACS MCP-1 mDC
Monoclonal antibody Magnetic-activated cell sorting Monocyte chemoattractant protein 1 Myeloid dendritic cell
MHC MIP
Major histocompatibility complex Macrophage inflammatory protein
MLR Mixed lymphocyte reaction MS Multiple sclerosis MST Median survival time mTOR Mammalian target of rapamycin NK NKT
Natural killer Natural killer T cell
NOX2 NADPH oxidase 2 NSG nTreg PAMP
NOD/SCID IL-2Rgnull Naturally-occurring regulatory T cells Pathogen-associated-molecular patterns
PBMC Peripheral blood mononuclear cells PBS Phosphate-buffered saline PD-1 PDGF
Programmed cell death protein 1 Peptide-derived growth factor
PE Phycoerythrin PerCP Peridinin chlorophyll A protein PI Propidium iodide PI3K Phosphoinositide 3-kinase PMA PRR
Phorbol 12-myristate 13-acetate Pattern recognition receptors
RA RANTES
Rheumatoid arthritis Regulated on activation, normal T cell expressed and secreted protein
xiii
SLE SOT
Systemic lupus erythematous Solid organ transplant
STAT Signal transducer and activator of transcription Tconv Conventional T cell Teff Effector T cell T1D TCR
Type one diabetes mellitus T cell receptor
TGF-β Transforming growth factor beta Th1 T helper type 1 Th2 T helper type 2 Th3 TLR TNF TRAIL TREC
T helper type 3 Toll-like receptor Tumour necrosis factor TNF-related apoptosis-inducing ligand T-cell receptor excision circles
Treg Regulatory T cell Tr1 Type 1 regulatory T cell Tsup Suppressor T cell UCB VEGF
Umbilical cord blood Vascular endothelial growth factor
xiv
LIST OF FIGURES
Figure 1. Phenotypic characteristics of DN Tregs isolated from peripheral blood. .......... 48
Figure 2. Cell surface expression of memory markers CCR7 and CD45RO on
peripheral blood T cells. ................................................................................... 50
Figure 3. Schematic representation of the method for ex vivo expansion of DN Tregs. ... 52
Figure 4. Isolation and expansion of DN Tregs. ............................................................... 53
Figure 5. The effect of supplementation of IL-7 and IL-15 during the expansion on
DN Treg proliferation and function.. ................................................................ 55
Figure 6. DN Tregs suppress proliferation of autologous T cells, and CD19+ B cells. ..... 57
Figure 7. Phenotypic characteristics of ex vivo expanded DN Tregs.. .............................. 59
Figure 8. Cytokine profile of DN Tregs. ........................................................................... 61
Figure 9. Addition of IL-2 and/or IL-7 directly to the suppression assay does not
impair functionality of DN Tregs. .................................................................... 63
Figure 10. DN Tregs produce IL-10 and IFN-γ. ................................................................ 64
Figure 11. Role of IFN-γ and IL-10 in the mechanism of DN Treg suppression. ............. 65
Figure 12. Mechanism of inhibition mediated by DN Tregs requires cell-to-cell
contact.. ............................................................................................................. 67
Figure 13. DN Tregs suppress secretion of IFN-γ by CD4+ T cells and CD8+ T cells. .... 68
Figure 14. DN Tregs do not kill autologous CD4+ or CD8+ T cells. ................................. 70
Figure 15. Ex vivo expanded DN Tregs can kill human cancer cells. ............................... 72
xv
Figure 16. Tracking and proliferation of human lymphocytes adoptively transferred to
NSG mice. ......................................................................................................... 74
Figure 17. Schematic protocol of the xenogeneic GVHD experiment. ............................. 76
Figure 18. Treatment with DN Tregs delayed the onset of xenogeneic GVHD. .............. 77
Figure 19. Rapamycin-treated DN Tregs manifest augmented regulatory function. ........ 80
Figure 20. Rapamycin-treated DN Tregs delayed the onset of xenogeneic GVHD.. ........ 81
xvi
LIST OF TABLES
Table 1. List of the antibodies used in this study. ...................................................... 40
1
CHAPTER 1. INTRODUCTION
2
Chapter 1
Introduction
1.1. Overview of the Immune System
The immune system encompasses many biological structures and processes that
are fundamental for protecting the organism from pathogens, and for distinguishing those
pathogens from the body’s healthy tissue. The immune system has been simplistically
subdivided into two “lines of defense”: the innate immunity and the adaptive immunity.
Although the two classes of the immune system have evolved independently, they are not
mutually exclusive, but rather complementary. Defects in either system, although rare,
result in host vulnerability to infections, malignancy, autoimmunity and
immunodeficiency disorders.
The innate subdivision of the immune system represents the first line of defense
against intruding pathogens and environmental agents. The underlying defense
mechanisms include anatomical barriers such as skin and mucosal tissues, and
physiological barriers such as acidic pH of the stomach, the onset of fever during the
infection, or activation of the complement system. The innate response evolved to non-
specifically recognize molecules that are evolutionary conserved amongst the broad
groups of microbes, referred to as pathogen-associated-molecular patterns (PAMPs),
through pattern recognition receptors (PRRs) (Mogensen, 2009). These receptors are
present on the cell surface of innate immune cells, such as dendritic cells, macrophages
and neutrophils, and the most extensively studied are Toll-like receptors (TLRs) (Kawai
and Akira, 2010). Activation of immune cells via PRRs leads to synthesis of pro-
inflammatory cytokines, chemokines, cell adhesion molecules, and other anti-microbial
agents, which ultimately lead to elimination of pathogens, and priming of the adaptive
immune response (Mogensen, 2009). Traditionally, the innate immune system was
classified as incapable of providing immunological memory. However, a growing body
of evidence suggests that the innate immune memory may be initiated upon priming
3
event during the first exposure (Netea et al., 2011), but preserves for a much shorter time
period than the second division of the immune system (Netea et al., 2015), known as
adaptive or acquired immunity.
The adaptive immune system has evolved sophisticated mechanisms to recognize
and fight a broad range of pathogens. The cells of paramount importance to the adaptive
immune system are T lymphocytes and B lymphocytes, which are capable of specific
elimination of pathogens via recognition of their antigens, which are unique molecular
structures distinctive for each pathogen. These cells express unique receptors on their cell
surface: T cell receptors (TCR) on T cells and B cell receptors (BCR) on B cells. TCR
and BCR possess enormous repertoire diversity due to somatic recombination that occurs
during the development of T cells in the thymus, and B cells in the bone marrow,
allowing specific recognition of antigens from nearly all pathogens. At the initial contact
with a pathogen, there is a lag time between exposure and mounting maximal immune
response, because naïve lymphocytes must first encounter the antigen in periphery,
followed by activation of lymphocytes and differentiation into effector, or long-lived
memory cells. Thanks to the immunological memory, subsequent encounters with the
same pathogen will enable the host to mount a more rapid and efficient response.
Activation of the innate immunity plays a crucial role in stimulating the adaptive
immunity (Kumagai and Akira, 2010; Netea et al., 2015). Innate immune cells migrate
towards the source of infection and discharge a vast array of antimicrobial weaponry,
which in turn facilitates recruitment of other immune cells to the injury site. The key
mediators between the innate and the adaptive immunity are dendritic cells (DC), which
together with macrophages and B cells, belong to a class of professional antigen
presenting cells (APCs). APCs digest antigens into smaller fragments and display the
peptide fragments coupled with major histocompatibility complexes (MHC) on their cell
surface in a process known as ‘antigen presentation.’ All nucleated cells present
endogenous peptides on MHC class I molecules, whereas APCs may also internalize
extracellular peptides and present it on MHC class II molecules. T cells can then
recognize these complexes using TCR on their cell surface.
4
T cells can be divided into two main groups based on the type of MHC-peptide
complexes they recognize, which is concurrent with differential expression of TCR co-
receptors on their cell surface. Thus T cells that express CD8 co-receptor recognize
antigens that are presented by MHC class I molecules. These cells differentiate into
cytotoxic T cells whose function is to destroy infected cells by killing. On the other hand,
T cells that express CD4 co-receptor recognize peptides presented by MHC class II
molecules and release cytokines, and growth factors that regulate other immune cells.
Depending on the cytokine milieu that CD4+ T cells encounter during TCR activation,
naïve CD4+ T cells may differentiate into several lineages of T helper (Th) and T
regulatory (Treg) cells, which are defined by their function and pattern of cytokine
production.
However, the TCR-MHC engagement may not be robust enough to effectively
activate T cells. In order to decrease the activation threshold APCs must present a broad
range of co-stimulatory molecules that will bind to the corresponding receptors on the T
cells. The activation, however, must be tightly regulated to ensure that activation is
resolved following the combat of pathogens by the T cells. Furthermore, T cells that
recognize self-antigens must be eliminated to avoid mounting an attack against the host
organism. Fortunately, the immune system has evolved multiple mechanisms to ensure
immune homeostasis and suppression of the immune responses against self-antigens.
1.2. Tolerance Mechanisms
1.2.1. Central Tolerance
Numerous tolerance mechanisms exist to prevent detrimental self-reactivity. The
first checkpoint of self-tolerance happens in the thymus during the final stages of T cell
development, but before the maturation and circulation of T cells. The primary
mechanism of central tolerance is ‘negative selection’, also known as ‘clonal deletion,’
which was first proposed by Frank MacFarlane Burnet who ultimately shared the Noble
5
Prize in 1960 with Peter Medawar for their work on immunological tolerance - a
phenomenon explained by clonal deletion (Liston, 2011). Since somatic recombination
produces TCRs of enormous diversity, it would be virtually impossible to avoid
production of self-reactive TCRs. Therefore, a vast majority of the T cell precursors that
express TCRs with high avidity for self-MHC complexes are eliminated via apoptosis;
only those T cell precursors that express low avidity TCRs for self-peptides will undergo
maturation and reach the periphery (Starr et al., 2003). Furthermore, some thymocytes
with high-avidity TCRs for self-peptides may avoid apoptosis by internalization of TCR
in order to undergo a secondary gene rearrangement to change the specificity of their
TCRα locus (McGargill et al., 2000; Wang et al., 1998). This process is known as
‘receptor editing’ and replaces a self-reactive TCR with a new, non-reactive one.
1.2.2. Peripheral Tolerance
Some autoreactive T cells may not have sufficient affinity for self-antigens and
the thymic epithelial cells may not express all the possible self-antigens. Therefore, these
low-avidity autoreactive T cells may escape central tolerance mechanisms and migrate to
the periphery. When left unchecked, the activation of these cells may lead to the
development of autoimmune disease. Fortunately, several peripheral tolerance
mechanisms evolved to restrain both the number and the function of autoreactive T cells.
These mechanisms include anatomical isolation, cellular inactivation, deletion by
activation-induced cell death, or direct suppression of self-reactive T cells by specialized
cells (Mueller, 2010).
Immune-privileged sites.
Certain specialized tissues in the human body are able to tolerate the introduction
of antigens without evoking an inflammatory response against them. In these sites tissue
allografts are readily accepted, and thus termed ‘privileged sites’ for having an advantage
over other sites, where usual immune response would be mounted (Forrester et al., 2008).
The privileged sites offer necessary protection to tissues where the inflammatory
response could cause permanent damage as it is in the case of brain, eyes, testes, placenta
6
and fetus. The anatomy of these sites limits the access to lymphatic drainage, thus
restraining the immune cells from entering such sites. Although the immune cells can
access these sites through circulation, markedly fewer cells from these sites can reach the
lymphatic organs. Some sites may be characterized by low expression of classical MHC
class I molecules, expression of immunoregulatory molecules and secretion of
immunosuppressive mediators that aid in the maintenance of immune privilege (Forrester
et al., 2008). An interesting phenomenon became evident through animal models, where
tolerance to antigen introduced in the privileged sites, such as brain or eye, was
transferable to other sites in an antigen (Ag)-specific manner (Galea et al., 2007;
Medawar, 1948).
Clonal anergy.
Effective activation of T cells requires simultaneous delivery of Ag-specific
signals and co-stimulatory signals. Anergy occurs when TCR recognizes peptide-MHC
complexes on tissues that lack co-stimulatory molecules, or display co-inhibitory
molecules. Instead of activation, such cells are induced into a state of long-term hypo-
responsiveness characterized by active repression of TCR signalling and thus the inability
to perform any effector functions (Wells, 2009).
Activation-induced cell death.
Autoreactive T cells chronically engaged by self-peptide-MHC complexes can
also die by apoptosis. This activation-induced cell death is a result of the combination of
two molecular mechanisms: Fas receptor engagement with FasL, and Bim-dependent
sequestering of a B cell lymphoma 2 (Bcl-2)- and B cell lymphoma extra-large (Bcl-xL)-
regulated mitochondrial death pathway (Mueller, 2010). Activated T cells upregulate Fas
on their cell surface and apoptosis is triggered upon engagement with cells expressing
FasL. Activated T cells that upregulate both Fas and FasL may also kill each other
directly.
Control by regulatory T cells.
7
Specialized immune cells termed regulatory T cells (Tregs) can actively suppress
self-reactive cells through many different mechanisms depending on the cell type
involved (discussed in the subsequent chapters). Tregs can be described as highly
heterogeneous population of cells of different developmental origins. They are crucial in
controlling adaptive immune responses ranging from autoimmune diseases to
inflammatory conditions.
Many mechanisms are responsible for eliciting immunological tolerance.
However, this state of unresponsiveness may be deleterious under certain circumstances.
The best examples of undesired tolerance are observed in the tumour microenvironments,
where excessive tolerance weakens immune response against tumour antigens, or during
chronic infections, where microbial or viral agents can be no longer eliminated. On the
other hand, these tolerance mechanisms can be harnessed to restrain autoimmune diseases,
graft rejections, and allergies. Considering all the different mechanisms discussed earlier,
Tregs represent the most feasible component of the immune system to be exploited for
clinical applications.
1.3. Overview of Tregs
In 1971, Gershon and Kondo published the first paper suggesting that thymus-
derived lymphocytes were required for tolerance induction (Gershon and Kondo, 1971).
The adoptive transfer of such cells from mice that were tolerated to the particular antigen
(sheep red blood cells) to naïve mice prevented the latter to respond to immunizing doses
of such antigens. The authors named this phenomenon “infectious immunological
tolerance” and suggested that these cells were Ag-specific, and produced
“immunosuppressive substances” as their mode of action (Germain, 2008). Many
subsequent studies and compelling “logic of the argument that the immune system
needed suitable brakes to prevent excessive activity” resulted in recognition of the
importance of this subset of cells by the scientific community, and these cells were
termed suppressor T cells (Tsup) (Germain, 2008; Kapp, 2008). However, interest in
8
Tsup fell drastically in the mid 1980s leading to downfall of the field for many reasons,
starting with the inability to generate stable long-term Tsup lines, the incapacity to
identify soluble factors mediating immunosuppression, negative findings regarding
identifying unique surface markers and transcription factors, and lastly, the shift in the
interests of immunological community to rather identify pivotal molecular elements of
the immune system (Ligocki and Niederkorn, 2015; Moller, 1988).
Sakaguchi resurrected interest in the suppressive lymphocytes in 1995
demonstrating that a small subset of CD4+ T cells that co-express IL-2Rα (CD25)
function as Tsup (Sakaguchi et al., 1995). Depletion of CD4+CD25+ cells in the normal
mice resulted in the spectrum of autoimmune diseases, when remaining CD4+CD25− cells
were transferred to the immunodeficient mice (Sakaguchi et al., 1995). Further studies
demonstrated the immunosuppressive activity of CD4+CD25+ in vitro and multiple
groups corroborated presence and function of these cells in humans (Baecher-Allan et al.,
2001; Jonuleit et al., 2001; Levings et al., 2001; Ng et al., 2001). With the ability to
finally isolate these cells and to avoid the negative connotation around Tsup, these cells
came to be known as Tregs.
Tregs can be classified into two groups based on their developmental origin:
naturally occurring (n) Tregs that differentiate in the thymus and inducible (i) Tregs that
differentiate in the periphery.
1.3.1. nTregs
Subsequent studies showed that the forkhead box P3 (Foxp3) transcription factor
is not only an essential intracellular marker but also an absolute requirement for function
and developmental fate of nTregs (Fontenot et al., 2003; Hori et al., 2003; Khattri et al.,
2003). The importance of Foxp3 in the development of Tregs, as well as the importance
of Tregs themselves, have been shown in scurfy mice and in human with immune
dysregulation, polyendocrinopathy, enteropathy, X-linked (IPEX) syndrome who develop
lethal autoimmune disease as the result of mutation in Foxp3 and thus Treg deficiency
(Bennett et al., 2001; Brunkow et al., 2001). Majority of IPEX patients expect to die
9
within the first year of life as uncontrolled activation and clonal expansion of CD4+ cells
results in autoimmune enteropathy, type-1 diabetes mellitus, infections and cutaneous
manifestations, such as lesions and severe eczema (Barzaghi et al., 2012).
Undeniable evidence that CD4+CD25+Foxp3+ cells are committed to regulatory
lineage during development in the thymus gave rise to term ‘naturally occurring Tregs’
(nTregs). The importance of nTregs was evident in mice who received neonatal
thymectomy three days after birth, which resulted in abrogated production of nTregs and
multi-organ autoimmune disease (Sakaguchi et al., 2007).
1.3.1.1. Development of nTregs
Since the function of nTregs is drastically different from that of Tconv cells, it
should be no surprise that nTregs follow developmental pathway different from that of
Tconv cells. During the development, most of double-positive thymocytes are unable to
recognize self-MHC complexes and thus die by apoptosis. Those cells whose TCR
engages with self-MHC complexes mature and become single-positive thymocytes
expressing either CD4 or CD8 co-receptor. The next step involves negative selection
(reviewed in section 1.2.1.), where the cells with high avidity or specificity for self-
peptides are eliminated (Hogquist et al., 2005). Only small proportion of the cells that
have low avidity for self-peptides leave the thymus to become Tconv cells; however,
cells at the single-positive stage with intermediate avidity for TCR complexes may
become nTregs (Kronenberg and Rudensky, 2005).
1.3.1.2. Mechanisms of nTreg Suppression
nTregs use numerous tools to suppress a large number of different target cells
including T cells, B cells, DCs, macrophages, osteoblasts, mast cells, NK cells and NKT
cells (Shevach, 2009). Shevach suggested that these mechanisms can be broadly divided
to those that target T cells, such as production of anti-inflammatory cytokines, killing the
target cells and cytokine consumption, and those that target APCs, which includes
decreasing co-stimulation or antigen presentation (Shevach, 2009). The mechanisms
10
listed below are not mutually exclusive and many mechanisms may be used
simultaneously.
Inhibitory Cell Surface Molecules.
nTregs may express on their cell surface a variety of inhibitory molecules whose
role is to dampen immune response. One of the most extensively studied molecules on
nTregs is a constitutively expressed cytotoxic T cell lymphocyte 4 (CTLA-4) receptor,
but on other T cell subsets only gets upregulated upon activation (Read et al., 2000;
Salomon et al., 2000; Takahashi et al., 2000). Interestingly, it is the Foxp3 transcription
factor that controls the expression of CTLA-4 in Tregs (Hori et al., 2003; Marson et al.,
2007; Ono et al., 2007; Wu et al., 2006; Zheng et al., 2007). Wing et al. has found that
CTLA-4 is crucial for Treg function as its deficiency impairs the regulatory function of
nTregs in vitro, and in vivo has been shown to induce systemic lymphoproliferation,
lethal autoimmune disease mediated by T cells and more effective immunity against
tumours (Wing et al., 2008). CTLA-4 acts as an immune checkpoint by direct binding
with CD80 and CD86 on Teff cells to inhibit expansion of these cells in vivo (Paust et al.,
2004), or by down-regulating CD80 and CD86 expression on DCs and thus inhibiting the
activation of Teff cells by DCs (Cederbom et al., 2000).
PD-1:PD-ligand 1 (PD-L1) pathway has also been strongly associated with nTreg
function, control of multiple tolerance checkpoints and is indispensible in delivering
inhibitory signals in persistent antigen stimulation, tumours, chronic infections, and play
an important role in T cell activation, tolerance and inflammatory-induced tissue damage
(discussed in great detail by (Francisco et al., 2010)). Since PD-1:PD-L1 pathway is
commonly used by a multitude of different cells, the intricacies of this pathways will not
be discussed here. However, it is important to mention that PD-1 ligation in the presence
of TCR stimulation and TGF-β may transform naïve T cells into Tregs. This is especially
detrimental in cancer setting where nTregs perpetuate highly immune-suppressive
environment preventing the body to fight cancer on its own. Blockade of this pathway in
cancer patients by administrating the antibodies against PD-1 or PD-L1, as well as
CTLA-4, show promising results in the treatment of various cancers, started being used
11
routinely in the clinic and there are multiple ongoing clinical trials around the world (Iwai
et al., 2017).
Lymphocyte-activation gene 3 (LAG-3), a CD4-related molecule, is another
inhibitory cell surface marker that has been associated with suppressive mechanisms of
nTregs through direct Teff-Treg interactions or by modulation of APC function. The
latter mechanism, involves binding of LAG-3 to MHC class II molecules on DCs, which
results in downregulation of the expression of co-stimulatory molecules on DCs (Liang et
al., 2008). Furthermore, CD4+CD25+ Tregs from LAG-3 knockout mice manifest
reduced regulatory activity, which can be restored by ectopic expression of LAG-3 via
retroviral transduction on CD4+ T cells (Huang et al., 2004). However, a recent study
suggests that LAG-3 may also negatively impact proliferation of Tregs and their function
at the sites of inflammation by downregulation of CD25 (Zhang et al., 2017).
An inhibitory molecule TIGIT were found to be expressed on activated Tregs and
directly supress immune responses of Th1 and Th17 cells, as well as induce production of
fibrinogen-like protein 2 (Fgl-2) (Joller et al., 2014). Fgl-2 production also contributes to
nTreg-mediated immunosuppression as Fgl2-/- knock out mice develop autoimmune
glomerulonephritis due to inability of Tregs to mount immunosuppression despite their
higher frequency (Shalev et al., 2008). Binding of TIGIT with the DC surface ligand
CD155 (also known as poliovirus receptor) can also inhibit T cell responses (Yu et al.,
2009).
Killing.
Both human and murine nTregs may also directly kill autologous CD4+ and CD8+
T cells, CD14+ monocytes and mature, and immature DCs using perforin/granzyme-
dependent pathway (Grossman et al., 2004). However nTreg do not kill by Fas/FasL or
TNF-related apoptosis-inducing ligand (TRAIL) (Shevach et al., 2006). nTregs were also
found to preferentially kill Ag-presenting B cells, but not bystander B cells (Zhao et al.,
2006).
Ectonucleotidase Activity.
12
Deaglio et al. identified that co-expression of ENTPD1 and ecto-5’-nucleotidase,
known as CD39 and CD73, respectively, is unique to nTregs population among T cells
(Deaglio et al., 2007). CD39 is critical to nTreg function, as CD39-deficient mice showed
diminished immunosuppressive function in vitro and were unable to prevent allograft
rejection in vivo (Deaglio et al., 2007). Together, CD39 and CD73 degrade extracellular
ATP, ADP, and AMP to generate pericellular adenosine. Adenosine reduces activation in
Teff cells by binding to A2A receptor, which is upregulated in T cells upon antigenic
stimulation, and thus the presence of adenosine reduces production and release of pro-
inflammatory cytokines by Teff cells (Ohta and Sitkovsky, 2001). This process can be
viewed as “immunological switch” that shifts ATP-driven pro-inflammatory state to an
anti-inflammatory state driven by the presence of adenosine (Antonioli et al., 2013).
Moreover, adenosine may participate in a positive feedback loop as nTregs also
upregulate A2A receptor expression upon TCR stimulation (Deaglio et al., 2007). In
contrast to Teff cells, adenosine binds to A2A receptor on nTregs, which in turn drives the
proliferation of Tregs and further influences their immunoregulatory function (Ohta et al.,
2012; Zarek et al., 2008). In humans, 90% of Foxp3+ Tregs have surface expression of
CD39 and although surface expression of CD73 on Tregs is insignificant, it is abundantly
present in the cytoplasm (Mandapathil et al., 2010).
Cytokine Depletion.
nTregs constitutively express CD25, a subunit of IL-2R, but are unable to
produce IL-2 (Chinen et al., 2016). On this basis alone, researchers suspected that
consumption of IL-2 produced by Teff could be a major force behind Treg suppression
(Pandiyan et al., 2007; Thornton et al., 2004). However, equally many studies emerged
that disputed this view and have shown that IL-2 consumption is not critical for Treg
function. For example, in a transwell setting where a membrane separates Tregs from
Teff cells abrogates regulatory function of Tregs (Thornton and Shevach, 1998). Hofer et
al. argues that it likely all comes down to spatio-temporal logistics of IL-2 consumption
competition between Tregs and Teff. Depending how quickly activated T cells can
upregulate CD25 on their cell surface to consume autocrine/paracrine IL-2, within this
13
window Tregs may reduce activation by consuming IL-2 given they are in close
proximity and high density (Hofer et al., 2012).
Anti-inflammatory Cytokines.
Upon activation, nTregs produce an arsenal of potent anti-inflammatory cytokines
with IL-10, TGF-β and IL-35 playing pivotal roles.
The role of TGF-β in immune homeostasis has been long known, as addition of
anti-TGF-β antibody abolishes suppression mediated by nTregs (Nakamura et al., 2001)
and mice deficient in TGF-β or molecules required for TGF-β signalling develop
systemic autoimmune disease (Li et al., 2006). However, Tregs themselves do not have
to produce TGF-β, because they can induce TGF-β production in other immune cells
(Kullberg et al., 2005). Furthermore, TGF-β produced by nTregs may also induce
expression of Foxp3 in Tconv cells rendering them regulatory (Chen and Konkel, 2010).
Inactive form of TGF-β is maintained on the surface of nTregs through binding with
latency-associated peptide (LAP) (Annes et al., 2003). This latent TGF-β is responsible
for converting Foxp3− to Foxp3+ cells upon TCR stimulation (Andersson et al., 2008).
IL-10 producing Tregs have been found indispensable in many in vivo models of
inflammation and homeostatic expansion, and have also been implicated in maintaining
the balance between pathogen elimination and immunopathology in viral, fungal and
parasitic infections (Couper et al., 2008). For example, administration of anti-IL-10 Ab
abrogates the ability of Tregs to inhibit intestinal colitis (Asseman et al., 1999), but the
suppression is dependent on the differentiation status of Teff cells with naïve Teff cells
being unresponsive to IL-10 (Asseman et al., 2003). Interestingly, nTregs may control
production of IFN-γ by Th1 cells without supressing Th1 differentiation via IL-10, but
not TGF-β or IL-35 (Sojka and Fowell, 2011).
Collison et al. identified IL-35 as novel inhibitory cytokine, which is produced by
murine nTregs and is required for maximal suppression of proliferation of responder cells
(Collison et al., 2007). In vitro, nTregs from mice deficient in the genes encoding for
subunits of IL-35: IL-12α and IL-27β were unable to supress responder cells and in vivo
14
they were unable to ameliorate inflammatory bowel disease. nTregs may also induce
naïve T cells to express IL-35 rendering them immunosuppressive. However, IL-35 is not
produced by human nTregs (Bardel et al., 2008).
1.3.2. CD4+ iTregs
In contrast to nTregs that develop in the thymus, Tregs may also be generated
extrathymically upon antigen stimulation in tolerogenic conditions. These induced Tregs
(iTregs), also known as adaptive Tregs, can be generated in peripheral lymphoid tissues
such as GALT, spleen, lymph nodes and even inflamed tissues, from naïve Tconv cells
that are often Foxp3−.
The de novo generation of functional iTregs has been observed in cases where the
antigens are encountered in the presence of TGF-β. To generate iTregs in vivo antigens
can be delivered through intravenous injection, oral administration, administration of
non-depleting CD4 antibodies or tolerogenic DCs (Schmitt and Williams, 2013). The
minimal requirement of iTreg generation in vitro is TCR stimulation of naïve CD4+ T
cells in the presence of TGF-β and IL-2. Additionally, retinoic acid, a vitamin A
metabolite, promotes development and function of CD4+ iTregs.
Many iTregs share similar phenotype with nTregs, as they are also
CD4+CD25+Foxp3+ and share many of the mechanisms of suppression as nTreg cells.
The extent to which these two populations have redundant versus complimentary roles in
the immune system remains unclear (Bilate and Lafaille, 2011). However, the two
populations have different TCR β-chain repertoires with the overlap between the two
being 10-42% suggesting different functions of these two types of Tregs (Haribhai et al.,
2011). There is a popular notion that the major function of nTregs is to prevent
autoimmunity, whereas that of iTregs is to raise the activation threshold of immune
responses directed against innocuous environmental antigens (Curotto de Lafaille and
Lafaille, 2009). However, both nTregs and iTregs were required to induce complete
tolerance in the mouse model of colitis (Haribhai et al., 2011), and both played active
15
roles in asthma tolerance (Huang et al., 2013). Generation of iTregs in the periphery may
also be one of the main barriers to the eradication of tumours and clearance of pathogens.
There are also many iTreg types that do not express Foxp3 transcription factor.
These include Th3, Tr1, most of CD8+ Tregs, and a unique population of
TCRαβ+CD4−CD8− cells. Other types of cells that may have a regulatory function under
certain conditions are TCRγδ+ T cells and NKT cells, but these will not be further
discussed.
1.3.2.1. Th3 Cells
A unique subset of TGF-β-secreting CD4+ cells was identified in the mucosal
tissues of mice and humans during studies concerning identifying mechanisms of oral
tolerance, and these cells were termed Th3 cells (Chen et al., 1998a; Fukaura et al., 1996;
Inobe et al., 1998). Although the developmental lineage of Th3 cells is still unclear, they
can be induced in the gut by DCs via TCR signalling and CD86 co-stimulation (Weiner,
2001b). Th3 induction can be further enhanced by the presence of TGF-β, IL-4, IL-10 or
anti-IL-12 (Inobe et al., 1998; Seder et al., 1998). Similarly to nTregs, Th3 cells
constitutively express CTLA-4 on their cell surface, initiation of which results in the
production of copious amounts of TGF-β, which in turn suppresses activation and
proliferation of Th1 and Th2 cells, and provides help for IgA production (Chen et al.,
1998b). The non-specific suppression mediated by Ag-specific Th3 cells gave rise to the
term ‘bystander suppression’ and is recognized as an important mechanism in the
induction of tolerance to exogenous antigens from food proteins and bacterial flora
(Weiner, 2001a).
1.3.2.2. Tr1 Cells
Type 1 Tregs (Tr1) are also a subset of CD4+, however, they do not express
Foxp3. Tr1 cells can be distinguished from other CD4+ cells by the co-expression of
LAG-3 and CD49b, in both mice and humans (Gagliani et al., 2013). Tr1 cells inhibit
immune-mediated inflammation via cytokine-dependent mechanisms by secreting
16
abundant amounts of anti-inflammatory IL-10, which diminishes the function of APCs
and Ag-specific Teff cells (Gagliani et al., 2013). Additionally, human Tr1 can
selectively kill APCs via a perforin/granzyme B-dependent mechanism that requires
recognition of MHC class I, CD2, and CD226 (Magnani et al., 2011). Similar to nTregs,
Tr1 cells can inhibit T cell responses in a cell contact-dependent manner facilitated by
CTLA-4 and PD-1 (Akdis, 2008), and by the disruption of metabolism of Teff cells via
production of the ectoenzymes CD39, and CD73 (Bergmann et al., 2007).
Since Tr1 cells belong to a family of inducible Tregs they can be induced in vitro
by the chronic antigen stimulation in the presence of IL-10 (Groux et al., 1997). In vivo,
Tr1 cells were generated in diabetic mice that underwent pancreatic islet transplantation
by exogenous administration of IL-10 in conjunction with rapamycin (Battaglia et al.,
2006a). Additionally, IL-27 production by tolerogenic DCs is crucial for in vivo
differentiation of Tr1 cells in both mice (Awasthi et al., 2007; Fitzgerald et al., 2007;
Stumhofer et al., 2007) and humans (Murugaiyan et al., 2009). Interestingly, IL-6 in the
absence of IL-27 also drives differentiation of murine Tr1 cells from naïve CD4+ T cells
in vitro (Jin et al., 2013).
In 1994 Bacchetta et al. published the first clinical evidence regarding the
tolerogenic role of Tr1 cells in severe combined immunodeficiency patients who
underwent HLA-mismatched fetal liver hematopoietic stem cells transplant (HSCT)
(Bacchetta et al., 1994). High endogenous production of IL-10 by Tr1 cells contributed to
the development of split chimerism characterized by the presence of T and NK cells of
donor origin, whereas B cells and APCs were host-derived. Interestingly, this split
chimerism was devoid of GVHD. Serafini et al. and Andreani et al. also have shown that
Tr1 cells were involved in induction of mixed chimerism in patients undergoing HSCT
for the treatment of thalassemia major and played a role in sustaining long-term allograft
tolerance despite the chronic donor-host allo-Ag stimulation (Andreani et al., 2014;
Serafini et al., 2009).
17
1.3.3. CD8+ Tregs
Gershon’s original suppressor T cells were defined as CD8+ (Gershon and Kondo,
1971). Thus in the 1970s and 1980s this population of Tregs was the most extensively
studied. However, since the discovery in 1995 of highly potent CD4+CD25+Foxp3+ cells,
there had been reluctance in the scientific community to re-explore the regulatory role of
CD8+ Tregs, particularly because nTregs suppress activation and proliferation of CD8+ T
cells (Shevach, 2006). Despite the decrease in interest to study these cells, several
subpopulations of CD8+ Treg have been characterized in the past 40 years (Ligocki and
Niederkorn, 2015). Many of the CD8+ subtypes share functional and phenotypic
similarities to nTregs, and these may include (i) CD8+CD122+ Tregs (Rifa'i et al., 2004);
(ii) CD8+Foxp3+ Tregs (Mahic et al., 2008); (iii) CD8+LAG-3+Foxp3+CTLA-4+ Tregs
(Boor et al., 2011); (iv) CD8+Foxp3+CCR7+ (Wen et al., 2016); (v) CD8+IL-
10+CCR7+CD45RO+ Tregs (Wei et al., 2005) and (vi) CD8+CD122+PD-1+ Tregs (Dai et
al., 2010) and many more.
Because of the highly diverse population of CD8+ Tregs, it should be no surprise
that CD8+ Tregs may either develop in the thymus or differentiate in the periphery from
naïve T cells. Murine and human CD8+CD28low Tregs arise in the thymus, despite the
fact that the lack of CD28 expression on their cell surface that would otherwise suggest
differentiation of chronically activated Tconv cells to Tregs in the periphery
(Vuddamalay et al., 2016). In vitro, human CD8+ Tregs were induced from naïve
CD8+CD25−CD45+ T cells by co-culture with IL-2 and TGF-β, or by stimulation with
allogeneic mDCs even in the absence of these cytokines. However, the combination of
IL-2 and TGF-β yielded the most potent suppressive phenotype (Bjarnadottir et al., 2014).
Murine CD8+ Tregs may also be induced in the presence of retinoic acid in addition to
the aforementioned conditions (Lerret et al., 2012).
CD8+ Treg suppression is Ag-specific and may be mediated by contact-dependent
mechanisms, or via secretion of immunosuppressive molecules and inhibitory cytokines
such as IL-10 and TGF-β (Endharti et al., 2005; Lerret et al., 2012; Zhang et al., 2009).
Contact-dependent suppression involves the direct killing of CD4+ T cells by perforin-
18
mediated cytotoxicity (Lu and Cantor, 2008), or by induction of apoptosis via Fas/FasL
interaction (Chen et al., 2013), as well as direct inhibition via inhibitory molecules
CTLA-4 or PD-1 (Boor et al., 2011). Interestingly, CD8+ Tregs can modify the
suppressive mechanisms they use based on whether or not they also interact with CD4+
Teff cells. Upon contact with Teff, CD8+ Tregs predominantly produce IFN-γ and Fgl-2,
whereas in the absence of contact with CD4+ cells, CD8+ Tregs induce Indoleamine 2,3-
dioxygenase (IDO) (Li et al., 2010). IDO is the first and rate-limiting enzyme in
tryptophan catabolism, thus causing depletion of tryptophan, which is essential for
protein synthesis to sustain life. In the recent report by Wen et al. new suppressive
mechanism of human CD8+Foxp3+CCR7+ Tregs has been identified that restrains
activation and proliferation of CD4+ T cells (Wen et al., 2016). CD8+ Tregs release
NADPH oxidase 2 (NOX2)–containing microvesicles into target CD4+ cells, which in
turn inhibit TCR signal transduction by increasing ROS and thus reducing
phosphorylation of the TCR-associated kinase ZAP70. Furthermore, donor-specific
CD8+Foxp3+ Tregs facilitated de novo generation of CD4+Foxp3+ Treg via a TGF-β
dependent mechanism in a phenomenon known as ‘infectious tolerance’ (Lerret et al.,
2012).
In humans, CD8+CD25+Foxp3+ Tregs may be involved in the prevention of
asthma (Eusebio et al., 2015). Moreover, in patients with systemic lupus erythematous
(Zhang et al., 2009), inflammatory bowel disease (Brimnes et al., 2005), or multiple
sclerosis (Baughman et al., 2011) defective functions and/or reduced numbers of
circulating CD8+ Tregs have been reported. Additionally, CD8+ Tregs may be one of the
key players in immunoregulatory processes to restrict recognition of foreign antigens in
the immune privileged anterior chamber of the eye and cornea (Niederkorn, 2006), may
play a major role in allograft tolerance in the kidney (Derks et al., 2007) and liver-
intestine (Sindhi et al., 2005), and they have been associated with fewer rejection
episodes in the heart-transplanted patients (Dijke et al., 2009).
19
1.3.4. DN Tregs
1.3.4.1. Phenotype of DN Tregs
In addition to Tregs mentioned above there are T cells that express TCRαβ, but do
not express CD4 or CD8 co-receptor, nor the conventional NK markers NK1.1 (mouse),
or CD56 (human). These TCRαβ+CD4−CD8− cells, known as double negative T
regulatory cells (DN Tregs) have been shown to play an important immunoregulatory
function in the variety of settings. Murine and human DN Tregs can be found in
lymphoid and non-lymphoid tissues such as thymus, spleen, lymph nodes, skin, bone
marrow, peripheral blood and lung (Reimann, 1991). DN Tregs also constitute a
substantial component of TCRαβ+ cells in the intestinal epithelium (Hamad, 2010) and
compose 70-90% of TCRαβ+ cells in the murine female genital tract (Johansson and
Lycke, 2003). DN Tregs remain a rare population of cells in the peripheral blood and
represent only 1-5% of total circulating CD3+ lymphocytes in mice and 1-2% in humans
(Fischer et al., 2005; Zhang et al., 2001).
The first phenotypic and functional characterization of human DN Tregs by
Fischer et al. revealed many similar properties to their murine counterparts (Fischer et al.,
2005). DN Tregs isolated from peripheral blood consist of naïve, as well as Ag-
experienced cells, and upon stimulation by allo-DCs they acquire effector memory
phenotype (Fischer et al., 2005; Voelkl et al., 2011). Although freshly isolated DN Tregs
are negative for the activation marker CD25, activation can induce its expression (Fischer
et al., 2005). Furthermore, both human and murine DN Tregs do not express Foxp3 or
CTLA-4, the two markers that are associated with other types of Tregs (Fischer et al.,
2005; Gao et al., 2011; Hillhouse et al., 2010). In contrast to NKT cells, DN Tregs carry
a polyclonal TCR repertoire and do not express CD16 or CD56 (Fischer et al., 2005).
Upon stimulation with αCD3/CD28 mAbs or allogeneic DCs, DN Tregs secreted IFN-γ,
IL-5, IL-4, IL-10 but no IL-2 (Fischer et al., 2005; Voelkl et al., 2011).
DN Tregs were shown to display high proliferative potential, which was further
enhanced by supplementing the culture medium with IL-2 (Fischer et al., 2005). Analysis
20
of T-cell receptor excision circles (TRECs) revealed that DN Tregs had undergone a
higher number of cell divisions as compared to CD4+ and CD8+ T cells, suggesting that
DN Tregs “are not recent thymic emigrants but rather an expanded T-cell subset”
(Fischer et al., 2005).
1.3.4.2. Overview of DN Treg Function
The first evidence that TCRαβ+ T cells that do not express CD4 and CD8 co-
receptors may have suppressive function comes from a study conducted by Strober et al.
in 1989. In this study, clonally expanded DN T cells obtained from murine spleens were
able to successfully suppress responder cells in allogeneic mixed lymphocyte reaction
(Strober et al., 1989). However, one of the major limitations of the study was the lack of
staining for NK markers, thus, it may have been possible that NK cells were responsible
for the observed suppression. However, a subsequent study by Bruley-Rosset et al.
showed that DN Tregs were responsible for protection against GVHD if the host was pre-
immunized with donor spleenocytes before the minor histocompatibility antigen (MHA)-
mismatched bone marrow transplant (Bruley-Rosset et al., 1990). Although the
suppressor cells were shown to have low expression of NK marker asialo-GM1, the
possibility that they may be NK cells can be excluded, because depletion of asialo-GM1+
cells removed all NK activity in vitro, but only slightly affected suppression of GVHD in
vivo.
In 2000, Zhang’s group was the first to phenotypically identify DN Tregs as
TCRαβ+CD4−CD8−NK1.1− cells that display regulatory function in vitro and in vivo
(Zhang et al., 2000). Since then, multiple rodent models had been adapted to study DN
Treg function in vivo. The adoptive transfer of DN Tregs induced donor-specific
tolerance to allogeneic islet, skin, and heart, as well as xenogeneic heart grafts (Chen et
al., 2003b; Chen et al., 2007; Chen et al., 2005; Ford et al., 2002; Lee et al., 2005; Ma et
al., 2008; Zhang et al., 2007; Zhang et al., 2000); infusion of DN Tregs attenuated
GVHD (He et al., 2007; Juvet et al., 2012; Young et al., 2003a); and DN Tregs
successfully controlled autoimmune diabetes (Duncan et al., 2010; Ford et al., 2007;
Hillhouse et al., 2010; Liu et al., 2016), lymphoproliferative syndrome (Ford et al.,
21
2002; Juvet et al., 2012) and infection (Cowley et al., 2005; Hossain et al., 2000;
Johansson and Lycke, 2003; Kadena et al., 1997).
1.3.4.3. Role of DN Tregs in GVHD
GVHD is a common, life-threating complication that occurs after hematopoietic
stem cell transplant (HSCT). GVHD follows when donor T cells recognize the host
antigens as non-self. This results in the activation and rapid expansion of donor T cells
that culminates in severe organ damage (further discussion of GVHD in chapter 1.4.2.).
Adoptive transfer of murine DN Tregs has been shown to prevent GVHD after bone
marrow transplant (BMT) (Juvet et al., 2012; Young et al., 2003a). Furthermore, infusion
of murine DN Tregs after allogeneic BMT promoted induction of tolerance to donor
antigens and establishment of mixed chimerism in the absence of GVHD (He et al.,
2007).
Perhaps the most important clinical evidence suggesting the value of harnessing
DN Tregs for immunotherapy comes from a study conducted by McIver et al. (McIver et
al., 2008) that followed a group of 40 patients who received allogeneic HSCT.
Significant difference was observed in the percentage of DN Tregs in peripheral blood of
patients who developed GVHD and those who did not. Interestingly, the inverse
correlation between the severity of GVHD (grade 1-4) and the percentage, as well as the
absolute number of DN Tregs was observed. As a matter of fact, all patients whose DN
Tregs expanded to more than 1% of peripheral T cells did not develop GVHD, strongly
suggesting that the expansion of DN Tregs may prevent the development of GVHD after
allogeneic HSCT. Additionally, the deficiency of DN Tregs was concomitant with
increased clonal T cell expansion, which in turn was positively correlated with the
occurrence of GVHD. Although the study did not offer any mechanistic explanation as
the role of DN Tregs in suppressing allo-reactivity and thus the results represent
correlation and may not necessarily mean causation. Nevertheless, before the publication
of this study only the nTregs were shown to correlate inversely with GVHD in humans
(Hess, 2006).
22
In a separate study, Ye at al. examined reconstitution of DN Tregs after allo-
HSCT (Ye et al., 2011). In patients that received HLA-mismatched transplant, the rate of
DN Treg reconstitution was reversely correlated with acute GVHD. Thus the patients
who did not develop GVHD were characterized by significant expansion of DN Tregs in
the peripheral blood. Moreover, the absolute counts of DN Treg 60 days post allo-HSCT
and the grade of GVHD were inversely correlated – results similar to those observed by
McIver et al. (McIver et al., 2008).
1.3.4.4. Role of DN Tregs in Cancer
Apart from the tolerogenic mechanisms, DN Tregs demonstrated anti-tumour
activity, even though the notion is counterintuitive to the established understandings of
Treg function. In mice, adoptive transfer of allogeneic DN Tregs has been shown to
prevent A20 lymphoma tumour growth without causing GVHD (Young et al., 2001;
Young et al., 2003b). Merims et al. have shown that human DN Tregs can effectively
kill allogeneic and autologous CD34+ leukemic blasts in a dose-dependent manner
(Merims et al., 2011). In this study, CD3+CD4−CD8−CD56− cells were isolated from the
peripheral blood of acute myeloid leukemia patients during chemotherapy-induced
remission. These cells underwent two weeks of ex vivo expansion using two rounds of
stimulation with anti-CD3 monoclonal antibody and IL-2. Since both TCRαβ+ and
TCRγδ+ cells were present in the ex vivo expanded cohort, the cells were sorted based on
their TCR expression. Both subsets of TCRαβ+ and TCRγδ+ showed similar cytotoxicity
against primary leukemic blasts (Merims et al., 2011).
1.3.4.5. DN Tregs in Lymphoproliferative Syndrome
Although DN Tregs have been recognized for more than three decades, ill
understanding of their pathophysiological roles led to many misconceptions and general
dismissal among immunologists that these cells were legitimate constituents of the
immune system (Martina et al., 2015). Historically, DN Tregs have been associated with
lpr and gld mice that have the loss-of-function mutation in genes encoding Fas and FasL,
respectively (Cohen and Eisenberg, 1991) and will be referred to in this section as lpr
23
DN Tregs. Since FasL is a death factor that binds to its receptor Fas and induces
apoptosis in the Fas-expressing cells, the defects in either one lead to lymphadenopathy
and splenomegaly, respectively, and systemic autoimmune disease characterized by
immense lymphoproliferation, and accumulation of lpr DN Tregs (Shirai et al., 1990).
Some studies have shown that a large abundance of lpr DN Tregs in lpr and gld
mice may be the result of conventional T cells down-regulating their co-receptor
(Bristeau-Leprince et al., 2008). However, the lpr DN Tregs have abnormal phenotype in
comparison with wild-type DN Tregs: lpr DN Tregs do not respond to antigenic
stimulation (Davignon et al., 1988), express an unusual B-cell-specific CD45RA isoform
called B220 (Cohen and Eisenberg, 1991), and overexpress and are dependent on the
transcription factor Eomes (Kinjyo et al., 2010). However, even lpr DN Tregs have been
shown to have potent immunoregulatory function against allo-antigens in vivo and in
vitro (Ford et al., 2002), and that their functions is dependent on autocrine IFN-γ
secretion (Juvet et al., 2012). Together, these results indicate that DN Tregs wrongfully
had bad reputation within the scientific community.
1.3.4.6. DN Tregs in Diabetes
Diabetes mellitus is an autoimmune disease characterized by destruction of
insulin-producing beta cells in the pancreas by autoreactive T cells and autoantibodies,
which results in life-long dependence on insulin injections (Xie et al., 2014). If left
untreated, diabetes results in death. The first report that DN Tregs may also alleviate
autoimmune disease comes from study conducted by Ford et al. where they showed that
peptide-activated transgenic DN Tregs prevented development of autoimmune diabetes
by killing autoreactive CD8+ T cells (Ford et al., 2007). In another study, adoptive
transfer of splenic DN Tregs obtained from non-obese diabetic (NOD) mice provided
long lasting protection against diabetes (Duncan et al., 2010). Furthermore, transferred
DN Tregs proliferated and differentiated into IL-10 secreting Tr1-like cells (Duncan et al.,
2010). Lastly, adoptive transfer of DN Tregs converted from CD4+ cells in combination
with anti-thymocyte serum administered to deplete pathogenic T cells have shown to
24
reverse autoimmune diabetes in NOD mice that recently developed diabetes (Liu et al.,
2016).
1.3.4.7. Mechanisms of DN Treg-mediated Suppression
DN Tregs can suppress immune responses via various mechanisms. Similar to
nTregs, murine DN Tregs suppress responder T cell proliferation in vitro and in vivo in a
cell contact-dependent manner (Sakaguchi et al., 2008; Zhang et al., 2000), may express
CTLA-4 on their cell surface and thus modulate co-stimulatory molecule CD80/CD86
expression on DCs (Gao et al., 2011; Wing et al., 2008), and require activation to carry
out their suppressive function. Unlike nTregs which can suppress immune responses in
an Ag-specific or non-specific manner, murine DN Tregs employ only Ag- or allo-Ag-
specific suppression, both in vitro and in vivo (Fischer et al., 2005; Gao et al., 2011;
Hillhouse and Lesage, 2013; Voelkl et al., 2011; Zhang et al., 2007). DN Tregs use an
interesting process to acquire antigen specificity called trogocytosis, in which DN Tregs
acquire allo-peptide-MHC complexes from APCs, which are then re-expressed on their
cell surface (Juvet and Zhang, 2012).
Mackensen’s group shed the light on the function of human DN Tregs, which in
many aspects is similar to their murine counterparts. In the first report by Fischer et al.
DN Tregs were shown to acquire allo-antigen peptide from APCs, and supress CD8+
responder cells activated with the same peptide (Fischer et al., 2005). In a subsequent
report, Voelkl et al. extended suppression of allo-Ag-specific DN Tregs to autologous
CD4+ and CD8+ T cells activated by allo-APCs or anti-CD3/CD28 microbeads (Voelkl et
al., 2011). However, the mechanism of suppression contrasted those observed in murine
models. Murine DN Tregs mediate suppression of responder T cells by eliminating them
via Fas/FasL interaction or perforin/granzyme B pathway (Chen et al., 2003a; Ford et al.,
2002; Young et al., 2002; Zhang et al., 2007; Zhang et al., 2006; Zhang et al., 2000);
however, blocking the same pathways in human DN Tregs failed to abolish their
suppressive capacity and, unlike murine DN Tregs, human DN Tregs did not induce
apoptosis in the responder T cells, even though the first report suggested otherwise
(Fischer et al., 2005). Even more striking is the fact that suppression was reversible as
25
responder T cells sorted from the mixed lymphocyte reaction (MLR) were able to
proliferate upon removal of DN Tregs (Voelkl et al., 2011). Unlike nTregs, suppressive
activity of DN Tregs is not modulated by the competition of growth factors or modulation
of APCs (Sakaguchi et al., 2008; Wing et al., 2008). However, DN Treg-mediated
allogeneic suppression is TCR-dependent, requires de novo protein synthesis and cell-to-
cell contact (Voelkl et al., 2011).
In the most recent study conducted by A. Mackensen group, the addition of IL-7
to MLR was found to impair suppressive function of DN Tregs and that the effects of IL-
7 were contributed to activation of Akt/ mechanistic target of rapamycin (mTOR)
pathway, which is directly related to cell proliferation, differentiation, and metabolism
(Allgauer et al., 2015). IL-7-treated DN Treg showed more activated phenotype,
increased proliferation and down-regulation of anergy-associated genes, which was
inversely correlated with their suppressive function, but can be reversed by selective
inhibition of Akt or mTOR protein (Allgauer et al., 2015). The researchers suggest that
DN Tregs, similarly to Tr1 cells (Roncarolo et al., 2014), require an anergic phenotype
for their suppressive activity.
1.3.4.8. Development of DN Tregs
The origin and development of DN Tregs remain elusive. Based on the current
evidence, DN Tregs may develop in the thymus, in the periphery, or they may ascend
from Tconv cells. Given that DN Treg population is heterogeneous, it is likely that more
than one if not all of these pathways are plausible (Juvet and Zhang, 2012).
During T cell development thymocytes progress through a series of maturation
stages as defined by their phenotype. These include four double negative (DN) stages and
double positive (DP) stage before committing to a CD4+ or CD8+ lineage (Koch and
Radtke, 2011). Therefore, DN Tregs may represent a subset of thymic DN precursors
that avoided development into single positive Tconv cells by escaping clonal deletion
(Egerton and Scollay, 1990), or they may represent thymic cells that went through all the
stages of the development and downregulated both co-receptors at the DP stage (Landolfi
26
et al., 1993). Based on TREC counts, human DN Tregs appear to have proliferated to a
larger extent than Tconv cells (Fischer et al., 2005), which supports the notion of thymic
development followed by expansion in the periphery. However, DN Tregs in the murine
genital tract that represent 70-90% of the total pool of T cells appear to have completely
developed extrathymically (Johansson and Lycke, 2003).
DN Tregs may also be generated de novo in the periphery from CD4+ or CD8+
Tconv cells that have downregulated its co-receptor (Petrie et al., 1990). For example,
murine DN Tregs can be generated by stimulating CD4+ T cells with allogeneic DCs in
the presence of IL-2 or IL-15 (Zhang et al., 2007). The disappearance of CD4 molecule
was a result of gene silencing and resulted in stable linage even after re-stimulation.
These cells were potent suppressors and resistant to AICD. Crispin et al. demonstrated
that the subset of human DN Tregs could derive from transformed CD8+ T cells, which
have down-regulated the CD8 co-receptor (Crispin et al., 2008; Crispin and Tsokos,
2009). However, Ford et al. has shown that murine DN Tregs can arise in the absence of
CD8+ or thymus, and that antigen stimulation of CD8+ cells in vivo does not convert
CD8+ cells to DN Tregs (Ford et al., 2006).
1.4. Immunotherapy with Tregs
In organ transplantation and autoimmune diseases the in vivo induction of
proliferation or adoptive transfer of Tregs would be beneficial, whereas in cancer setting
Treg depletion and/or functional blockade of Tregs would enhance the immune responses
against tumour antigens. The idea that Tregs could be used for immunotherapy was first
proposed by Gershon back in the 1970s (Gershon, 1975). Since then the efficacy of Tregs
has been shown in many preclinical models and the improvements in the methods for ex
vivo expansion of Tregs over the past years made it possible to use Tregs in adoptive
cellular therapy (ACT). In general, ACT involves isolation of cells of interest from a
patient or donor, followed by ex vivo expansion of the cellular product and reinfusion
back into the patient. Understanding the role of the different subset of Tregs in each
27
disease will allow for the development of highly effective immunotherapies specifically
tailored to combat particular illness.
1.4.1. Treg Expansion Methods
There are many steps involved in the successful expansion of Tregs and each step
comes with its own unique challenges. The first problem lies in obtaining immune tissue
from which Tregs could be isolated. For this purpose, peripheral blood of healthy adults,
banked umbilical cord blood (UCB) and discarded human thymi can serve as Treg
sources (Dijke et al., 2016; Singer et al., 2014). The next challenge involves
identification and isolation of Tregs from the bulk population. Unfortunately, the task is
challenging, because there are many phenotypes of Tregs (as discussed in the earlier
chapters) and no definitive cell surface markers that would distinguish Tregs from Tconv
have been identified. Therefore, staining for multiple cell surface markers is required and
the combination of the selected markers would ensure a relatively pure population of
Tregs. These cells can then be isolated by magnetic sorting, which is probably the most
common method employed by the researchers, or via fluorescence activated droplet cells
sorting, microfluidics switch device, or microchip based sorting (Trzonkowski et al.,
2015). The last step involves expansion of these cells in culture and as with any Tconv
cells it includes activation and provision of growth factors, and nutrients. The easiest
method is to use anti-CD3/CD28 beads, which will generate a population of polyclonal
Tregs due to non-specific TCR stimulation. However, generation of Ag- or allo-Ag-
specific Tregs would significantly reduce the numbers of Tregs that need to be infused
into a patient, as they would be more potent as seen in the pre-clinical studies (Golshayan
et al., 2007; Sagoo et al., 2011).
1.4.1.1. Role of IL-2, IL-7 and IL-15 Growth Factors in Treg Maintenance and Proliferation
Interleukin 2 (IL-2), interleukin 7 (IL-7) and interleukin 15 (IL-15) belong to the
family of common gamma chain (γc) cytokines that are essential growth factors for T cell
development and homeostasis. All of these cytokines bind to receptors that use the γc-
28
receptor (CD132) to signal through two major signalling pathways the JAK-STAT
pathway and the phosphoinositide 3 kinase (PI3K)-AKT pathway to induce expression of
genes associated with T cell survival and proliferation. Mutations in the genes associated
with γc-receptor lead to severe combined immunodeficiency (SCIDX1) characterized by
severe reduction in circulating peripheral T cells, severe infections and patients urgently
need BMT within first year of life (Di Santo et al., 1995).
IL-2 has been long known as one of the most important cytokines for adaptive
immune response and has been the first cytokine used for propagation of T cells in vitro
(Smith, 1988). Therefore, it has been baffling that the absence of IL-2 does not interfere
with proliferation, or composition of T cells in the periphery (Schorle et al., 1991), but
rather hyperactivation of CD4+ cells (Sadlack et al., 1993). Mice deficient in IL-2Rα
(CD25), or IL-2Rβ (CD122) also succumb to hyperplasia and systemic inflammation
(Suzuki et al., 1995; Willerford et al., 1995). The paradox that IL-2 seems to limit rather
than enhance immune responses has been resolved with the discovery that IL-2 is critical
for Treg development and expansion (Nelson, 2004; Sakaguchi et al., 2008). Since
nTregs express high levels of CD25 and dysregulation of IL-2, or its receptor, is
associated with autoimmunity, it was reasonable to predict that administration of low
doses of IL-2 may have positive effect on the function and expansion of Tregs in the
periphery (Klatzmann and Abbas, 2015). Several clinical trials have shown promising
results in reducing incidence of GVHD (Kennedy-Nasser et al., 2014; Koreth et al.,
2016), T1D (Hartemann et al., 2013) and SLE (He et al., 2016).
Under biological conditions, IL-7 regulates homeostasis of naïve and memory T
cells and preserves T cell repertoire diversity, as it is responsible for V(D)J
rearrangement during early T cell development (ElKassar and Gress, 2010). IL-7 is
continuously produced by epithelial and stromal cells in the thymus and bone marrow,
and by fibroblastic reticular cells in the T cell zones of secondary lymphoid organs
(Rochman et al., 2009). IL7-Rα (CD127) is downregulated upon T cell activation, thus
most likely IL-7 does not have an effect on activated T cells. This in turn decreases IL-7
consumption by activated T cells increasing the availability of IL-7 for other cells poised
to receive survival signals. IL-7 impacts T cells homeostasis in at least two different ways
29
(Rochman et al., 2009): first, IL-7 promotes survival of T cells by activating PI3K-AKT
signalling pathway, increasing the expression of survival factors and inhibiting the
expression of the pro-apoptotic factors (Mazzucchelli and Durum, 2007; Surh and Sprent,
2008); and second, levels of circulating IL-7 are enhanced in lymphopenic conditions
(Surh and Sprent, 2008) and it may promote expansion of T cells in tumours (Dummer et
al., 2002), GVHD (Dean et al., 2008) and autoimmunity (Katzman et al., 2011; Monti et
al., 2008). But even in the autoimmune setting IL-7 signalling had positive effect on
Tregs as seen in an experimental model of autoimmune encephalomyelitis (EAE) in
which IL-7 signalling pathway has been found to drive accelerated differentiation and
proliferation of Tregs in the thymus leading to increased output of thymic and self-
regulation of EAE (Chen et al., 2009). Although one of the characteristics of Tregs is low
expression of IL7Rα, low levels of IL-7 in the periphery are required for Treg survival
and to support Foxp3 expression to sustain CD25 expression on the Treg cell surface in
vivo and modulate the ability of Tregs to efficiently bind IL-2, and transduce IL-2
signalling (Kim et al., 2012; Simonetta et al., 2014).
Although IL-15 induces similar signalling pathways as IL-2 or IL-7, its role in
Treg homeostasis is not well established. However, beneficial effects of IL-15 on the
expansion of Tregs have been reported previously. IL-15 has been found to promote de
novo generation of Tregs in the thymus and to enhance their regulatory function (Ben
Ahmed et al., 2009). In cell culture, IL-15 has been used in expansion of nTregs from
UCB (Asanuma et al., 2011) and adult peripheral blood, however, it offered no advantage
as compared to IL-2 alone (Lin et al., 2014). Additionally, IL-2 along with IL-15 was
used for the expansion of allo-specific Tregs (Veerapathran et al., 2013).
1.4.1.2. Role of Rapamycin in Treg Expansion and Suppressive Function
The mammalian target of rapamycin (mTOR) is an evolutionary conserved 289-
kDa serine/threonine protein kinase. mTOR is a key regulator of metabolism that
integrates environmental cues in terms of nutrients, energy and growth factors, and is
involved in cell growth, proliferation, and survival. The activation of mTOR is tightly
regulated, induced by multiple stimuli and signals through PI3K-AKT-dependent fashion,
30
which leads to an increase in glucose uptake, and a switch from oxidative
phosphorylation to glycolysis (Fox et al., 2005).
As the name suggests mTOR can be inhibited by rapamycin (sirolimus). Other
mTOR inhibitors are referred to as rapalogs (analogs of rapamycin). These potent
inhibitors of metabolism are commonly used as immunosuppressants to prevent organ
rejection. Rapamycin promotes generation of Tregs as evident in the induction of de novo
expression of Foxp3 in naïve T cells, thus it may be used to expand Tregs in vitro
(Battaglia et al., 2006b; Battaglia et al., 2005; Golovina et al., 2011). Addition of
rapamycin directly to nTreg culture also aids in their in vitro expansion, and these nTregs
have amplified suppressive function relative to non-treated cells (Battaglia et al., 2006b;
Battaglia et al., 2005; Golovina et al., 2011; Singh et al., 2012). Their improved
regulatory function is at least in part due to upregulation of CD25 and CTLA-4 on their
cell surface (Singh et al., 2012). Allgauer also reported that human DN Tregs increase
their suppressive activity when treated with mTOR or AKT inhibitors (Allgauer et al.,
2015).
1.4.2. Treatment of GVHD
One particular situation where the infusion of Tregs would be exceptionally
beneficial therapeutically is graft-versus-host disease. GVHD is a major complication
after allogeneic hematopoietic stem cell transplant (HSCT). HSCT is a life-saving
treatment for many haematological malignancies and is beneficial in the treatments of
benign disorders, such as autoimmune diseases. GVHD occurs when the donor T cells
recognize the recipient (the host) as ‘non-self’ and attack the recipient tissues. The
immune response results in potent inflammatory reaction, despite the routine
administration of immunosuppressive drugs. Systemic inflammation eventually exhausts
all the immunoregulatory mechanisms and culminates in severe organ damage. The
global incidence of GVHD is related to HLA disparity and the larger the degree of
mismatch, the more likely the occurrence of developing GVHD (Choi et al., 2010).
Although recent advances in HLA typing allow for a much safer procedure,
31
approximately half of the patients that undergo allo-HSCT will develop acute GVHD and
about half of those patients will eventually progress to chronic GVHD (Jacobsohn and
Vogelsang, 2007). Overall, more than 10% of GVHD cases will result in death (Jamil and
Mineishi, 2015). Worldwide, more than 20,000 allogeneic HSCT are performed annually,
and the number continues to increase mainly due to advances in the safety profile, such as
improved HLA matching, but also by extending HSCT for treatment of autoimmune
diseases (Burt et al., 2015). In contrast to solid organ transplants, recipients of HSCT will
ultimately develop systemic tolerance, because the donor’s lymphocytes and APCs will
eventually replace the host’s leukocytes. Since the risk of GVHD is the highest in the first
few months after HSCT, adoptive transfer of Tregs would be particularly suitable for the
prevention and treatment of acute GVHD. Moreover, the feasibility of Treg
immunotherapy is improved with the availability of Tregs from donors.
Trzonkowski et al. conducted the first-in-man clinical trial in the treatment of two
cases of GVHD (Trzonkowski et al., 2009). Tregs were isolated from the HSCT donors,
polyclonally expanded ex vivo and transferred to the recipients. The patient with chronic
GVHD significantly improved, whereas the patient with grade IV acute GVHD only
transiently experienced amelioration of the symptoms. The authors of this study argue
that the lack of significant improvability of the latter was most likely due to the relatively
late administration of Tregs.
In another Phase I dose-escalation clinical trial, 23 patients received a dose of 1-
3×107 Tregs/kg after partially HLA-matched UCB transplantation, which ensued in
significant reduction in the incidence of Grade II-IV GVHD in comparison to 108
controls (Brunstein et al., 2011). In another trial conducted by Di Ianni et al., the HSCT
transplant patients received ex vivo expanded Tregs together with Tconv cells isolated
from the HSCT donor (Di Ianni et al., 2011). Less than 10% of patients developed
chronic GVHD in the absence of any post-transplantation immunosuppression and
maintained graft-versus-leukemia (GVL) effect. In the subsequent Phase II clinical trial
Treg-Teff adoptive immunotherapy was found to prevent post-transplant leukemia
relapse and significantly decreased the incidence of GVHD, and out of 43 patients only
15% developed GVHD (Martelli et al., 2014).
32
1.4.3. Treatment of Autoimmune Diseases
There are over 80 identified autoimmune diseases and the most common include
rheumatoid arthritis (RA), type one diabetes mellitus (T1D), multiple sclerosis (MS) and
systemic lupus erythematous (SLE). In autoimmune diseases, immune cells recognize
self-antigens as foreign and thus attack own healthy cells. The incidence of 29
autoimmune diseases is rising globally with an estimated prevalence of 7.6-9.4% (Cooper
et al., 2009). Currently, there are no curative therapies for autoimmune diseases. Standard
treatment involves administration of immunosuppressive drugs, which attempt to
alleviate the symptoms by restraining the magnitude of immune response. However, the
use of immunosuppressants is associated with severe and debilitating long-term side
effects, such as increased risk of infection and cancer. The improved understanding of
pathophysiology of autoimmune disease allowed for the development of new therapeutic
interventions that are highly specific, with minimal off-target side effects.
In the first-in-man treatment of T1D with Tregs, recently diagnosed young
patients received either one or two doses of ex vivo expanded autologous Tregs (Marek-
Trzonkowska et al., 2014). Administration of Tregs was safe and effective, because in
two-thirds of patients the therapy prolonged survival of beta cells and reduced the
requirement for exogenous insulin. A clinical trial in newly diagnosed adult T1D patients
is ongoing.
More ongoing and scheduled clinical trials are underway involving infusion of
autologous/donor polyclonal/Ag-specific fresh/expanded Tregs, or Tr1 cells, for the
treatment of lupus, autoimmune hepatitis, acute/chronic/steroid dependent/refractory
GVHD, prevention of GVHD and complications after liver, and kidney transplantation
(Gliwinski et al., 2017). There are also many planned and ongoing clinical trials in a
large collaborative project named The ONE Study, which will evaluate safety, efficacy,
and feasibility of Treg therapies for SOT tolerance (Gliwinski et al., 2017; Schliesser et
al., 2012).
33
1.5. Hypothesis and Specific Research Aims
Adoptive cellular therapy using Tregs is one of the most promising non-
pharmacological approaches for prevention of GVHD, induction of tolerance to
transplanted organs and treatment of autoimmune diseases. Over the past few decades,
multiple Tregs populations have been identified. Although the different subtypes of Tregs
share many similarities, the subtle differences in the mechanisms of action and the
preferences for particular tissues make them more, or less suitable for different clinical
applications. So far the focus in the scientific community to harness the
immunoregulatory function of Tregs in the clinic has been mostly on CD4+CD25+Foxp3+
Tregs, as their phenotype, mechanism of action and their role in many diseases have been
more extensively studied in comparison to other Tregs. However, exploiting the Treg
type that has been associated with the improved outcome in certain diseases may lead to
the development of therapies tailored to the specific needs of patients - a hallmark of
personalized medicine.
The population of TCRαβ+CD4−CD8− NK lineage negative cells (DN Tregs) has
been identified in mice, rats, monkeys and humans, and has been shown in rodent models
to attenuate GVHD, control autoimmune diabetes, lymphoproliferative syndrome, and
infection, and to prolong survival of heart and skin allo-, and xeno-grafts. In humans, a
higher frequency of DN Tregs in the peripheral blood of patients undergoing HSCT has
been correlated with less severe GVHD. The current therapies for GVHD involve
systemic administration of immunosuppressive drugs. However, the risk of developing
GVHD remains high and the use immunosuppressive agents are associated with high
toxicity. Therefore, new treatments that would induce transplantation tolerance without
the debilitating toxicity are needed. Despite the extensive animal studies demonstrating
regulatory function of DN Tregs in various diseases, studying DN Tregs in humans has
been hampered due to their low frequency in peripheral blood and the lack of adequate
expansion method.
34
Therefore, I instigated a hypothesis that human DN Tregs can be expanded ex
vivo and exhibit regulatory function both in vitro and in vivo. The specific objectives of
this study were:
1. To develop a protocol for ex vivo expansion of human DN Tregs. To date, there
was no established protocol for a large-scale ex vivo expansion of human DN
Tregs. Therefore, I tested a combination of variety of expansion methods
previously used for large-scale expansion of conventional and regulatory T cells.
Once the protocol for producing the largest yield of DN Treg was established, I
tested the addition of growth factors IL-7 and IL-15 to the expansion cell culture
to determine whether supplementation of these cytokines had any effect on the
number, or function of DN Tregs.
2. To determine the phenotype of expanded DN Tregs. The limited studies involving
human DN Tregs described the phenotype of activated Ag- or allo-Ag-specific
DN Tregs. In this study, I generated DN Tregs with polyclonal reactivity, thus it
was important to determine whether the ex vivo expanded DN Tregs had the
phenotype comparable to what had been previously observed, both in mice and
humans. To this end, I performed a detailed analysis of cell surface markers
expression pre- and post-expansion, as well as cytokine profile of the ex vivo
expanded cells.
3. To assess the function of expanded DN Tregs in vitro and in vivo. In mice and
humans, DN Tregs were found to suppress CD4+ and CD8+ T lymphocytes in an
Ag-specific manner. Furthermore, human DN Tregs were found to be cytotoxic
towards leukemic cells. Since the expanded DN Tregs have polyclonal specificity,
I evaluated whether DN Tregs can also mediate non-specific suppression. To test
this, I sorted and used autologous CD4+, CD8+ and CD19+ cells as responder cells
and ex vivo expanded DN Tregs as putative suppressors in the CFSE-based in
vitro suppression assay. Since ex vivo expanded DN Tregs exerted suppressive
activity in vitro, I adapted a xenogeneic GVHD model to assess their regulatory
35
function in vivo. Lastly, DN Treg cytotoxicity against leukemic and lung cancer
lines was evaluated by a flow-based killing assay.
4. To investigate the mechanisms of DN Treg-mediated suppression in vitro. It has
been shown that murine DN Tregs can suppress Ag-specific responses via various
mechanisms. The mechanisms of human DN Treg-mediated suppression of CD4+
and CD8+ T cells have not been fully dissected, and seem to differ from their
murine counterparts. Therefore, I first determined whether any soluble factors
possibly identified in Objective 3 would mediate regulatory function. Next, I
determined whether cell-to-cell contact is important. Since there has not been a
consensus regarding the ability of human DN Tregs to induce apoptosis in
responder cells, I also attempted to determine whether ex vivo expanded DN Tregs
suppressed responder cells by killing.
5. To evaluate the effect of rapamycin treatment on DN Treg function in vitro and in
vivo. Treatment of nTregs, as well as human naïve or Ag-specific DN Tregs, with
mTOR inhibitors has been shown to increase their suppressive activity. Therefore,
I treated DN Tregs post ex vivo expansion with rapamycin, an mTOR-inhibitor, to
determine whether DN Treg with polyclonal specificity can be rendered more
potent. Since DN Tregs showed augmented regulatory function following
rapamycin treatment in vitro, the role of treated DN Tregs was also evaluated in
vivo and compared to untreated DN Tregs.
36
CHAPTER 2. MATERIALS AND METHODS
37
Chapter 2 Materials and Methods
2.1. Blood Samples
This study was approved by Research Ethics Board (REB# 05-0221-T).
Peripheral blood samples were obtained from 11 healthy individuals (6 females and 5
males) upon receiving informed consent from the study subjects. The volume of 50 to
100 ml of blood was collected in BD Vacutainer® blood collection tubes with sodium
heparin.
2.2. Cell Isolation and Magnetic Sorting
Mononuclear cells were isolated from peripheral blood by Ficoll-Hypaque density
gradient centrifugation. Cells recovered from the gradient interface were washed three
times in PBS, counted and immediately enriched for DN Tregs using magnetic-activated
cell sorting technology (MACS), according to the manufacturer’s instructions (Miltenyi
Biotec). Briefly, DN Tregs were purified from peripheral blood mononuclear cells
(PBMCs) with LD column by negative selection using FITC-conjugated monoclonal
antibodies (mAbs) directed against CD4, CD8, CD56 and TCRγδ cell surface markers,
and anti-FITC magnetic beads. The remaining monocytes temporarily played a role of
supporting cells. Purity ranged from 96% to 99% as measured by flow cytometry. For the
isolation of CD4+, CD8+ and CD19+ cells, positive selection was performed by direct
magnetic labeling and separation with LS column. Cells were washed, counted and
cultured immediately or cryopreserved for later use.
38
2.3. Freezing and Thawing of Cells
Cells were cryopreserved if they were not used immediately for cell culture. To
this end, cells were counted, centrifuged, gently resuspended in freezing medium (90%
fetal bovine serum (FBS) and 10% dimethyl sulfoxide (DMSO)) and aliquoted to
cryogenic storage vials. The cryogenic vials containing the cells were placed in an
isopropanol chamber and stored at −80°C overnight. The cells were transferred the next
day to the liquid nitrogen tank and stored in the gas phase above the liquid nitrogen.
To thaw frozen cells, the cryogenic vials containing cells were placed in water
bath at 37°C. The cells were then diluted slowly in preheated RPMI medium and
centrifuged at 400g for 6 min. The supernatants were carefully decanted without
disturbing the cell pellets and washed again to remove residual DMSO. The cells were
gently resuspended in the complete growth medium (RPMI media supplemented with
10% FBS, penicillin, and streptomycin).
2.4. Expansion of DN Tregs
Sorted cell fractions were resuspended in the complete growth medium
supplemented with recombinant human (rh) IL-2 (250 U/ml). To activate DN Tregs, 2-
3×106 cells/well were seeded in a 24-well tissue culture plate coated with 2.5 µg/well of
anti-CD3 mAb (OKT3, eBioscience). On day 3 of culture, cells were harvested, washed
and cultured for another 4 days in the presence of rhIL-2. DN Tregs were restimulated on
days 7, 12 and 17 with lethally irradiated (150 Gy) artificial antigen presenting cells
(aAPCs) (human K562 cell line with surface expression of a transduced membranous
form of anti-CD3 mAb, CD80, CD83, CD86 and 4-1BBL; a gift from Dr. Naoto Hirano)
(Butler et al., 2012) in the presence of rhIL-2 and/or rhIL-7 (10 ng/ml, PeproTech) and/or
rhIL-15 (10 ng/ml, PeproTech). DN Tregs were used for functional assay and phenotypic
studies on day 21. Viability and purity of DN Tregs were regularly assessed by flow
cytometry and the cells were purified via MACS if purity was found to be less than 90%.
39
2.5. Antibodies and Flow Cytometry
Cell surface staining
For investigation of the cell surface proteins by flow cytometry, cells were stained
with fluorochrome-coupled Abs. Briefly, cells were harvested, placed in FACS tubes and
washed twice with the staining media (PBS supplemented with 0.05% BSA). All
centrifugation steps were performed at 500g at 4°C for 5 min. The supernatants were
decanted, the pellet resuspended with the staining media and the amount of antibodies
specified by the manufacturer was added, followed by 15 min incubation at 4°C. After
the final wash step, cells were resuspended in 300 µl staining media or fixation buffer
(2% paraformaldehyde in PBS) and filtered. For some experiments, cells were stimulated
for 4 hours with Cell Stimulation Cocktail (2 µl/ml, eBioscience) containing PMA and
ionomycin.
Intracellular staining
For intracellular cytokine staining, cells were activated with Cell Stimulation
Cocktail (2 µl/ml, eBioscience) in the presence of monensin (1 µl/ml, GolgiStop, BD
Pharminogen) for 4 hours. After washing cells were stained for surface proteins, as well
as stained with fixable viability dye eFluor450 for live/dead recognition, fixed,
permeabilized (all reagents were included in the Fixation and Permeabilization Buffer Set,
eBioscience), and finally stained for intracellular cytokines.
The antibodies listed in Table 1 were used to label cells. Dead cells were excluded
with PI (Sigma-Aldrich), DAPI (Biolegend) or 7-AAD (eBioscience). Flow cytometry
data were acquired using LSR II (Becton Dickinson) or Accuri C6 (Accuri Cytometers),
and analyzed using FlowJo 10 software (Tree Star).
40
Table 1. List of the antibodies used in this study.
Fluorochrome Antibody Clone Company
FITC Anti-CD4 A161A1 Biolegend
Anti-CD8 HIT8a
Anti-CD25 BC96
Anti-CD45RO UCHL1
Anti-CD56 HCD56
Anti-CD127 A019D5
Anti-TCRγδ n/a Beckham Coulter
PE Anti-CD19 HIB19 Biolegend
Anti-CD45RA HI100
Anti-CD62L DREG-56
Anti-CD215 JM7A4
Anti-CD152 14D3 eBioscience
Anti-Granzyme B GB11
Anti-Perforin dG9
PE-Cy7 Anti-CD3 HIT3a Biolegend
Anti-CD56 HCD56
Anti-CD122 TU27
Anti-CD197 n/a BD Pharmingen
PerCP/Cy5.5 Anti-CD8 HIT8a Biolegend
APC Anti-TCRαβ IP26 Biolegend
Anti-CD132 TUGh4
Anti-CD197 TG8/CCR7
Anti-CD279 eBioJ105 eBioscience
APC-Cy7 Anti-CD4 RPA-T4 Biolegend
Pacific Blue Anti-TCRαβ IP26 Biolegend
41
2.6. Detection of Cytokines and Chemokines Secreted by DN Tregs
DN Tregs and CD8+ T cells obtained from the same donors were grown using DN
Treg expansion protocol. On day 21, cells were stimulated for 4 hours with Cell
Stimulation Cocktail (eBioscience). Cell supernatants were collected and stored at −20°C,
thawed and processed using the Luminex® platform and the Bio-Rad Human Cytokine
27-plex Array (FGF basic, Eotaxin, G-CSF, GM-CSF, IFN-γ, IL-1β, IL-1Ra, IL-2, IL-4,
IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12 (p70), IL-13, IL-15, IL-17, IP-10, MCP-1
(MCAF), MIP-1α, MIP-1β, PDGF-BB, RANTES, TNF-α, VEGF).
For IL-10 and IFN-γ analyses, the same aliquots used for Luminex cytokine
assays, and the aliquots of supernatants collected from suppression assays after 4 or 5
days of co-culture were assayed by enzyme-linked immunosorbent assays (ELISA).
2.7. In Vitro T cell and B cell Suppression Assays
DN Treg suppressive activity was measured by quantifying inhibition of
proliferation of autologous lymphocytes under polyclonal stimulation, as previously
described (Mond and Brunswick, 2003; Venken et al., 2007). Briefly, purified CD4+,
CD8+ or CD19+ cells (responders) were labeled with 1µM CFSE (5(6)-
Carboxyfluorescein N-hydroxysuccinimidyl ester) and seeded in a 96-well U-bottom
plate (2.5×104 cells/well). T cells and B cells were stimulated with anti-CD3/CD28-
coupled (αCD3/CD28) magnetic beads (Dynabeads, Life Technologies) or F(ab’)2
fragment of goat anti-human IgM (Jackson Laboratories), respectively, in the presence or
absence of autologous DN Tregs at increasing effector-to-responder ratios. After 4 to 5
days of co-culture, cells were harvested and stained with Abs against different surface
markers to discriminate between responders and suppressors. CFSE signal of gated
responder cells was analyzed by flow cytometry. The suppressive capacity of DN Tregs
towards responders in co-culture was expressed as the relative inhibition of the
percentage of proliferating cells as assessed by CFSE dilution using the formula:
42
% 𝐼𝑛ℎ𝑖𝑏𝑖𝑡𝑖𝑜𝑛 𝑜𝑓 𝑝𝑟𝑜𝑙𝑖𝑓𝑒𝑟𝑎𝑡𝑖𝑜𝑛 = 100− % !"#$%&'"()%#* !"#! !" !"#$%
% !"#$%&'"()%#* !"#!!"# !" !"#$%×100
and expressed as mean ± SD of 3 replicates.
For the IFN-γ and IL-10 blocking studies, suppression assays were prepared as
described above, except that CD4+ and CD8+ cells were blocked for 15 min with human
serum (from human male AB plasma, Sigma-Aldrich) prior to co-culture. 10 µg/ml of
purified mAbs to IFN-γ (MD-1, Biolegend), IL-10 (JES3-9D7, Biolegend) or both were
added to the co-cultures. Purified rat IgG1, κ Isotype Ab (RTK2071, Biolegend) was used
as control.
2.8. In Vitro Suppression Assays with Rapamycin-treated DN Tregs
Ex vivo expanded DN were treated with 2 µM rapamycin (Sigma) added to the
culture media incubated at 37°C. After 2 hours, cells were harvested and washed three
times with large volume of warmed PBS to ensure rapamycin had been washed away
completely. Rapamycin-treated DN Tregs or untreated DN Tregs were used as suppressor
cells in the suppression assay described above, with the exception of shortened duration
to 3 days.
2.9. Transwell® Experiments
Transwell® experiments were performed in 24-well plates with 0.4 µm pore size
(Millipore). Purified CD4+ and CD8+ T cells (5×105) were CFSE labeled and cultured in
the bottom chamber with αCD3/CD28 beads. DN Tregs were placed directly into the
bottom chamber or placed in the top chamber with αCD3/CD28 beads. For some co-
culture experiments, CD4+ or CD8+ T cells were also added to the top chamber. After 4
days, the proliferation of responders was measured by CFSE dilution using flow
cytometry.
43
2.10. Lymphocyte Cytotoxicity Assay
Flow-based Assay
Naïve CD4+ T cells were activated with αCD3/CD28 beads for 4 days. Activated
cells were harvested, washed and co-cultured either alone or in the presence of varying
ratios of DN Tregs. After 24 hours, the presence of AnnexinV, an early apoptotic marker
was quantified on the surface of CD4+ T cells by flow cytometry.
51Cr-release Assay
Naïve CD4+ or CD8+ T cells were activated for 4 days with αCD3/CD28 beads.
Activated cells were labeled with 51Cr and co-cultured with DN Tregs for 4 or 22 hours.
The amount of 51Cr released was quantified using scintillation counter. The percentage of
specific lysis was calculated using the standard formula and expressed as the mean of
triplicates:
% 𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝑘𝑖𝑙𝑙𝑖𝑛𝑔 = 100× (!"#!$%&!'()*!!"#$%&$'#(! !"#"$%")(!"#$!%! !"#$ !!"#$%&$'#(! !"#"$%")
2.11. Cancer Cells Cytotoxicity Assay
To assess DN Treg-induced killing of cancer cell lines, a flow cytometry-based
killing assay was adapted in which target cells were stained with PKH-26 before
culturing with DN Tregs. Human leukemic and lung cancer cell lines were cultured for 2
and 16 hours, respectively, in triplicates, at increasing effector/target ratios. Cell death
was assessed by staining with AnnexinV-FITC, an early apoptotic marker, and 7-AAD, a
cell viability dye. The percentage of specific killing mediated by DN Treg in the PKH-
26-gated cell population was calculated by subtracting non-specific AnnexinV-FITC- and
7-AAD-positive target cells measured in targets co-cultured in the absence of DN Tregs
using the formula:
44
% 𝑆𝑝𝑒𝑐𝑖𝑓𝑖𝑐 𝑘𝑖𝑙𝑖𝑛𝑔 = 100×(% !!!"! !"!"#$! !!"#! !" !"#$
! ! % !!!"!!""#$%" !!"#!!"# !" !"#$! )
!""!% !!!"!!""#$%" !!"#!!"# !" !"#$!
2.12. Mice and Xenogeneic GVHD Model
Animal experiments were approved by University Health Network (UHN)
Animal Care Committee (AUP#322.21).
NSG mice were obtained from Jacksons Laboratory and bred in the pathogen-free
colony at UHN Animal Facility. 6 to 8 week old mice received whole body irradiation
(250 cGy) from 137Cs source 24 hours prior to tail vein injection. For xenogeneic GVHD
induction, mice were injected with 5×106 freshly isolated human PBMCs in 100 µl of
PBS using an insulin syringe. Treated mice received either 1 or 3 injections of 107
autologous DN Tregs on day 0, or on day 0, 3 and 7, respectively. Some mice were
injected with only 107 DN Tregs, or with PBS. All mice were monitored daily for the
signs of GVHD including weight loss, hunched posture, ruffling of the fur, diarrhoea,
reduced mobility or tachypnea. At the time of severe GVHD, defined as weight loss
greater than 20% of initial weight, or presence of two or more symptoms described above,
mice were sacrificed in accordance to the local animal welfare regulations, and an end
point of survival was recorded.
2.13. Monitoring of Lymphocyte Migration and Proliferation In Vivo
Ex vivo expanded human DN Tregs or freshly isolated PBMCs were labeled with
5 µM CFSE. NSG mice were irradiated at 250 cGy. 5×106 CFSE labeled DN Tregs or
PBMCs were intravenously injected into the lateral tail vein 24 hours later. At days 1, 3,
5, 7 and 10 post-injection, 2-3 mice per group were sacrificed. Peripheral blood, spleen,
lymph nodes, bone marrow, liver, lungs, and kidneys were harvested. Single cell
suspensions were obtained from all organs by crushing, and Ficoll-Hypaque
45
centrifugation was performed to obtain mononuclear cells. Cells from bone marrow were
obtained by flushing femurs and tibiae. The peripheral blood samples was depleted of red
blood cells using ACK lysis buffer. The cells were counted using Vi-CELL cell counter
(Beckman Coulter) and analyzed by flow cytometry.
2.14. Data Analysis
Graphical presentation and statistical analysis of data were performed with Prism
6 (GraphPad) or SPSS 22 (IBM). All results are presented as ‘means ± SD,’ unless
otherwise specified. Survival analysis was performed using the log-rank test. Results
were assessed for normal Gaussian distribution and then variance between the groups was
assessed with either Student’s t test or ANOVA, or the Mann-Whitney, or Kruskal-Wallis
test in the case of non-parametric data. p values were considered statistically significant,
when p<0.05, p<0.01, or p<0.001 represented in the figures as *, **, or ***, respectively.
46
CHAPTER 3. RESULTS
47
Chapter 3 Results
3.1. Frequencies and Phenotypic Analysis of TCRαβ+ CD3+ CD4−
CD8− DN Tregs in the Peripheral Blood of Healthy Adults
To validate that DN Tregs are naturally present in the peripheral blood of healthy
adults, PBMCs were collected from 4 donors and analyzed for the expression of cell
surface markers: TCRαβ, CD4 and CD8. As shown in Figure 1a the DN Treg population,
as identified by the lack of expression of CD4 and CD8 co-receptors, comprised on
average only 1.48 ± 0.54% (n=4, Figure 1b) of the total TCRαβ+ cells, which is similar to
what has been observed in the previous reports (Allgauer et al., 2015; Fischer et al.,
2005; Voelkl et al., 2011). To further characterize the phenotype of peripheral blood DN
Tregs cell surface expression of common activation markers, memory markers and
selected cytokine associated receptors were assessed by flow cytometry. As shown in
Figure 1c & d, DN Tregs have a slightly different pattern of activation and cytokine
receptor expression, when compared to CD4+ and CD8+ T cells obtained from the same
donor. Markedly, DN Tregs had significantly lower expression of CD62L (49.48 ±
15.11%), a homing receptor to secondary lymphoid organs rapidly shed after TCR
triggering or following lymph node migration (Chao et al., 1997; Kishimoto et al., 1990),
when compared to CD4+ T cells (87.28 ± 5.42%) and CD8+ T cells (75.38 ± 5.50%),
suggesting that DN Tregs represent a more Ag-experienced T cell lineage. DN Tregs had
significantly lower expression of CD25 (3.34 ± 2.03% vs. 9.47 ± 3.6%) and CD127
(77.45 ± 12.15% vs. 94.9 ± 1.01%), as compared to CD4+ T cells and significantly lower
expression of CD122 (11.93 ± 3.76% vs. 23.58 ± 5.42%) in comparison to CD8+ T cells.
48
Figure 1. Phenotypic characteristics of DN Tregs isolated from peripheral blood. (a) Representative histogram of CD4+, CD8+ or CD4−CD8− cell frequency in the TCRαβ+ population from one donor. The numbers refer to frequencies of cells in each quadrant. (b) Summary of percent expression of DN Tregs in peripheral blood from 4 different healthy donors, as defined by TCRαβ+CD4−CD8− phenotype. (c) Evaluation of the expression of cell surface proteins shown as a gating strategy from one donor. All cells were gated on 7-AAD−TCRαβ+ population and on CD4+, CD8+ or CD4−CD8− (DN Tregs). Each gate was based on FMO control. (d) Summary of the cell surface markers expression from 4 different donors. Bars represent mean ± SD percentage expression of indicated markers. *p<0.05. **p<0.01. *** p<0.001.
C
CD4 CD8
DN Treg
Control
CD25
D
0 10 20 30 40
DN Treg
CD8
CD4
% Expression CD122
CD4 CD8
DN Treg *
% Expression CD122
0 5 10 15 20
DN Treg
CD8
CD4
% Expression CD25
CD4 CD8
DN Treg
% Expression CD25
** *
CD4 C
D8
A
70 80 90 100
DN Treg
CD8
CD4
% Expression CD132
% Expression CD132
60 70 80 90 100
DN Treg
CD8
CD4
% Expression CD127% Expression CD127
*
0 2 4 6 8 10
DN Treg
CD8
CD4
% Expression CD215% Expression CD215
0 20 40 60 80 100
DN Treg
CD8
CD4
% Expression CD62L% Expression CD62L
*
***
CD62L
CD215 CD132
CD127
CD122
CD4 CD8
DN Treg
Control
B
0.0
0.5
1.0
1.5
2.0
2.5
Day 0
%T
CRαβ
(+)
CD
4(-
)CD
8(-
) 2.5 2.0
1.5 1.0
0.5 0.0
% T
CRαβ
+
CD
4−C
D8−
49
Memory is a hallmark of the acquired immune system as it results from clonal
expansion and differentiation of Ag-specific lymphocytes (Sallusto et al., 2004). Effector
memory T (TEM) cells migrate to peripheral tissues and display immediate effector
functions, while central memory T (TCM) cells home to secondary lymphoid organs and
have little to no effector function, but readily proliferate and differentiate into effector
cells in response to antigenic stimulation (Lanzavecchia and Sallusto, 2000). The two
common markers that differentiate between naïve and memory phenotypes are CCR7 and
CD45RO (Figure 2a). Therefore, naïve T cells can be considered as CCR7+ and
CD45RO−, TCM cells as CCR7+ and CD45RO+, while effector memory cells are CCR7−
and can be either CD45RO+ (TEM) or CD45RA+ (TEMRA). As seen in Figure 2b, the
majority of CD4+ and CD8+ T cells show a naïve phenotype (66.1% and 54%,
respectively). In contrast, only a small proportion of DN Tregs (15.1%) display a naïve
phenotype. The majority of DN Tregs can be considered Ag-experienced cells with TEM
and TEMRA phenotype (45.3% and 31.1%, respectively).
50
Figure 2. Cell surface expression of memory markers CCR7 and CD45RO on peripheral blood T cells. (a) Schematic of T cell differentiation stages from naïve to effector memory phenotype based on the surface expression of CCR7 and CD45RO. TCM, central memory T cells. TEM, effector memory T cells. TEMRA, effector memory CD45RA T cells. (b) Flow cytometry detection of T cell differentiation stages of CD4+, CD8+ or CD4−CD8− (DN Tregs) cells. A representative figure is shown. Similar results were observed in different donors.
DN Treg CD4+ CD8+
CD45RO
CC
R7
Naïve T cells
TCM cells
TEMRA cells
TEM cells
CD45RO
CC
R7
A
Figure 2
Figure 2. Cell surface expression of memory memory markers CCR7 and CD45RO on peripheral blood T cells. (A) Schematic of T cell differentiation stages from naïve to effector memory phenotype based on the surface expression of CCR7 and CD45RO. TCM, central memory. TEM, effector memory. TEMRA, effector memory RA. (D) Flow cytometry detection of T cell differentiation stages of CD4+, CD8+ or CD4-CD8- (DN Tregs) cells. Representative figure is shown. Similar results were observed in different donors.
B
51
3.2. Human DN Tregs Can Be Expanded Ex Vivo
It has been shown that human DN Tregs can be generated by stimulation with
allogeneic mDCs (Fischer et al., 2005). However, the process is complex and
therapeutically relevant numbers of DN Tregs cannot be generated. In order to explore
the potential of clinical use of DN Tregs, we developed a novel protocol that allows for
large-scale expansion of polyclonal human DN Tregs, with the steps summarized in
Figure 3.
Briefly, PBMCs were isolated from peripheral blood of healthy donors. The
population of CD4+, CD8+ and TCRγδ+ T cells, and CD56+ NK cells was depleted by
magnetic beads sorting to enrich for the DN Treg population (Figure 4a). From 50 to 100
ml of blood, we were able to obtain 5-25×105 DN Tregs (n=8). After sorting, cells were
activated with anti-CD3 mAb in the presence of rhIL-2, followed by co-culture with
lethally irradiated aAPCs every 5 days. aAPC delivered optimal stimulation through the
expression transduced anti-CD3 Ab on the cell surface, as well as co-stimulatory
molecules, such as CD80, CD83, CD86 and 4-1BBL. Expanded DN Tregs were
harvested and used for functional and phenotypic studies on day 21. The phenotype of
DN Tregs was assessed regularly to monitor for potential outgrowth of CD4+, CD8+,
CD56+ or TCRγδ+ cells, which once found were depleted using magnetic beads sorting.
We were able to expanded DN Tregs from 8 different donors and generated 1-10×108
cells within 3 weeks, with the average fold expansion of 3422 ± 2341 (Figure 4b). By day
21, the average purity of DN Tregs was 95.48 ± 2.49% (n=8; Figure 4c).
52
Figure 3. Schematic representation of the method for ex vivo expansion of DN Tregs. PBMCs were isolated from peripheral blood of healthy donors by Ficoll-Hypaque gradient. DN Tregs were enriched for by depleting CD4+, CD8+, CD56+ and TCRγδ+ cells from PBMCs via magnetic beads sorting. On day 0, the sorted population was stimulated by plate-bound anti-CD3 mAb. On days 7, 12 and 17, DN Tregs were harvested, washed and co-cultured with irradiated (150 Gy) artificial APC cells (aAPCs), in the presence of the combination of cytokines. DN Tregs were collected on day 21 and used for phenotypic and functional studies.
53
Figure 4. Isolation and expansion of DN Tregs. (a) DN Tregs were enriched by staining PBMCs with CD4-FITC, CD8-FITC, CD56-FITC, TCRγδ-FITC Abs and depleting the stained cells using anti-FITC magnetic beads and MACS technology. Representative flow cytometry histograms of pre- and post-sort are shown. (b) Fold increase in DN Tregs cells after 21 days of ex vivo expansion of cells cultured in the presence of rhIL-7 (n=8). (c) The purity of DN Tregs was assessed by flow cytometry on day 0 and day 21. Numbers represent frequency in each gate or quadrant. Gating was based on FMO controls. A representative figure is shown.
0
2
4
6
8
Day 21
Fo
ld E
xpan
sio
n (1
03 )
C
TCRαβ
SS
C
CD4
CD
8
d0
d21
Post-sort Pre-sort
TCRγδ, CD4, CD8, CD56
B
Day 21
Fold
Exp
ansi
on (1
03) A
Figure 4. Isolation and expansion of DN Tregs. (A) DN Tregs were enriched by staining PBMCs with CD4-FITC, CD8-FITC, CD56-FITC, TCRγδ-FITC Abs and depleting the stained cells using anti-FITC magnetic beads and MACS technology. Representative flow cytometry histograms of pre- and post-sort are shown. (B) Fold increase in DN Tregs cells after 22 days of ex vivo expansion of cells cultured in the presence of rhIL-7 (n=8). (C) Purity of DN Tregs was assessed by flow cytometry on day 0 and day 21. Numbers represent frequency in each gate or quadrant. Gating was based on FMO controls. Representative figure is shown.
Figure 4
CD56
d0 d21 Positive control
54
3.3. Supplementation of IL-7 During Expansion Enhance Proliferation and Suppressive Function of DN Tregs
The homeostatic cytokines IL-7 and IL-15 have been shown to facilitate nTreg
expansion and function (Asanuma et al., 2011; Ben Ahmed et al., 2009; Simonetta et al.,
2014; Veerapathran et al., 2013). Since DN Tregs purified from peripheral blood express
all components of IL-7 and IL-15-receptors, which are composed of IL-7Rα (CD127),
IL-15Rα (CD215), as well as common beta and gamma subunit of IL-2R (CD122 and
CD132, respectively) (Figure 1c & d), we wanted to evaluate if addition of these
cytokines to the cultures could increase the yield of functional DN Tregs.
We found that supplementation to the culture media of either IL-7 or IL-15 in
addition to IL-2 increased proliferation of DN Tregs, as demonstrated by the higher yield
of DN Tregs on day 28 of co-culture, when compared to DN Tregs grown in the presence
of IL-2 alone (Figure 5a). However, when the viability of DN Tregs was examined at the
end point of expansion, 20% of DN Tregs cultured in the presence of IL-15 were dead, in
comparison to <10% of DN Tregs grown in the presence of IL-2 with or without IL-7
(Figure 5b).
In a recent report, the addition of IL-7 directly to the suppression assay was found
to abrogate the immunosuppressive function of human allo-Ag-specific DN Tregs
(Allgauer et al., 2015). Consequently, we wanted to determine whether the different
growth conditions affected the functionality of DN Tregs. Surprisingly, we found that DN
Tregs expanded in the presence of IL-2 + IL-7 demonstrated three-fold increase in their
suppressive activity on per cell basis as compared to the DN Tregs grown in the presence
IL-2 + IL-15 (65.42 ± 17.56% vs 18.41 ± 9.31%, respectively, p<0.05) (Figure 5c). These
data indicate that supplementation of either IL-7 or IL-15, in addition to IL-2, increases
the total yield of DN Tregs, but only IL-2 + IL-7-grown DN Tregs have potent
suppressive function. Therefore, DN Tregs that were grown in the presence of IL-2 and
IL-7 were used in the subsequent functional studies.
55
Figure 5. The effect of supplementation of IL-7 and IL-15 during the expansion on DN Treg proliferation and function. (a) Supplementation of rhIL-7 (10 ng/ml) or rhIL-15 (10 ng/ml) to cell cultures on day 14 aided DN Tregs in proliferation in comparison to cells grown in rhIL-2 (250 U/ml) only. (b) The viability of DN Tregs grown in the presence of various cytokines as assessed by 7-AAD live/dead staining. (c) DN Tregs expanded in the presence of rhIL-7 exhibit amplified suppressive function against autologous CD4+ cells activated with αCD3/CD28 beads, as assed by CFSE suppression assay. Grouped data of percentage inhibition of suppression at 8:1 suppressor-to-responder ratio is shown. Each point on the graph represents a different donor. *p<0.05. **p<0.01. ***p<0.001. n.s., non-significant.
15 20 25 300
2
4
6
8
Days
Cel
l Co
un
t (x1
06 ) IL-2IL-2 + IL-7IL-2 + IL-15
B A
Figure 5
C
Figure 5. DN Tregs grown in rhIL-7 exhibit augmented regulatory function. (A) Supplementation of rhIL-7 (20 ng/ml) or rhIL-15 (20 ng/ml) to cell cultures on day 14 aided DN Tregs in proliferation in comparison to cells grown in rhIL-2 (250 U/ml) only. Similar results were obtained from 3-6 different donors. (B) Viability of DN Tregs grown in the presence of different cytokines as assessed by 7-AAD live/dead staining. (D) DN Tregs grown in the presence of rhIL-7 exhibit amplified suppressive function against autologous CD4+ cells activated with αCD3/CD28 beads, as assed by CFSE suppression assay. Grouped data of percentage inhibition of suppression at 8:1 suppressor-to-responder ratio is shown. Each point on the graph represents a different donor. * p < 0.05. n.s., non-significant.
15 20 25 300
2
4
6
8
Days
Cel
l Co
un
t (x1
06 ) IL-2IL-2 + IL-7IL-2 + IL-15
IL-2 + IL-15 IL-2 + IL-7IL-2
IL-2
IL-2
+IL-7
IL-2
+IL-1
50
5
10
15
20
25
Growth Conditions
% 7
-AAD
(+) D
N T
regs
********
0
20
40
60
80
100
DN Treg:CD4+ (8:1)
% S
uppr
essi
on
**n.s.
*
DN Treg : CD4+ (8:1)
% S
uppr
essi
on
Days
Cel
l Cou
nt (x
106 )
Growth Conditions
% 7
-AA
D(+
) DN
Tre
gs
56
3.4. Ex Vivo Expanded DN Tregs are Potent Suppressors In Vitro
Previously, human DN Tregs activated by mature allogeneic DCs, or αCD3/CD28
beads, have been shown to suppress proliferation of CD4+ and CD8+ T cells (Allgauer et
al., 2015; Fischer et al., 2005; Voelkl et al., 2011). However, whether polyclonally
activated and ex vivo expanded DN Tregs also manifest suppressive function remained
unknown. Consequently, we evaluated the ability of ex vivo expanded DN Tregs to
inhibit proliferation of autologous CD4+ and CD8+ T cells, as well as CD19+ B cells. To
this end, CFSE-labeled T cells or B cells were polyclonally stimulated with αCD3/CD28
coated beads, or F(ab’)2 fragment of IgM, respectively, and co-cultured in the presence
or absence of DN Tregs at increasing ratios. At the end of the co-culture, proliferation of
the responder cells was assessed by CFSE-dilution using flow cytometry. As shown in
Figure 6a, CD4+ and CD8+ T cells failed to proliferate in the absence of stimulation, but
when αCD3/CD28 beads were added to the co-culture, almost all of the responder cells
proliferated within 5 days. Upon addition of DN Tregs at 8:1, suppressor-to-responder
ratio, the proliferation of responder cells was reduced by half. The results from triplicate
cultures are summarized in Figure 6b and show a dose-dependent suppression of CD4+
and CD8+ T cells. Interestingly, suppression of CD4+ T cells by DN Tregs was
consistently stronger than suppression of CD8+ T cells for all the donors that we tested.
Additionally, DN Tregs are also potent suppressors of proliferation of CD19+ B
cells (Figure 6c) in a dose-dependent manner. Together these data indicate that ex vivo
expanded polyclonal DN Tregs are effective at suppressing proliferation of autologous B
cells, as well as CD4+ and CD8+ T cells in vitro.
57
Figure 6. DN Tregs suppress proliferation of autologous T cells, and CD19+ B cells. Purified naïve CD4+, CD8+ or CD19+ cells were labeled with CFSE and co-cultured with αCD3/CD28 beads or F(ab’)2 fragment of goat anti-human IgM, respectively, in the presence of ex vivo expanded DN Tregs for 4-5 days. (a) The proliferation of responder cells was measured by CFSE dilution. Shown here is an example of gating strategy. Numbers represent the percentage of CD4+ proliferating cells. (b, c) The data are expressed as mean percentage inhibition of three replicate cultures. Error bars represent SD. The experiment was repeated 8 times (b) or 3 times (c) with cells obtained from different donors with similar results.
1.25 2.5 5 100
20
40
60
80
DN Treg : B (x:1)
% In
hibi
tion
of P
rolif
erat
ion
DN Treg : CD19+ (x:1)
% S
uppr
essi
on
1 2 4 80
20
40
60
80
DN Treg:Tconv (x:1)
% In
hibi
tion
of P
rolif
erat
ion
CD4+
CD8+
DN Treg : TCONV (x:1)
% S
uppr
essi
on
B
A
C
αCD3/CD28 αCD3/CD28 + DN Treg
CFSE
CD4+
CD8+
No stimulation
58
3.5. Phenotype of Ex Vivo Expanded DN Tregs
To define characteristics of expanded DN Tregs, we analyzed various markers
associated with Treg phenotype and/or regulatory function with the findings summarized
in Figure 7a. Expanded DN Tregs maintained high expression of CD3 and TCRαβ
complexes, which were slightly downregulated upon strong non-specific stimulation with
PMA/ionomycin - an observation that is consistent with the biology of T cells. CD25
expression was induced after expansion and retained upon stimulation. Since low
concentration of IL-7 had been regularly supplemented to the expansion medium, DN
Treg had little to no expression of CD127, an IL7Rα.
CD62L, also known as L-selectin, has been shown to be important for the function of
Tregs, because it allows homing to lymphoid tissues necessary for induction of peripheral
suppression (Ermann et al., 2005). Ex vivo expanded DN Tregs retained high levels of
CD62L, which was downregulated upon PMA/ionomycin stimulation. Surprisingly,
CD69, an early activation marker upregulated only upon stimulation, whereas CD278
(ICOS), a late activation marker, was constitutively expressed on the cell surface of DN
Tregs, which expression was further upregulated through stimulation. Expanded DN
Tregs have TEM phenotype, as they are CCR7−CD45RO+ (Figure 7b). Low intracellular
expression of regulatory molecules CTLA-4 and PD-1 upon stimulation suggests that
these molecules are unlikely facilitators of DN Treg-mediated immunosuppression
(Figure 7c).
59
Figure 7. Phenotypic characteristics of ex vivo expanded DN Tregs. DN Tregs expanded for 21 days were harvested and stimulated for 4 h with PMA/Ionomycin. (a) Change in expression of cell surface markers upon stimulation as assessed by flow cytometry. (b) Expanded DN Tregs express effector memory phenotype. (c) DN Tregs express low intracellular levels of suppressive molecules CTLA-4 and PD-1. (a-c) Representative histograms from one donor are shown. Similar results were obtained from 4 different donors.
FMO Unstimulated
Stimulated
CD3 TCRαβ CD25
CD127 CD62L CD69 CD278
A
CTLA-4 PD-1 CD45RO
CC
R7
B C
CD56
60
3.6. Cytokine Profile of Ex Vivo Expanded DN Tregs
Previous analysis of cytokine profile of human DN Tregs activated with
allogeneic mature DCs revealed high production of IFN-γ, and some IL-4, IL-5 and IL-10,
which is similar to what has been reported in murine DN Tregs (Fischer et al., 2005; Ford
et al., 2002). We sought to investigate more thoroughly the cytokine profile of ex vivo
expanded polyclonal DN Tregs. We tested supernatants of DN Tregs and CD8+ T cells,
expanded by the same method, with Luminex™ platform that simultaneously measured
concentration of 27 different cytokines and chemokines in a single sample. Despite the
fact that expanded DN Tregs displayed an activated phenotype (Figure 7a), the only
notable cytokine that they produced without additional stimulation, and that has not been
produced by CD8+ T cells, was IL-5, as shown in Figure 8. However, upon stimulation
with PMA/ionomycin, a robust non-specific pro-inflammatory response was detected.
DN Tregs released pro-inflammatory Th1 cytokines IL-2, IFN-γ and TNFα, but to a
lesser extent than CD8+ T cells. DN Tregs also secreted Th2 cytokines IL-4, IL-5, IL-6,
IL-9, IL-10 and IL-13, as well as other pro-inflammatory cytokines such as IL-8, all to a
much higher extent than CD8+ T cells. Moreover, upon stimulation secretion of
chemokines MIP-1α, MIP-1β and RANTES was detected. A rather interesting finding
was high production of an anti-inflammatory cytokine IL-1 receptor antagonist (IL-1RA)
upon stimulation.
61
Figure 8. Cytokine profile of DN Tregs. Purified DN Tregs and CD8+ T cells obtained from the same donor were expanded in the same manner and harvested on day 21. DN Tregs and CD8+ T cells were cultured in media alone, or stimulated with PMA/Ionomycin. Supernatants were collected after 4 h of co-culture. Change in production of cytokines and chemokines upon stimulation measured by Luminex™ platform. Bars represent mean concentration ± s.e.m. of 2 different donors. All values are in pg/ml. ND, non-detectable values. OOR, values outside of detectable range.
0
10
20
30
IL-1β
0
2000
4000
6000
8000
10000
IL-5
0
500
1000
1500
2000
2500
IL-9
NDND
0
2
4
6
8
10
IL-15
0
10
20
30
40
50
G-CSF
0
5
10
15
20
MCP-1
ND
0
5000
10000
15000
RANTES
0
1000
2000
3000
4000
IL-1RA
ND
0
500
1000
1500
IL-6
0
1000
2000
3000
IL-10
ND
0
1000
2000
3000
4000
5000
IL-17
NDND
0
2000
4000
6000
8000
GM-CSF
NDND
0
1000
2000
3000
4000
MIP-1α
OOR OOR
0
10000
20000
30000
40000
50000
TNFα
NDND
0
10000
20000
30000
40000
50000
IL-2
NDND
0
2
4
6
8
IL-7
0
5
10
15
20
25
IL-12(p70)
0
50
100
150
200
250
Eotaxin
0
5000
10000
15000
20000
IFNγ
OOR
0
5
10
15
PDGF-bb
0
50
100
150
200
VEGF
0
200
400
600
800
IL-4
NDND
0
20000
40000
60000
80000
100000
IL-8
NDND
0
10000
20000
30000
40000
IL-13
0
100
200
300
FGF basic
ND
0
100
200
300
400
500
IP-10
ND
0
5000
10000
15000
MIP-1b
IL-1β IL-1RA IL-2 IL-4
IL-5 IL-6 IL-7 IL-8
IL-9 IL-10 IL-12(p70) IL-13
IL-15 IL-17
TNFα
MIP-1b
RANTES VEGF
MCP-1 MIP-1α PDGF-bb
GM-CSF IFN-γ IP-10
Eotaxin FGF basic
G-CSF
Stimulated
Unstimulated Stimulated
Unstimulated DN
Treg
CD8
pg/m
l
62
3.7. DN Treg-mediated Suppression Is Not Facilitated by IFN-γ or IL-10 Cytokines and Requires Cell-to-Cell Contact
Since addition of extrinsic IL-2 or IL-7 to the in vitro suppression assay can
abrogate suppressive function of nTregs or human allo-specific DN Tregs, we evaluated
whether addition of IL-2 or IL-7 had any effect on in vitro function of DN Tregs
expanded only in the presence of IL-2. As seen in Figure 9, the addition of either IL-2 or
IL-7 did not impair regulatory effect of DN Tregs exerted on autologous CD4+ T cells.
IFN-γ has been shown to be important in the function of murine DN Tregs in
order to upregulate surface FasL expression and ameliorate GVHD in vivo, and suppress
and kill CD4+ T cells in vitro (Juvet et al., 2012). Moreover, both IFN-γ and IL-10 are
produced by Ag- and allo-Ag-specific human DN Tregs (Fischer et al., 2005; Voelkl et
al., 2011). Therefore, we investigated whether immunosuppression in polyclonal DN
Tregs is mediated by IFN-γ, or IL-10. First, we confirmed by ELISA the Luminex assay
results. As seen in Figure 10a, PMA/ionomycin stimulated DN Tregs produced large
amounts of IFN-γ although significantly lower than stimulated CD8+ T cells.
Interestingly, stimulated DN Tregs produce significantly larger quantities of IL-10 in
comparison to stimulated CD8+ T cells (Figure 10b).
However, blocking IFN-γ and/or IL-10 by addition of neutralizing Abs to either
cytokine during suppression assay did not inhibit proliferation of autologous CD4+ T
cells (Figure 11a), while CD8+ T cells suppression was slightly reduced upon addition of
either cytokine (Figure 11b), but the results were not statistically significant. However,
when both cytokines were added, the effect was not synergistic. Therefore, we concluded
that unlike Ag-specific mouse DN Tregs, IFN-γ and IL-10 are not important mediators of
polyclonal activated human DN Treg suppression.
63
Figure 9. Addition of IL-2 and/or IL-7 directly to the suppression assay does not impair functionality of DN Tregs. CFSE-labeled CD4+ cells were cultured at increasing ratios with DN Tregs that were grown only in the presence of IL-2. The co-cultures were supplemented with either IL-2, or IL-2 and IL-7, or devoid of any cytokines. After 4 days, the proliferative response of CD4+ T cells stimulated with αCD3/CD28 beads was determined by CFSE dilution. Data is expressed as mean ± SD of three replicate co-cultures. Similar results were obtained with cells from another donor.
1 4 80
20
40
60
80
100
DN Treg:CD4 (x:1)
% In
hibi
tion
of P
rolif
erat
ion
IL-2IL-2 + IL-7∅
Figure 7
DN Treg : CD4+ (x:1)
% S
uppr
essi
on
Figure 6. Addition of IL-2 and/or IL-7 directly to the suppression assay does not impair functionality of DN Tregs. CFSE-labeled CD4+ cells were cultured with DN Tregs at increasing ratios in the presence of IL-2, IL-2 and IL-7 or devoid of any cytokines. After 4 days, the proliferative response of CD4+ T cells stimulated with αCD3/CD28 beads was determined by CFSE dilution. Data is expressed as mean ± SD of three replicate co-cultures. Similar results were obtained with cells from another donor.
64
Figure 10. DN Tregs produce IL-10 and IFN-γ. (a,b) Change in the production of IFN-γ (a) and IL-10 (b) upon PMA/Ionomycin stimulation as assessed by ELISA. Similar results were obtained from 2 different donors. **p<0.01. ***p<0.001.
Stimulated Unstimulated
Figure 10. Cytokine profile of DN Tregs. Purified DN Tregs and CD8+ T cells obtained from the same donor were expanded in the same manner and harvested on day 21. DN Tregs and CD8+ T cells were cultured in media alone (resting, grey bars) or stimulated with PMA/Ionomycin (black bars). Supernatants were collected after 4 h of culture. (A) Change in production of cytokines and chemokines upon stimulation measured by Luminex™ platform. Bars represent mean concentration ± s.e.m. of 2 different donors. All values are in pg/ml. ND, non-detectable values. OOR, values outside of detectable range. (B) Change in production of IFN-γ and IL-10 upon stimulation as assessed by ELISA. Similar results were obtained from 2 different donors.
A
B
DN Treg CD8+0.0
5.0×102
1.0×103
1.5×103
2.0×103
[IL-1
0] (
pg/m
l)
******
**
DN Treg CD8+0.0
5.0×103
1.0×104
1.5×104
2.0×104
[IFN
-γ] (
pg/m
l)
***
******
[IFN
-γ] (
pg/m
l)
[IL-1
0] (p
g/m
l)
DN Treg
DN Treg
CD8+
CD8+
65
Figure 11. Role of IFN-γ and IL-10 in the mechanism of DN Treg suppression. (a, b) Anti-IFN-γ and anti-IL-10 blocking mAbs were added to the DN Treg-mediated suppression assay against CD4+ (a) and CD8+ (b) responder cells obtained from the same donor. After 4 days cells were harvested and responder cells were assessed for proliferation by CFSE dilution, and flow cytometry. Similar results were obtained from 3 different donors.
A
B
1 4 80
20
40
60
80
100
DN Treg:CD8 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n
IgG1
IFNγ AbIL-10 + IFNγ Abs
IL-10 Ab
1 4 80
20
40
60
80
100
DN Treg:CD4 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n
IgG1IL-10 AbIFNγ AbIL-10 + IFNγ Abs
1 4 80
20
40
60
80
100
DN Treg:CD4 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n
IgG1IL-10 AbIFNγ AbIL-10 + IFNγ Abs
% S
uppr
essi
on
DN Treg:CD4+ (x:1)
% S
uppr
essi
on
DN Treg:CD8+ (x:1)
66
Next, we investigated whether cell contact or other soluble factors are important
for DN Treg-mediated suppression. To this end, we have performed a transwell
suppression assay that prevents cell-to-cell contact, but allows diffusion of soluble factors
across the membrane. When both DN Tregs and CD4+ (Figure 12a) or CD8+ (Figure 12b)
responders were cultured in the same chamber thus allowing direct cell-to-cell contact, a
dose-dependent suppression was observed. However, separating DN Tregs in the upper
compartment from the responder T-cells and αCD3/CD28 beads in the lower
compartment completely prevented the suppression of responder T-cell proliferation, and
it was not due to a lack of activation of DN Tregs, as αCD3/CD28 beads were also added
to the upper chamber. Moreover, we tested whether any soluble factors that get released
only upon interaction of DN Tregs with responder cells would promote suppression. To
this end, we co-cultured responder cells with DN Tregs and αCD3/CD28 beads in the
upper chamber, but no inhibition of proliferation of responder T-cells in the lower
chamber was observed. These findings demonstrate that direct cell-to-cell contact is
necessary for DN Treg function.
Lastly, we analyzed the supernatants from the suppression assay co-cultures to
quantify the amount of IFN-γ present. Both CD4+ and CD8+ responder T cells produced
high amounts of IFN-γ upon stimulation with αCD3/CD28 beads (Figure 13). However,
upon addition of DN Tregs, the presence of IFN-γ in the supernatants was significantly
reduced. These findings demonstrate that DN Tregs prevented CD4+ T cells and CD8+ T
cells from releasing IFN-γ.
67
Figure 12. Mechanism of inhibition mediated by DN Tregs requires cell-to-cell contact. (a, b) CD4+ or CD8+ responder T cells were cultured in the bottom chamber of Transwell™ flat-bottom culture plate in media containing αCD3/CD28 beads. DN Tregs were placed in the top chamber with media containing αCD3/CD28 beads in the presence of absence of responder cells. Inhibition of autologous CD4+ T cells (a) and CD8+ T cells (b) was abrogated during co-culture suppression assays in a Transwell™ system. Similar results were obtained from three different donors.
1 4 80
10
20
30
40
DN Treg:CD4 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n No barrierCD4 | DNTCD4 | DNT+ CD4
Figure 14
B A
Figure 13: Mechanism of inhibition mediated by DN Tregs requires cell-to-cell contact. CD4+ or CD8+ responder T cells were cultured in the bottom chamber of Transwell™ flat-bottom culture plate in media containing αCD3/CD28 beads. DN Tregs were placed in the top chamber with media containing αCD3/CD28 beads in the presence of absence of responder cells. Inhibition of autologous CD4+ (A) and CD8+ (B) cells was abrogated during co-culture suppression assays in a Transwell™ system. Similar results were obtained from three different donors.
DN Treg:CD4+ (x:1)
% S
uppr
essi
on
1 4 8-5
0
5
10
15
20
DN Treg:CD8+ (x:1)%
Inhi
bitio
n of
Pro
lifer
atio
n No barrierCD8 | DNTCD8 | DNT + CD8
DN Treg:CD8+ (x:1) %
Sup
pres
sion
68
Figure 13. DN Tregs suppress secretion of IFN-γ by CD4+ T cells and CD8+ T cells. The concentration of IFN-γ was measured in the supernatants obtained from a 4-day suppression assay of DN Tregs and CD4+ T cells, or CD8+ T cells, at 4:1 suppressor-to-target ratio. The values represent mean ± SD of 3 replicates. The amount secreted by DN Tregs in the co-cultures has been subtracted. Similar results were obtained from 3 different suppression assays, each executed with a different donor. * p<0.05. *** p<0.001. n.d., non-detected.
0
20
40
60
80
*
*
******
[IFN
-γ] (
ng/m
l)
CD4+ + + - - - CD8+ - - + + -
DN Treg - + - + +
69
3.8. DN Tregs Do Not Suppress by Killing Responder Cells
There is a large body of evidence that murine DN Tregs suppress immune
responses by direct cytotoxicity to responder cells through various mechanisms (Juvet
and Zhang, 2012). Whether human DN Tregs function by directly killing responder cells
remains controversial. To determine whether DN Tregs are also cytotoxic toward
responder cells, we first directly measured the viability of CD4+ and CD8+ T cells
isolated after 4 days of co-culture with DN Tregs (Figure 14a & b). Upon increasing the
ratio of DN Tregs to responder cells, the viability of the responder cells did not change,
as measured by the percentage of AnnexinV+ and 7-AAD+ cells, suggesting that DN
Tregs do not suppress by killing the responder cells.
Since 4 days is a relatively long time, there was a chance that the time window
when DN Tregs induced apoptosis in the responder cells might have been missed.
Therefore, a standard chromium-release assay was used to validate the results. To this
end, freshly isolated CD4+ or CD8+ T cells were activated with αCD3/CD28 beads in the
absence of DN Tregs for 4 days. Activated CD4+ or CD8+ T cells were then isolated,
labeled with 51Cr and co-cultured with ex vivo expanded DN Tregs for 4 or 22 hours. The
death of CD4+ or CD8+ T cells was assessed by the amount of 51Cr released. Results
indicate that DN Tregs are not cytotoxic towards activated autologous CD4+ or CD8+ T
cells, regardless of the length of co-culture with DN Tregs (Figure 14c & d). Together,
these data indicate that unlike murine DN Tregs, polyclonally expanded human DN Tregs
do not suppress Tconv cells by direct cytotoxicity.
70
Figure 14. DN Tregs do not kill autologous CD4+ or CD8+ T cells. (a) After 4 days of suppression assay, CD4+ cells were assessed for viability (7-AAD) and apoptotic markers (AnnexinV) through flow cytometry. Bar graphs represent mean ± SD from 3 replicates. Similar results were obtained with DN Tregs from at least 4 different donors. (b) Naïve CD4+ cells were stimulated for 4 days with αCD3/CD28 beads. Cells were harvested, counted and co-cultured with ex vivo expanded DN Tregs at different ratios. After 24 h, apoptosis among targets was assessed by AnnexinV staining. The figure represents one replicate out of three. Similar results were obtained from two independent experiments. (c, d) The results were confirmed by 51Cr-release assay. Activated CD4+ T cells or CD8+ T cells were labeled with 51Cr and co-cultured with DN Tregs for 4 or 24 h. The amount of 51Cr released was quantified using scintillation counter. Similar results were obtained with cells from a different donor.
0.6 1.25 2.5 5 100
2
4
6
DN Treg:CD4+ (x:1)
% S
peci
fic L
ysis
4 h22 h
n.d. n.d.
0.6 1.25 2.5 5 100.0
0.5
1.0
1.5
DN Treg:CD8 (x:1)
% S
peci
fic L
ysis
4 h 22 h
n.d. n.d.
0 1 2 4 860
70
80
90
DN Treg:CD8+ (x:1)
% 7
AA
D (-
) in
CD
8+ p
op
ula
tion
A
C
0 1 2 4 860
70
80
90
DN Treg:CD4+ (x:1)
% 7
AA
D (-
) in
CD
4+ p
op
ula
tion
CD4
B
D
% V
iabl
e C
D4+
% V
iabl
e C
D8+
DN Treg : CD4+ (x:1) DN Treg : CD8+ (x:1)
% S
peci
fic L
ysis
% S
peci
fic L
ysis
DN Treg : CD4+ (x:1) DN Treg : CD8+ (x:1)
71
3.9. Ex Vivo Expanded DN Tregs Kill Human Cancer Cells
Previously, we demonstrated that both TCRαβ+ and TCRγδ+ populations sorted
from DN T cells expanded from peripheral blood of AML patients in complete remission
were cytotoxic against autologous and allogeneic leukemic blasts in vitro (Merims et al.,
2011). Therefore, we wanted to test whether DN Tregs expanded by the means of a novel
protocol also demonstrate cytotoxic function against various human cancer lines. To
assess DN Treg-induced killing of cancer cells, a flow cytometry-based killing assay was
adapted in which cancer cells were stained with PKH-26 before co-culture with DN
Tregs. The percentage of cytotoxicity in the PKH-26-gated cell population was calculated
as described in Materials and Methods, and summarized in Figure 15. DN Tregs
efficiently killed leukemic MV4-11 and K562 cell lines, but were less cytotoxic towards
AML-3 and KG1a cells lines (Figure 15a). Furthermore, DN Tregs exerted a dose-
dependent cytotoxicity against all human primary lung cancer cell lines that were tested:
186-144, 426-177 and H125 lines (Figure 15b). These findings demonstrate that DN
Tregs exhibit not only immunoregulatory function, but are also cytotoxic to various
cancer cells in vitro.
72
Figure 15. Ex vivo expanded DN Tregs can kill human cancer cells. (a,b) Cytotoxicity against human leukemic (a) or lung cancer (b) cell lines by ex vivo expanded DN Tregs was determined by flow cytometry-based killing assay. In short, cancer cells were labeled with PKH and co-cultured for 2 hours (a) or 16 hours (b) with DN Tregs at the indicated ratios. Specific killing of cancer cells was determined by calculating the proportion of cells remaining alive after co-incubation with DN Tregs. Similar results were obtained from 3 different donors.
1 4 8 160
20
40
60
80
DN Treg:Target (x:1)
% S
peci
fic K
illin
g
2h
KG1aAML-3K562MV4-11
Targets:
1 4 8 160
20
40
60
80
DN Treg:Target (x:1)
% S
peci
fic K
illin
g
16h
186-144426-177
H125
Targets:
A
B
DN Treg : Target (x:1)
% S
peci
fic K
illin
g
DN Treg : Target (x:1)
% S
peci
fic K
illin
g
73
3.10. Spatial and Temporal Dynamics of Human DN Tregs In Vivo
Due to the limitation in obtaining large numbers of human DN Tregs, the survival,
proliferative potential and distribution after adoptive transfer of DN Tregs have never
been studied before. The expansion method presented here allows for large-scale
production of DN Tregs, thus allowing for the first time to investigate the features of DN
Tregs in vivo. For this purpose, ex vivo expanded DN Tregs were stained with CFSE and
infused intravenously to sublethally irradiated NSG mice. The same numbers of freshly
isolated PBMCs obtained from the same donor as DN Tregs, were labelled with CFSE
and injected into a group of mice as controls. Every few days, 2 to 3 mice from each
cluster were sacrificed and various organs were collected to determine proliferative
capacity of the infused lymphocytes. As shown in Figure 16a & b, DN Tregs, just like
Tconv cells, traveled to all hematopoietic and lymphoid tissues, including peripheral
blood (PB), bone marrow (BM), spleen and lymph nodes (LNs), as well as other organs
including lung, liver, and kidneys. However, the proliferative response of DN Tregs was
lower when compared to that of Tconv cells, as determined by the CFSE dilution (Figure
16a). Over the span of 10 days, the frequency of DN Tregs in all of the studied tissues
decreased, while the frequency of Tconv cells increased (Figure 16b). Together, these
findings suggest that only small proportion of DN Tregs proliferate in vivo, which may
indicate that DN Tregs are less responsive or even completely unresponsive to xeno-
antigens in comparison to Tconv cells, which within 10 days proliferated to a high extent
in all the tissues tested.
74
Figure 16. Tracking and proliferation of human lymphocytes adoptively transferred to NSG mice. Freshly isolated PBMCs or ex vivo expanded DN Tregs from the same healthy donor were stained with CFSE and injected into sublethally irradiated (250 cGy) NSG mice. On days 1, 3, 5, 7 and 10 post infusion, 2 to 3 mice per group were sacrificed, and peripheral blood (PB), bone marrow (BM) and following tissues were harvested: spleen, kidneys, liver, and lungs. (a) Analysis of proliferation of lymphocytes was assessed by CFSE dilution and gated on 7-AAD− CD45+
CD3+. (b) The percentage of human CD45+ CD3+ cells within 7-AAD− fraction from each tissue was assessed by flow cytometry.
d1
d3
d5
d7
d10
d1
d3
d5
d7
d10
Spleen
PBMC
DN Treg
PB
CFSE
Kidney Lung Liver A
0 2 4 6 8 100
10
20
30
Blood
Days
% C
D45
+ C
D3+
%
CD
45+
CD
3+
0 2 4 6 8 100
5
10
15
20
25
Bone Marrow
Days
% C
D45
+ C
D3+
0 2 4 6 8 100
10
20
30
40
Spleen
Days
% C
D45
+ C
D3+
0 2 4 6 8 100
20
40
60
Days
% C
D45
+ C
D3+
Lung
0 2 4 6 8 100
20
40
60
80
100
Liver
Days
% C
D45
+ C
D3+
Days
0 2 4 6 8 100
20
40
60
80
Kidney
Days
% C
D45
+ C
D3+
PBMCDN Treg
PB BM Spleen
Lung Liver Kidney
B
75
3.11. Ex Vivo Expanded DN Tregs Delayed Onset of Xenogeneic GVHD in NSG Mice
Since DN Tregs suppress autologous T cells and travel to the same organs as
Tconv cells, the ability of ex vivo expanded human DN Tregs to suppress the immune
responses in vivo was evaluated by using a xenogeneic GVHD mouse model (Figure 17).
To induce xenogeneic GVHD, PBMCs were freshly isolated from peripheral blood and
5×106 cells were infused into sub-lethally irradiated NSG mice. As shown in Figure 18a,
all mice injected with PBMC only died within 31 days with the median survival time
(MST) of 19 days, as manifested by severe weight loss (Figure 18b). In contrast, infusion
of the same number of ex vivo expanded DN Tregs from the same donor did not cause
any weight loss, or induce xenogeneic GVHD in the recipient mice (Figure 18a & b), and
all mice remained alive and healthy up to 100 days post injection.
To determine whether DN Tregs could ameliorate GVHD, PBMC-injected mice
were treated with three doses of 107 DN Tregs on days 0, 3 and 7 after induction of
GVHD (Figure 17). Infusion of DN Tregs protected mice from GVHD as it significantly
delayed an onset of disease (MST=28, p<0.01, Figure 18a), with one of the mice
surviving for 77 days. In addition to prolonged survival, three injections of DN Tregs also
significantly protected mice from weight loss, as compared to PBMC only group (p<0.05,
Figure 18b). These data indicate that infusion of ex vivo expanded polyclonal human DN
Tregs does not cause xenogeneic GVHD in immunodeficient mice and can attenuate
GVHD induced by human PBMCs.
76
Figure 17. Schematic protocol of the xenogeneic GVHD experiment. 6-8 week old NSG mice were sublethally irradiated (250 cGy) and i.v. injected with 5×106 PBMCs obtained from a healthy human donor or 107 DN Tregs ex vivo expanded from the same donor. Treated mice were infused with PBMCs followed by injection of 3 doses of 107 DN Tregs on day 0, 3 and 7. Mice were monitored daily for the signs of GVHD.
Irradiation 250cGy
Day -1 Day 0 Days 0-7
Treatment
huPBMC inj. 5x106
Days 10+ Mice monitored
for signs of GVHD
huDN Treg inj. 107
huDN Treg inj. 107
77
Figure 18. Treatment with DN Tregs delayed the onset of xenogeneic GVHD. (a) Survival curves of the mice injected with PBMC (n=23), DN Tregs (n=14) or PBMCs and 3 doses of DN Tregs (n=14). Statistical differences between the curves were compared using log-rank test. (b) Weight curves of the same animals as depicted in (a). Mice that lost >20% of the initial weight were euthanized. For consistency, the last weight measurement of each diseased animal was kept in the analysis until the last mouse in the group was sacrificed. Statistical differences between the curves were calculated with ANOVA. Survival and weight monitoring data were pooled from three separate experiments. *p<0.05. **p<0.01. ***p<0.001.
0 20 400
20
40
60
80
100
0 20 4080
100
120
Days Post Injection
Initi
al B
ody
Wei
ght (
%)
*
***
DN TregPBMCPBMC + DN Treg (3 inj.)
Days Post Injection
% S
urvi
val
Days Post Injection
% In
itial
Wei
ght
***
**
A
B
78
3.12. Rapamycin Augmented Immunosuppressive Function of DN Tregs In Vitro and In Vivo
Rapamycin (sirolimus, RAPA) is an mTOR inhibitor that facilitates expansion
and promotes the function of nTregs and iTregs in mice and humans (Shan et al., 2014).
In the recent report, the addition of IL-7 directly to the suppression assay abrogated the
immunosuppressive function of human allo-Ag-specific DN Tregs via activation of
Akt/mTOR pathway, whereas inhibition of this pathway reversed the IL-7 effect
(Allgauer et al., 2015). We found that addition of IL-7 directly to the suppression assay
did not affect the function of polyclonally activated DN Tregs (Figure 9). However, we
investigated whether the inhibition of Akt/mTOR pathway by rapamycin could enhance
the immunosuppressive function of DN Tregs.
For this purpose, ex vivo expanded DN Tregs were pre-incubated with rapamycin
for 2 hours, extensively washed and used in a 3-day suppression assay. Blockade of the
Akt/mTOR pathway by rapamycin rendered DN Tregs more immunosuppressive, as they
inhibited proliferation of autologous CD4+ and CD8+ T cells to a greater extent than non-
treated DN Tregs (Figure 19a). Suppression of CD4+ T cells increased significantly by
almost 52 ± 2 % at the 1:1 suppressor-to-responder ratio; while at the 5:1 ratio the
difference was non-significant, likely due to saturation of the suppression by untreated
DN Tregs (Figure 19b). Since DN Tregs always exerted a more modest suppression
against CD8+ cells as compared to CD4+ cells, rapamycin-treated DN Tregs demonstrated
significant improvements in their suppressive activity by 17 ± 1% at 1:1 suppressor-to-
responder ratio and by 27 ± 4% at 5:1 ratio (Figure 19c). These data indicate that
rapamycin treatment significantly augments the suppressive function of DN Tregs.
Whether rapamycin-treated DN Tregs can also exert enhanced
immunosuppressive function in vivo was evaluated. To this end, PBMCs were infused
into irradiated NSG mice to instigate xenogeneic GVHD. Treatment groups received one
injection of untreated DN Tregs or rapamycin-treated DN Tregs. As shown in Figure 20a,
79
one injection of rapamycin-treated DN Tregs significantly delayed the onset of
xenogeneic GVHD (MST=32 days, p<0.001) in comparison to PBMC-injected mice
(MST=18 days), while infusion of untreated DN Tregs did not have any significant effect
on survival (MST=20 days). In addition to prolonged survival, one injection of
rapamycin-treated DN Tregs also significantly protected mice from weight loss as
compared to PBMC only group (p<0.05, Figure 20b), whereas one injection of untreated
DN Tregs had no significant effect on weight loss. Together, these data indicate that
treatment of DN Tregs with rapamycin renders them more immunosuppressive both in
vitro and in vivo.
80
Figure 19. Rapamycin-treated DN Tregs manifest augmented regulatory function. Ex vivo expanded DN Tregs were pre-incubated with rapamycin for 2 h, washed extensively and used as suppressor cells in a suppression assay against autologous CD4+ and CD8+ cells stimulated with αCD3/CD28 beads. (a) On day 3, the proliferation of CD4+ and CD8+ responder cells was quantified by CFSE dilution. The histograms represent proliferation of responder cells at 1:1, DN Treg-to-responder ratio. The number represent percentage of proliferating responder cells. (b,c) The average suppression of proliferation of CD4+ (b) and CD8+ (c) responder cells from triplicates. Similar results were observed in 3 independent experiments. **p<0.01. n.s., non-significant.
B
A
C
CFSE
Untreated DN Treg
RAPA-treated DN Treg
CD4+
CD8+
No DN Treg
1 50
20
40
60
80
100
DN Treg:CD4 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n
ControlRapa-treated
**n.s.
DN Treg : CD4+ (x:1)
% S
uppr
essi
on
1 50
20
40
60
80
100
DN Treg:CD8 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n
ControlRapa-treated
**
**
1 50
20
40
60
80
100
DN Treg:CD8 (x:1)
% In
hib
itio
n o
f Pro
lifer
atio
n
ControlRapa-treated
**
**
DN Treg : CD8+ (x:1)
% S
uppr
essi
on
81
Figure 20. Rapamycin-treated DN Tregs delayed the onset of xenogeneic GVHD. 6-8 week old NSG mice were sublethally irradiated (250 cGy) and i.v. injected with 5×106 PBMCs (n=28) to induce GVHD. Treated mice were infused with PBMCs followed by one injection of 107 untreated DN Tregs (n=19) or one injection of 107 rapamycin-treated DN Tregs (n=9). (a) Survival curves of the recipient mice. Statistical differences between the curves were compared using log-rank test. (b) Weight curves of the same animals as shown in (a). The weight of mice was monitored daily to assess severity of GVHD. Mice that lost >20% of their initial weight were euthanized. For consistency, the last weight measurement of each diseased animal was kept in the analysis until the last mouse in the group expired. Statistical differences between curves were calculated with ANOVA. Survival and weight monitoring data were pooled from four separate experiments. ***p<0.001. *p<0.05. n.s., non-significant.
A
B PBMCPBMC + DN Treg (1inj.)PBMC + RAPA-DN Treg (1inj.)
0 10 20 30 400
20
40
60
80
100
Days Post Injection
Sur
viva
l (%
)
Days Post Injection
% S
urvi
val
0 20 4080
100
120
Days Post Injection
Initi
al B
ody
Wei
ght (
%)
n.s.*
Days Post Injection
% In
itial
Wei
ght
***
n.s.
82
CHAPTER 4. DISCUSSION
83
Chapter 4 Discussion
4.1. General Discussion
DN Tregs comprise only 1 to 2% of lymphocytes in the human peripheral blood.
The major obstacles in studying DN Tregs are their low frequency and the lack of
effective expansion method. In this study, we presented a novel protocol for a large-scale
ex vivo expansion of human DN Tregs. The methodologies in the previously published
studies required a large starting population, generation of allogeneic mature DCs as a
source of stimulation and activation, and extensive supplementation of growth factors,
and cytokines (Fischer et al., 2005; Voelkl et al., 2011). We developed a less complicated
and arguably cheaper protocol that allows for the generation of up to ~109 DN Tregs
within 21 days. With the method here presented, a much smaller starting population can
be used as opposed to leukophoresis product. On average we were able to expand DN
Tregs ~3500-fold and generate clinically significant numbers with only 50 to 100 ml of
fresh peripheral blood obtained from healthy donors, or 3-10×107 cryopreserved PBMCs.
DN Tregs do not express any specific markers that would allow for one-step
isolation from PBMCs. Furthermore, DN Tregs being such a rare population of
lymphocytes are unsuitable for FACS sorting, because it would have negative effect on
the viability and the cell number of the starting population due to prolonged sorting times.
To bypass these obstacles and to reduce the complexity of the process of selection from
PBMCs, we decided to deplete other T cells and NK cells by magnetic beads sorting, as
these are the cells would proliferate extensively under the DN Treg expansion conditions.
Therefore, the initial DN Treg population contained other mononuclear cells such as
monocytes and B cells, which briefly played a role of supporting cells until they died off
due to lack of specific activation. Currently, the most common T cell expansion methods
include supplementation of IL-2 and stimulation of cells with anti-CD3 mAb,
84
αCD3/CD28 beads, allogeneic APCs in the form of B cells or DCs, or artificial APCs
(aAPCs). Since autologous APCs were present in the initial co-culture, the stimulatory
signal was provided by plate bound anti-CD3 mAb to ensure DN Treg activation. Efforts
in using soluble anti-CD3 mAb to facilitate further expansion of DN Tregs were
undertaken, but were fruitless suggesting that DN Tregs need other co-stimulatory signals.
This may be due to the fact that anti-CD3 mAb delivers a moderate proliferative signal
through the T cell receptor complex (signal 1), but in the absence of additional co-
stimulatory signals (signal 2) the resulting proliferation is often followed by premature T
cell apoptosis or anergy (Schwartz, 1990). However, too strong of stimuli can also lead to
the exhaustion of the cell colony. Thus every round of stimulation is a balancing act
between stimuli strong enough to drive proliferation, but weak enough to not cause
activation induced cell death (AICD).
Previously reported was that human DN Tregs exerted their suppressive activity
exclusively after pre-activation (Voelkl et al., 2011). Thereby in preceding studies DN
Tregs were activated with allogeneic mDCs (Allgauer et al., 2015; Fischer et al., 2005;
Voelkl et al., 2011). However, the proliferation of DN Tregs was modest even though
mDCs would provide signals 1 and 2. Furthermore, using mDCs for the expansion
purposes is challenging, as they have to be differentiated in vitro from mononuclear cells
isolated from the donors’ blood prior to DN Treg expansion. Since mDCs are unsuitable
for cryopreservation, the number of generated mDCs would be a limiting factor if used in
a large-scale expansion. Since mDCs would have to be constantly in culture, it would
impose additional materials and labour costs. aAPCs, a K562 cells line that expresses
transduced anti-CD3 Ab and co-stimulatory molecules 4-11BL, CD80 and CD86 among
other things, provides co-stimulatory signals similar to mDCs, and may offer an
alternative method of the expansion of DN Tregs with the exception in TCR-specificity,
as mDC-stimulated DN Tregs would have allogeneic specificity, whereas aAPC-
stimulated DN Tregs would be polyclonal. There are many advantages in using aAPCs as
opposed to mDCs: aAPCs being an immortalized cell line are easy to grow, suitable for
cryopreservation for later use and represent a readily available off-the-shelf product
appropriate for the expansion of DN Tregs from virtually any donor. The final protocol
85
that yields up to 1-10×108 DN Tregs calls for 1 round of stimulation with anti-CD3 mAb
to specifically activate DN Tregs and 3 rounds of stimulation with irradiated aAPCs to
provide a broad range of co-stimulatory signals, for the total expansion time of 21 days.
We were able to expand cells from all donors that we tested (n=8) and using this protocol
we achieved ~3500 fold expansion of DN Tregs. DN Tregs continue to proliferate after
21 days upon addition of irradiated aAPCs every 5-7 days. However, the viability of the
cells progressively decreases, along with their function, most likely due to AICD. After
42 days of expansion, DN Tregs manifested severely diminished function (data not
shown).
One of the most surprising findings of this study was the difference in the
magnitude of suppressive function of DN Tregs that were grown in the presence of IL-7
or IL-15, in addition to IL-2. Although the scientific community is still uncertain about
the requirement and the role of these two cytokines on the subject of homeostasis and
function of Tregs, both IL-7 and IL-15 have been used as growth factors in Treg
expansion protocols. The fact that freshly isolated DN Tregs expressed all the
components of IL-7 and IL-15 receptors indicated that DN Treg may be receptive to these
cytokines. The addition of IL-15 apart from IL-2 resulted in more robust proliferation in
comparison to DN Tregs grown solely in IL-2, or IL-2 + IL-7. However, the viability of
IL-2 + IL-15-grown DN Tregs was significantly reduced as compared to either IL-2 or
IL-2 + IL-7-grown DN Tregs, which may be a sign of AICD. Also, the addition of IL-15
resulted in the outgrowth of residual NK cells that had to be depleted on a regular basis, a
phenomenon absent in the other culture conditions. Looking at the viability and the
proliferative potential, supplementing expansion culture with IL-2 + IL-7 seemed to
generate most promising candidates. When we compared the ability of differently grown
DN Tregs to supress autologous CD4+ T cells, DN Tregs that were expanded in the
presence of IL-2 + IL-7 significantly outperformed other contenders. Therefore, DN Treg
expansion culture was supplemented with IL-2 and IL-7.
Interestingly, when the phenotype of IL-7 and IL-15-grown DN Tregs was
compared it did not reveal any significant differences with regards to the expression of
cell surface markers that were tested (data not shown). However, detailed analysis of the
86
activation, as well as inhibitory molecules, may give guidance to the mechanisms of
action of DN Tregs.
The phenotype of ex vivo expanded DN Tregs using the proposed method is
comparable to what was previously observed with allo-Ag-specific DN Tregs (Fischer et
al., 2005; Voelkl et al., 2011). Expanded DN Tregs have higher expression of CD25 and
CD122, the components of IL-2 receptor, which is on par with what is generally observed
in any T cell expansion protocol due to supplementation of exogenous IL-2. CD25 is a
well-known activation marker, thus multiple rounds of activation with aAPCs would
further induce its expression. Furthermore, expanded DN Tregs have somewhat
surprising expression of other activation markers. CD278, known as inducible T cell co-
stimulator (ICOS), is a late activation marker that is expressed on the expanded DN Tregs.
An early activation marker CD69 is not expressed on the expanded DN Tregs and only
gets upregulated upon very strong stimulation with PMA/ionomycin. This property of
DN Tregs differs from that of conventional T cells. However, we are unable to propose
explanation of this phenomenon at this moment. Kinetic studies involving induced
activation and changes in the receptor expression need to be conducted. Lastly, DN Tregs
express high levels of L-selectin (CD62L), a homing receptor to secondary lymphoid
organs, even after a three-week expansion. L-selectin has been shown to be critical in
nTreg function, as its expression indicated highly suppressive Tregs within the
CD4+CD25+Foxp3+ cohort. Upon further stimulation, L-selectin gets downregulated and
thus would render DN Treg unable to enter secondary lymphoid tissues. In this study we
were unable to determine the importance of L-selectin expression on the DN Treg
function. However, using DN Tregs activated with PMA/Ionomycin in the suppression
assay could reveal the role of L-selectin as it may be important in in vivo studies and
clinical applications.
A potential problem with the expansion that has not been addressed could be a
reduction in DN Treg TCR repertoire and presence of non-regulatory cells in the ex vivo
expanded population. Data from multiple studies regarding Vβ and Vα gene usage of DN
Tregs is conflicting (Bristeau-Leprince et al., 2008; Brooks et al., 1993; Fischer et al.,
2005; Niehues et al., 1994) and most likely reflects the heterogeneity of this population.
87
We have shown that DN Tregs are composed of naïve and effector memory phenotypes,
and it has been reported that different expansion methods may preferentially target one
cell type over the other, as it has been observed apropos expansion of Tconv cells.
Higher L-selectin expression in the ex vivo expanded DN Treg population than on freshly
isolated DN Tregs suggests that DN Tregs expanded mostly likely from the naïve cohort.
If our suspicion is correct these cells should have highly diverse TCR, which would not
be the case with memory or effector phenotypes. However, only comparative analysis of
the TCR Vβ repertoire of freshly isolated and ex vivo expanded DN Tregs would reveal
whether all, or only specific DN Tregs clones were expanded further confirming the
polyclonal status of DN Tregs.
Previous studies demonstrated that DN Tregs can induce Ag-specific immune
tolerance in various murine models (Chen et al., 2003a; Chen et al., 2005; Ford et al.,
2002; Minagawa et al., 2004; Thomson et al., 2007; Zhang et al., 2007) and human DN
Tregs have been shown to supress in vitro in Ag- or allo-Ag-specific manner (Allgauer et
al., 2015; Fischer et al., 2005; Voelkl et al., 2011). For the first time, we show that DN
Treg-mediated suppression can also be non-specific, as ex vivo expanded DN Tregs with
polyclonal specificity successfully suppressed proliferation of autologous CD4+ T cells
and CD8+ T cells stimulated with αCD3/CD28 beads. In all the donors that we tested,
CD4+ T cells were consistently suppressed more than CD8+ T cells. One possibility for
this discrepancy may come from the fact that CD4+ T cells were positively selected using
solely CD4 mAb, and thus 5-10% of total CD4+ T population may represent CD4+CD25+
T cells, the majority of which would be nTregs that might add to the overall suppression.
However, no suppression of responder CD4+ T cells has been observed in the absence of
DN Tregs. Thus nTregs, although potent, had minimal effect on the suppression of CD4+
T cells, likely due to insufficient stimulation. However, it is possible although unlikely
that DN Tregs may directly interact with nTregs or induce naïve CD4+ T cells to
differentiate into iTregs, either directly or via soluble factors such as IL-10, thus
contributing to the overall suppression. Studies need to be conducted that would evaluate
the relationship of DN Tregs with nTregs and the ability of DN Tregs to aid in
differentiation of iTregs.
88
DN Tregs exhibit a unique cytokine production profile that differs from that of
CD4+ T cells and CD8+ T cells. Since a proportion of DN Tregs may be CD8+ T cells that
downregulated CD8 co-receptor, we compared the cytokine production of DN Tregs to
CD8+ T cells. Similarly to CD8+ T cells, minimal cytokine production by the ex vivo
expanded DN Tregs was detected in the absence of stimulation. However, upon potent
non-specific stimulation with PMA and ionomycin, DN Tregs secreted a robust array of
different cytokines, which differed from those secreted by CD8+ T cells in the amount
and type of cytokine secreted. Stimulated DN Tregs were skewed towards the production
of Th2 cytokines, which in nTregs were found to promote immune tolerance (Tran et al.,
2012), and thus may act as an important regulatory factors in DN Treg suppression.
Furthermore, a lot of cytokines have pro- and anti-inflammatory properties depending on
the context of stimulation. The ability to robustly produce such a wide array of different
cytokines may also indicate heterogeneity of DN Tregs. This could mean two different
things: 1) there could be many subtypes of DN Tregs and each subtype may produce
different set of cytokines; or 2) a single DN Treg cell may be able to produce different set
of cytokines depending on the milieu, site, Ag recognition or interactions with different
immune cells.
Studies on nTregs, as well as different types of iTreg subsets, revealed that Tregs
regulate immune responses via production of immune cytokines such as IL-10 and TGF-
β. Human DN Tregs are potent producers of IL-10 and IFN-γ upon stimulation, and these
cytokines played a significant role in the murine DN Treg-mediated suppression (Dugas
et al., 2010; Ford et al., 2002; Juvet et al., 2012). However, neutralization of IL-10 or
IFN-γ during the inhibition of proliferation assays of CD4+ T cells and CD8+ T cells had
insignificant effect on the DN Treg-mediated suppression. Furthermore, we showed that
polyclonal DN Tregs require cell-to-cell contact to mediate suppression, since the
presence of the membrane in transwell assay prevented suppression of the responder cells,
indicating that immunosuppression is not dependent on IL-10, IFN-γ, or other soluble
factors. Furthermore, the addition of exogenous IL-2 or IL-7 to the suppression assay
failed to abolish DN Treg regulatory function, indicating that unlike nTregs, they do not
suppress through consumption of IL-2.
89
Multiple studies have reported that murine DN Tregs mediate suppression by
eliminating T cells through Fas/FasL interaction or via perforin/granzyme, which would
ultimately induce apoptosis in the responder cells (Chen et al., 2003a; Ford et al., 2002;
Young et al., 2002; Zhang et al., 2007; Zhang et al., 2006; Zhang et al., 2000). However,
there are conflicting studies regarding human DN Tregs mediation of suppression by
elimination of responder T cells (Fischer et al., 2005; Voelkl et al., 2011). AnnexinV/7-
AAD staining did not reveal a decrease in the viability of CD4+ T cells or CD8+ T cells
upon addition of DN Tregs, suggesting that DN Tregs do not induce apoptosis in the
responder T cells. Further study was conducted to measure directly the ability of DN
Tregs to kill activated responder T cells and revealed that DN Tregs are not cytotoxic
towards responder T cells. Interestingly, DN Tregs are capable of cytotoxic activity, as
they successfully killed various leukemic and lung cancer cell lines.
We have shown that DN Tregs exhibit dual function; they can successfully kill
some human cancer lines, as well as suppress the proliferation of lymphocytes. This dual
function is somewhat unusual. However, it is yet to be determined if the suppressive and
cytotoxic functions are performed by the same cells. If one cell is capable of both
functions, then depending on the milieu cytotoxic function may override regulatory
function and vice versa. However, since both regulatory and cytotoxic capacity of DN
Tregs is lower in comparison to nTregs and TCRγδ+CD4−CD8− T cells, respectively, it is
also possible that certain subsets of DN Tregs are specialized to carry out different
effector functions. Finding a definitive phenotypic marker that would distinguish a truly
regulatory population would certainly magnify appeal of these cells as potential
candidates for adoptive cellular therapy.
Furthermore, for the first time we show that human DN Tregs are capable of
suppression of proliferation of autologous CD19+ B cells, just as murine DN Tregs do so
but in an Ag-specific manner (Hillhouse et al., 2010; Ma et al., 2008; Zhang et al., 2006).
This dose-dependent suppression of B cells opens doors for wider applications of DN
Tregs in the clinic, as autoreactive B cells mediate many autoimmune diseases (Browning,
2006). The in vitro results suggest that DN Tregs participate not only in the regulation of
cellular immunity, but also humoral immunity. Due to time constraints, in this study we
90
were unable to determine whether DN Tregs also hamper B cell activation, class-
switching or antibody production. However, all these questions should be addressed in
the future.
The major caveat of the proposed expansion protocol is that expanded DN Tregs
have polyclonal reactivity and thus exhibit reduced suppressive function in comparison to
Ag- or allo-Ag-specific DN Tregs, i.e. more DN Tregs are needed to generate similar
levels of suppression. However, a reduction in polyclonal nTreg potency following ex
vivo expansion has been observed previously. Therefore, we are working towards
production of allo-Ag-specific DN Tregs by using allogeneic B cells as activators instead
of mDCs. Working with B cells is arguably less challenging than working with mDCs, as
large quantity of B cells may be produced in relatively short period of time using CD40L-
transduced fibroblasts. B cells are not temperature sensitive and can be cryopreserved for
immediate, off-the-shelf use. However, this expansion approach requires further
optimization, as we could only produce limited amount of cells thus far. Nevertheless, we
believe that stimulation with aAPCs post allogeneic B cell activation would increase the
yield of allo-Ag-specific DN Tregs, which would essentially increase DN Treg potency
and reduce the number of DN Tregs needed.
To increase the potency of polyclonal DN Tregs we further evaluated whether
treatment of DN Tregs with rapamycin, an mTOR inhibitor, would have any effect on the
suppressive function of DN Tregs. In a recently published study, human allo-Ag-specific
DN Tregs were found to be sensitive to IL-7 and addition of IL-7 to the suppression assay
severely hampered the suppressive function of DN Tregs (Allgauer et al., 2015). The
authors found that IL-7 hyperactivated Akt/mTOR pathway, however inhibition of this
pathway by Akt or mTOR inhibitors restored the functionality of DN Tregs. Furthermore,
it has been found that rapamycin increased suppressive capacity of nTregs and iTregs, as
well as enhanced generation of iTregs (Lu et al., 2014). In this study, we showed that DN
Tregs were most immunosuppressive, when they were grown in the presence of IL-7.
Additionally, supplementation of exogenous IL-7 to the suppression assay did not have
any effect on DN Treg function. We found that a 2-hour pre-incubation with rapamycin
prior to the suppression assay significantly increased the DN Treg-mediated suppressive
91
function on par with what has been observed in other Tregs (Lu et al., 2014). However,
the molecular mechanisms behind this suppression are unknown. We hypothesize that
rapamycin may induce T cell anergy by blocking the Akt/mTOR pathway that is
responsible for survival and proliferation, and switching to STAT5 signalling pathway
that drives anergy associated molecules to be upregulated on the cell surface of these
cells. Further studies are needed to evaluate changes in any cell surface markers upon
mTOR inhibition. It is yet to be determined what impact the mTOR inhibition has on the
proliferation of DN Tregs, as rapamycin drives proliferation in nTregs and iTregs but not
conventional T cells (Lu et al., 2014). Whether there are any alterations in the phenotype
or expression of any suppressive molecules in rapamycin treated-DN Tregs should also
be evaluated in the future. Comparison of the phenotypes of the rapamycin-treated and
untreated DN Tregs could be useful in revealing markers that partake in the
immunosuppressive function of DN Tregs.
We demonstrate for the first time that DN Treg infusion attenuated xenogeneic
GVHD. However, DN Tregs only provided a temporary protection from GVHD since all
mice that received DN Treg infusion eventually developed signs of GVHD, such as
weight loss, hunched posture and apathy. This could be due to lack of persistence, or
survival of the infused human DN Tregs in the mouse environment. Since the in vitro
studies revealed that DN Treg-mediated suppression is cell-contact dependent and does
not involve elimination of the responder T cells, the ability of DN Tregs to protect mice
for GVHD would continue as long as the DN Tregs survive inside the host. Furthermore,
we propose that the presence of DN Tregs may increase the activation threshold of the
responder cells, thus the magnitude of suppression mediated by DN Tregs in vivo may
directly depend on the number of injected DN Tregs. One injection of DN Tregs was
insufficient to significantly delay the onset of GVHD. Three injections over the span of
one week, however, significantly delayed onset GVHD by ~10 days and reduced the
severity of GVHD based on the weight loss data. We have also evaluated whether
injection of rapamycin treated-DN Tregs would minimize the amount of DN Tregs
required to have protective effect from GVHD, as rapamycin treatment rendered DN
Tregs more immunosuppressive than untreated DN Tregs in vitro. Intrestingly, one
92
injection of rapamycin treated-DN Tregs protected mice to similar levels as treatment
with three doses of untreated DN Tregs, when we compare the median survival time
between the two groups. Further studies are needed to assess whether infusion of more
rapamycin treated-DN Tregs, or injection of higher doses of DN Treg would halt GVHD
progression entirely. Also, it is yet to be determined whether approaches combining DN
Treg infusion together with rapamycin, or low-dose IL-7 would be more effective than
infusion DN Tregs alone.
Importantly, unlike the infusion of PBMCs, the infusion of DN Tregs did not
induce xenogeneic GVHD. DN Tregs migrate to lymphatic organs and other tissues in a
similar fashion to PBMCs. However, DN Tregs only expanded marginally in vivo, as
evident by the decreasing numbers of DN Tregs infiltrating the organs of NSG mice. This
unresponsiveness of DN Tregs to xeno-antigens may be attributable to the their lack of
expression of TCR co-receptors, CD4 and CD8, on their cell surface, which is a very
attractive and important feature in the clinical applications of DN Tregs.
DN Tregs appear to have an interesting property of being able to inhibit GVHD,
while mediating cytotoxicity towards cancer cells. Harnessing their potential in the clinic
as candidates for adoptive cellular therapy would have enormous implications in the
prevention of GVHD in leukemia patients undergoing allogeneic HSCT, as well as aid
solid organ transplant recipients with pre-existing malignancies in remission. Further
understanding of DN Treg mechanisms of action may help in development of novel
immunotherapies.
93
4.2. Future Directions
Understanding of the DN Treg development and biology is severely lagging
behind other immune cells due to their low frequency, the lack of markers to
identify/isolate them, and the lack of effective expansion methods. Over the past years
many studies of murine DN Tregs have surfaced and there is no doubt that they are strong
suppressors of the immune system. However, there are still a lot of unanswered questions
with regards to human DN Tregs. Herein, we presented a method for a large-scale
expansion of human DN Tregs that will hopefully invite more researchers to explore in-
depth the DN Treg biology and function.
The presence DN Tregs have been found in many organs, but their niche has not
been completely identified. Many questions remain, for instance what activates DN
Tregs? What drives their proliferation? Since they are capable of producing many
different cytokines upon non-specific stimulation, what cytokines get released under what
pathophysiological settings? How do DN Tregs interact and communicate with other
cells in the body? The answers to the stated questions may not only advance our
understanding of the mechanisms of suppression mediated by DN Tregs, but may also
facilitate the clinical application of DN Tregs. In the future, infusion of DN Tregs into
human patients may provide necessary means to induce transplantation tolerance along
with the reduced risk of developing GVHD. In turn this would lead to improved patient
survival and decreased likelihood of graft rejection, ultimately reducing health care costs
and enhancing the well being of transplant patients. However, before this could be
possible the studies outlined below will aid in understanding of these cells brining them
one step closer to the clinic.
94
4.2.1. To Identify Markers That Are Critical for DN Treg Function
A major challenge that is yet to be addressed is identifying markers that are
associated with truly regulatory, or truly cytotoxic DN Tregs, as well as markers that
would set ‘functional’ DN Tregs apart from the ‘pathological loss of function’ DN Tregs
in ALPS, or CD4+ and CD8+ T cells that have downregulated their co-receptors.
Furthermore, the origin and thymic development of DN Tregs are still largely unknown.
Although multiple mechanisms of suppression have been described from murine models
to date no specific mechanisms have been proposed for human DN Tregs. We and others
have demonstrated that DN Treg suppression is cell contact-dependent, but does not
involve killing of the responder T cells (Voelkl et al., 2011). Furthermore, DN Tregs
require TCR signalling and de novo protein synthesis (Fischer et al., 2005; Voelkl et al.,
2011) implying that certain protein molecules are produced. These may include cell
surface proteins and/or soluble factors. Using flow cytometry, DN Tregs were examined
for the multitude of surface proteins found to be involved in inhibitory function in their
murine counterparts, as well as other Treg subsets, such as Fas/FasL, CTLA-4,
perforin/granzyme and PD-1. Although DN Tregs are capable of producing
perforin/granzyme upon TCR stimulation, DN Treg did not induce cell death in the
responder cells, thus eliminating the possibility of perforin/granzyme involvement.
Moreover, we have found that IL-2 + IL-7-grown DN Tregs had lower expression of
CTLA-4 and PD-1 as compared to IL-2 + IL-15-grown DN Tregs (data not shown),
suggesting other molecules are involved in the suppression mechanisms. Lastly, DN
Tregs did not express Fas on their cell surface.
To identify new molecules that may be associated with DN Treg-mediated
suppression, we can take advantage of high-throughput screening (HTS) technology that
allows for screening of over 370 cell surface molecules including a majority of
functionally known CD molecules, as well as non-CD molecules including a variety of
TCR and HLA molecules. Experiments can be performed on any flow cytometer with a
high-throughput sampler. Since DN Tregs grown in the presence of IL-2 + IL-7 or IL-2 +
IL-15 manifest different inhibitory abilities, we hypothesize that the differences in the
95
receptor expression between the two groups may pinpoint to the molecules involved in
the mechanism of suppression of DN Tregs, as well the expression of markers unique to
DN Tregs if compared to other subtypes of T cells from the same donor. An even better
comparison of the cell surface molecules may be between rapamycin-treated and
untreated DN Tregs. However, this comparison may be proven problematic due to the
differences in the receptor expression on the cell surface rather than presence or absence
of individual markers. An alternative method could be an exploitation of microarray-
based gene expression profiling, which can be used to identify genes whose expression
are changed in response to either IL-2 + IL-7 or IL-2 + IL-15 cytokine treatment during
the expansion, or rapamycin-treatment post expansion. The comparison of gene
expression in these two conditions may pinpoint to the molecules involved in the
suppression.
4.2.2. To Determine the Effects of DN Treg-Immunosuppressive Agents Combination Therapy on the Treatment of Xenogeneic GVHD
Another obstacle that would hinder clinical application of DN Tregs is the fact
that DN Tregs do not prevent GVHD from occurring in a xenogeneic mouse model, but
rather the infusion of DN Tregs delays its onset and severity. Therefore, modified
treatment regimens should be explored that would ideally prevent GVHD from occurring
all together. Since rapamycin inhibits activation of B cells and T cells, and we have
shown that the pre-treatment of DN Tregs with rapamycin prior to the infusion augments
their suppressive function, it would be worthwhile to evaluate the effects of the co-
injection of rapamycin with DN Tregs in the xenogeneic GVHD model. We suspect that
the rapamycin would hinder the proliferation of conventional T cells, while at the same
time promote the regulatory function of DN Tregs, rendering them more
immunosuppressive.
Furthermore, the data from the tracking study revealed that DN Tregs proliferate
poorly in vivo likely due to the lack of human cytokines that would support their survival
96
and proliferation. Therefore, the next step should involve evaluation of injection of
exogenous IL-2 and/or IL-7 on the suppressive function, proliferation and survival of
human DN Tregs in mice. Furthermore, the in vitro mechanisms of suppression may be
different from the mechanisms engaged during suppression in vivo, as it has been
exemplified by nTregs and their effect of IL-10 production. Since we have showed that
DN Tregs are capable of producing copious amounts of IL-10, yet IL-10 does not seem to
play a role in DN Treg-mediated immunosuppression in vitro, it is important to evaluate
whether IL-10 plays any role in vivo as it may for example induce generation of iTregs
from conventional cells. Additionally, DN Tregs seem to reduce the ability of responder
T cells to secrete IFN-γ. Therefore, it would be worthwhile to measure the differences in
the levels of pro- and anti-inflammatory cytokines in the mouse sera between the DN
Treg-treated and untreated groups. This may be achieved by using a simple ELISA
method or with the help of Luminex platform that allows for simultaneous measurement
of more than 27 different cytokines. These studies will shed more light on the function of
DN Tregs in vivo.
4.2.3. To Determine Whether Human DN Tregs Suppress DCs
Both donor and recipient DCs play a critical role in mounting allogeneic immune
responses through direct and indirect antigen presentation. Moreover, both human and
mouse nTregs and iTregs can suppress allo-reactive CD4+ and CD8+ T cell proliferation
by down regulating CD80/CD86 on mDCs through CTLA-4. Similarly, murine DN Tregs
also express a high level of CTLA-4 that is critical for down regulation of CD80 and
CD86 expression on DCs, as this capacity is lost in murine DN Tregs from CTLA-4
deficient mice (Kowalczyk et al., 2014). Since human DN Tregs do not express high
levels of CTLA-4, whether human DN Tregs can alter the function or kill allo-Ag-
expressing DCs is still unknown. To see if human DN Tregs are cytotoxic to allogeneic
DCs, allo-Ag-specific DCs need to be generated first and co-cultured with DN Tregs in
an in vitro suppression assay. After 4 days, DCs can be stained with AnnexinV and 7-
AAD to determine their viability and whether they are susceptible to DN Tregs. If DCs
97
would be found to be alive, to study the function of the DCs after the co-culture human
DN Tregs will be removed using CD3-coated beads and the ability of remaining DCs to
stimulate naïve CD4+ and CD8+ T cell proliferation will be assessed by CFSE dilution.
These studies will determine if human DN Tregs are cytotoxic towards DCs and/or
whether DN Tregs can impair their function despite the negligible expression of CTLA-4.
4.2.4. To Determine Whether Trogocytosis is Critical for Human DN Treg Function
In this study, we presented for the first time the ability of polyclonally stimulated
ex vivo expanded human DN Tregs to suppress the responder cells in vitro. However, it is
a well-known phenomenon that Ag-specific Tregs are more effective at suppressing
allograft rejection than polyclonally activated Tregs (Putnam et al., 2013; Sagoo et al.,
2011). Furthermore, CD40 ligand-stimulated human B cells are more potent than DCs at
inducing proliferation of donor-specific human Tregs (Tang et al., 2012; Zheng et al.,
2010). However, why activated B cells are better than DCs at generating donor-specific
human Tregs is not known. Since we had some success in generating (but not expanding)
allo-Ag-specific DN Tregs using allogeneic B cells, it would be important to determine
whether differences in the acquisition of allo-Ag from donor APC alter human DN Treg
ability to mount Ag-specific suppression. To this end, human DN Tregs can be isolated
from HLA-A2− donors and co-cultured with either CD40L-activated HLA-A2+ donor B
cells or mDCs. The expression of HLA-A2 on human DN Tregs can be measured using
anti-HLA-A2 Ab at varying time points after co-culture. To determine the importance of
acquired allo-Ag in human DN Tregs-mediated suppression, A2+ APC-primed HLA-A2+
and A2− DN Tregs can be sorted by FACS and used to suppress autologous CD4+ and
CD8+ T cells in vitro. DN Tregs from the same donor activated by the anti-CD3 and
aAPC that do not express HLA (Butler et al., 2012) will be used as a control. The
specificity of acquisition of allo-Ag can be determined by blocking TCR-HLA
interactions with anti-HLA-A2 Ab. Together, these studies will establish whether
CD40L-activated B cells are better than mDCs at donating allo-Ag to human DN Tregs
98
and enhancing their suppressive function and whether acquisition of allo-Ag by human
DN Tregs is critical for their ability to suppress allo-reactive T cell responses in vitro.
4.2.5. To Determine Signalling Pathways That Govern DN Tregs
It is unknown, which nutrients and growth factors are necessary for DN Treg
function. DN Treg preference for metabolic pathways is also unknown. Since DN Tregs
behave differently when grown in the presence of IL-7 versus IL-15, it strongly suggests
that common gamma chain receptors are important modulators. To determine the
signalling pathways that govern DN Treg function, a series of experiments measuring the
downstream signalling molecules should be performed. Developed in the last decade, the
phosphoflow technique could be used to track and measure the phosphorylation of the
key signalling molecules such as “STATs, members of the MAPK and stress-activated
protein kinase families, other cell survival kinases and adaptor molecules” (Wu et al.,
2010). This technique would determine which pathways are involved when different
growth factors or antigen stimuli are present, together with the metabolic pathway
experiments that would define the preference for nutrients and measure metabolic waste,
and respiration, would shine the light on signalling pathways and metabolic requirements
of DN Tregs. Furthermore, qPCR could be used to determine which transcription factors
play a role in DN Treg function.
99
4.3. Conclusion
Presented here is a novel method that allows for large-scale ex vivo expansion of
human DN Tregs. Using this protocol, up to ~109 DN Tregs of very high purity can be
obtained from 50 to 100 ml of blood within 3 weeks. Importantly, these ex vivo expanded
DN Tregs are functional in vitro as they suppress the proliferation of autologous T cells
and B cells.
For the first time, we show that DN Treg suppressive function is modulated by the
presence of cytokines during the expansion. Addition of IL-7 to the expansion co-culture
was found to increase suppressive function of DN Tregs, whereas addition of IL-15
promoted expansion of DN Tregs, but resulted in hampered suppressive function. The
mechanism of DN Treg suppression is cell-contact-dependent and results in reduced
production of IFN-γ by CD4+ and CD8+ T cells. However, DN Tregs do not supress by
inducing apoptosis in the responder cells, even though they are capable of cytotoxic
function as they successfully eliminated various leukemic and lung cancer cells.
The function of ex vivo expanded human DN Tregs in vivo has been investigated
for the first time by employing xenogeneic GVHD mouse model. The infusion of human
PBMC into immunodeficient NSG mice induced severe acute xenogeneic GVHD.
However, injection of 3 doses of DN Tregs within a week of initiation of GVHD resulted
in significantly delayed onset of GVHD and significant reduction in weight loss.
Furthermore, a single infusion of ex vivo expanded DN Tregs to NSG mice did not cause
GVHD or tissue damage in the recipient, highlighting the safety of DN Tregs.
Lastly, pre-treatment of ex vivo expanded DN Tregs with an mTOR inhibitor
rapamycin significantly improved regulatory function of DN Tregs both in vitro and in
vivo. These finding hold important implications for the future use of DN Tregs in the
clinic, since it can significantly reduce the amount of DN Tregs required for infusion,
ultimately decreasing the need to expand large amount of cells, which would widen the
feasibility of DN Treg application in adaptive cellular therapy.
100
The ability to obtain large quantity of pure and functional DN Tregs not only
makes it possible to study their function and mechanisms of action in vivo, but also
highlights the possibility of using ex vivo expanded human DN Tregs in immunotherapy.
DN Tregs have enormous potential of acting as double-edge swords in the treatment and
prophylaxis of GVHD, especially after BMT for leukemia, as DN Tregs would not only
inhibit GVHD, but also promote anti-cancer effect in these patients.
101
REFERENCES
Akdis, M (2008). T-cell tolerance to inhaled allergens: mechanisms and therapeutic approaches. Expert Opin Biol Ther 8, 769-777.
Allgauer, A, Schreiner, E, Ferrazzi, F, Ekici, AB, Gerbitz, A, Mackensen, A, and Volkl, S (2015). IL-7 abrogates the immunosuppressive function of human double-negative T cells by activating Akt/mTOR signaling. J Immunol 195, 3139-3148.
Andersson, J, Tran, DQ, Pesu, M, Davidson, TS, Ramsey, H, O'Shea, JJ, and Shevach, EM (2008). CD4(+) FoxP3(+) regulatory T cells confer infectious tolerance in a TGF-beta-dependent manner. J Exp Med 205, 1975-1981.
Andreani, M, Gianolini, ME, Testi, M, Battarra, M, Tiziana, G, Morrone, A, Sodani, P, Lucarelli, G, Roncarolo, MG, and Gregori, S (2014). Mixed chimerism evolution is associated with T regulatory type 1 (Tr1) cells in a beta-thalassemic patient after haploidentical haematopoietic stem cell transplantation. Chimerism 5, 75-79.
Annes, JP, Munger, JS, and Rifkin, DB (2003). Making sense of latent TGF-beta activation. J Cell Sci 116, 217-224.
Antonioli, L, Pacher, P, Vizi, ES, and Hasko, G (2013). CD39 and CD73 in immunity and inflammation. Trends Mol Med 19, 355-367.
Asanuma, S, Tanaka, J, Sugita, J, Kosugi, M, Shiratori, S, Wakasa, K, Shono, Y, Shigematsu, A, Kondo, T, Kobayashi, T, et al. (2011). Expansion of CD4(+)CD25(+) regulatory T cells from cord blood CD4(+) cells using the common gamma-chain cytokines (IL-2 and IL-15) and rapamycin. Ann Hematol 90, 617-624.
Asseman, C, Mauze, S, Leach, MW, Coffman, RL, and Powrie, F (1999). An essential role for interleukin 10 in the function of regulatory T cells that inhibit intestinal inflammation. J Exp Med 190, 995-1004.
Asseman, C, Read, S, and Powrie, F (2003). Colitogenic Th1 cells are present in the antigen-experienced T cell pool in normal mice: control by CD4(+) regulatory T cells and IL-10. J Immunol 171, 971-978.
Awasthi, A, Carrier, Y, Peron, JP, Bettelli, E, Kamanaka, M, Flavell, RA, Kuchroo, VK, Oukka, M, and Weiner, HL (2007). A dominant function for interleukin 27 in generating interleukin 10-producing anti-inflammatory T cells. Nat Immunol 8, 1380-1389.
102
Bacchetta, R, Bigler, M, Touraine, JL, Parkman, R, Tovo, PA, Abrams, J, de Waal Malefyt, R, de Vries, JE, and Roncarolo, MG (1994). High levels of interleukin 10 production in vivo are associated with tolerance in SCID patients transplanted with HLA mismatched hematopoietic stem cells. J Exp Med 179, 493-502.
Baecher-Allan, C, Brown, JA, Freeman, GJ, and Hafler, DA (2001). CD4(+)CD25(high) regulatory cells in human peripheral blood. J Immunol 167, 1245-1253.
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.
Barzaghi, F, Passerini, L, and Bacchetta, R (2012). Immune dysregulation, polyendocrinopathy, enteropathy, x-linked syndrome: a paradigm of immunodeficiency with autoimmunity. Front Immunol 3, 211.
Battaglia, M, Stabilini, A, Draghici, E, Gregori, S, Mocchetti, C, Bonifacio, E, and Roncarolo, MG (2006a). Rapamycin and interleukin-10 treatment induces T regulatory type 1 cells that mediate antigen-specific transplantation tolerance. Diabetes 55, 40-49.
Battaglia, M, Stabilini, A, Migliavacca, B, Horejs-Hoeck, J, Kaupper, T, and Roncarolo, MG (2006b). 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.
Battaglia, M, Stabilini, A, and Roncarolo, MG (2005). Rapamycin selectively expands CD4(+)CD25(+)FoxP3(+) regulatory T cells. Blood 105, 4743-4748.
Baughman, EJ, Mendoza, JP, Ortega, SB, Ayers, CL, Greenberg, BM, Frohman, EM, and Karandikar, NJ (2011). Neuroantigen-specific CD8(+) regulatory T-cell function is deficient during acute exacerbation of multiple sclerosis. J Autoimmun 36, 115-124.
Ben Ahmed, M, Belhadj Hmida, N, Moes, N, Buyse, S, Abdeladhim, M, Louzir, H, and Cerf-Bensussan, N (2009). IL-15 renders conventional lymphocytes resistant to suppressive functions of regulatory T cells through activation of the phosphatidylinositol 3-kinase pathway. J Immunol 182, 6763-6770.
Bennett, CL, Christie, J, Ramsdell, F, Brunkow, ME, Ferguson, PJ, Whitesell, L, Kelly, TE, Saulsbury, FT, Chance, PF, and Ochs, HD (2001). The immune dysregulation, polyendocrinopathy, enteropathy, X-linked syndrome (IPEX) is caused by mutations of FOXP3. Nat Genet 27, 20-21.
103
Bergmann, C, Strauss, L, Zeidler, R, Lang, S, and Whiteside, TL (2007). Expansion and characteristics of human T regulatory type 1 cells in co-cultures simulating tumor microenvironment. Cancer Immunol Immunother 56, 1429-1442.
Bilate, AB, and Lafaille, JJ (2011). It takes two to tango. Immunity 35, 6-8.
Bjarnadottir, U, Lemarquis, AL, Halldorsdottir, S, Freysdottir, J, and Ludviksson, BR (2014). The suppressive function of human CD8(+) iTregs is inhibited by IL-1beta and TNF-alpha. Scand J Immunol 80, 313-322.
Boor, PP, Metselaar, HJ, Jonge, S, Mancham, S, van der Laan, LJ, and Kwekkeboom, J (2011). Human plasmacytoid dendritic cells induce CD8(+) LAG-3(+) Foxp3(+) CTLA-4(+) regulatory T cells that suppress allo-reactive memory T cells. Eur J Immunol 41, 1663-1674.
Brimnes, J, Allez, M, Dotan, I, Shao, L, Nakazawa, A, and Mayer, L (2005). Defects in CD8+ regulatory T cells in the lamina propria of patients with inflammatory bowel disease. J Immunol 174, 5814-5822.
Bristeau-Leprince, A, Mateo, V, Lim, A, Magerus-Chatinet, A, Solary, E, Fischer, A, Rieux-Laucat, F, and Gougeon, ML (2008). Human TCR alpha/beta+ CD4-CD8- double-negative T cells in patients with autoimmune lymphoproliferative syndrome express restricted Vbeta TCR diversity and are clonally related to CD8+ T cells. J Immunol 181, 440-448.
Brooks, EG, Balk, SP, Aupeix, K, Colonna, M, Strominger, JL, and Groh-Spies, V (1993). Human T-cell receptor (TCR) alpha/beta(+) CD4(-)CD8(-) T cells express oligoclonal TCRs, share junctional motifs across TCR V beta-gene families, and phenotypically resemble memory T cells. Proc Natl Acad Sci U S A 90, 11787-11791.
Browning, JL (2006). B cells move to centre stage: novel opportunities for autoimmune disease treatment. Nat Rev Drug Discov 5, 564-576.
Bruley-Rosset, M, Miconnet, I, Canon, C, and Halle-Pannenko, O (1990). Mlsa generated suppressor cells. I. Suppression is mediated by double-negative (CD3+CD5+CD4-CD8-) alpha/beta T cell receptor-bearing cells. J Immunol 145, 4046-4052.
Brunkow, ME, Jeffery, EW, Hjerrild, KA, Paeper, B, Clark, LB, Yasayko, SA, Wilkinson, JE, Galas, D, Ziegler, SF, and Ramsdell, F (2001). Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat Genet 27, 68-73.
104
Brunstein, CG, Miller, JS, Cao, Q, McKenna, DH, Hippen, KL, Curtsinger, J, Defor, T, Levine, BL, June, CH, Rubinstein, P, et al. (2011). Infusion of ex vivo expanded T regulatory cells in adults transplanted with umbilical cord blood: safety profile and detection kinetics. Blood 117, 1061-1070.
Burt, RK, Balabanov, R, Han, X, Sharrack, B, Morgan, A, Quigley, K, Yaung, K, Helenowski, IB, Jovanovic, B, Spahovic, D, et al. (2015). Association of nonmyeloablative hematopoietic stem cell transplantation with neurological disability in patients with relapsing-remitting multiple sclerosis. JAMA 313, 275-284.
Butler, MO, Imataki, O, Yamashita, Y, Tanaka, M, Ansen, S, Berezovskaya, A, Metzler, G, Milstein, MI, Mooney, MM, Murray, AP, et al. (2012). Ex vivo expansion of human CD8(+) T cells using autologous CD4(+) T cell help. PLoS One 7, e30229.
Cederbom, L, Hall, H, and Ivars, F (2000). CD4(+)CD25(+) regulatory T cells down-regulate co-stimulatory molecules on antigen-presenting cells. Eur J Immunol 30, 1538-1543.
Chao, CC, Jensen, R, and Dailey, MO (1997). Mechanisms of L-selectin regulation by activated T cells. J Immunol 159, 1686-1694.
Chen, W, Ford, MS, Young, KJ, Cybulsky, MI, and Zhang, L (2003a). Role of double-negative regulatory T cells in long-term cardiac xenograft survival. J Immunol 170, 1846-1853.
Chen, W, Ford, MS, Young, KJ, and Zhang, L (2003b). Infusion of in vitro-generated DN T regulatory cells induces permanent cardiac allograft survival in mice. Transplant Proc 35, 2479-2480.
Chen, W, Jin, W, Cook, M, Weiner, HL, and Wahl, SM (1998a). Oral delivery of group A streptococcal cell walls augments circulating TGF-beta and suppresses streptococcal cell wall arthritis. J Immunol 161, 6297-6304.
Chen, W, Jin, W, and Wahl, SM (1998b). Engagement of cytotoxic T lymphocyte-associated antigen 4 (CTLA-4) induces transforming growth factor beta (TGF-beta) production by murine CD4(+) T cells. J Exp Med 188, 1849-1857.
Chen, W, and Konkel, JE (2010). TGF-beta and 'adaptive' Foxp3(+) regulatory T cells. J Mol Cell Biol 2, 30-36.
Chen, WH, Diao, J, Stepkowski, SM, and Zhang, L (2007). Both infiltrating regulatory T cells and insufficient antigen presentation are involved in long-term cardiac xenograft survival. Journal of Immunology 179, 1542-1548.
105
Chen, WH, Zhou, DJ, Torrealba, JR, Waddell, TK, Grant, D, and Zhang, L (2005). Donor lymphocyte infusion induces long-term donor-specific cardiac xenograft survival through activation of recipient double-negative regulatory T cells. Journal of Immunology 175, 3409-3416.
Chen, X, Fang, L, Song, S, Guo, TB, Liu, A, and Zhang, JZ (2009). Thymic regulation of autoimmune disease by accelerated differentiation of Foxp3(+) regulatory T cells through IL-7 signaling pathway. J Immunol 183, 6135-6144.
Chen, Z, Han, Y, Gu, Y, Liu, Y, Jiang, Z, Zhang, M, and Cao, X (2013). CD11c(high)CD8(+) regulatory T cell feedback inhibits CD4(+) T cell immune response via Fas ligand-Fas pathway. J Immunol 190, 6145-6154.
Chinen, T, Kannan, AK, Levine, AG, Fan, X, Klein, U, Zheng, Y, Gasteiger, G, Feng, Y, Fontenot, JD, and Rudensky, AY (2016). An essential role for the IL-2 receptor in Treg cell function. Nat Immunol 17, 1322-1333.
Choi, SW, Levine, JE, and Ferrara, JL (2010). Pathogenesis and management of graft-versus-host disease. Immunol Allergy Clin North Am 30, 75-101.
Cohen, PL, and Eisenberg, RA (1991). Lpr and gld: single gene models of systemic autoimmunity and lymphoproliferative disease. Annu Rev Immunol 9, 243-269.
Collison, LW, Workman, CJ, Kuo, TT, Boyd, K, Wang, Y, Vignali, KM, Cross, R, Sehy, D, Blumberg, RS, and Vignali, DA (2007). The inhibitory cytokine IL-35 contributes to regulatory T-cell function. Nature 450, 566-569.
Cooper, GS, Bynum, ML, and Somers, EC (2009). Recent insights in the epidemiology of autoimmune diseases: improved prevalence estimates and understanding of clustering of diseases. J Autoimmun 33, 197-207.
Couper, KN, Blount, DG, and Riley, EM (2008). IL-10: the master regulator of immunity to infection. J Immunol 180, 5771-5777.
Cowley, SC, Hamilton, E, Frelinger, JA, Su, J, Forman, J, and Elkins, KL (2005). CD4(-)CD8(-) T cells control intracellular bacterial infections both in vitro and in vivo. Journal of Experimental Medicine 202, 309-319.
Crispin, JC, Oukka, M, Bayliss, G, Cohen, RA, Van Beek, CA, Stillman, IE, Kyttaris, VC, Juang, YT, and Tsokos, GC (2008). Expanded double negative T cells in patients with systemic lupus erythematosus produce IL-17 and infiltrate the kidneys. J Immunol 181, 8761-8766.
106
Crispin, JC, and Tsokos, GC (2009). Human TCR-alpha beta(+) CD4(-) CD8(-) T cells can derive from CD8(+) T cells and display an inflammatory effector phenotype. J Immunol 183, 4675-4681.
Curotto de Lafaille, MA, and Lafaille, JJ (2009). Natural and adaptive Foxp3+ regulatory T cells: more of the same or a division of labor? Immunity 30, 626-635.
Dai, H, Wan, N, Zhang, S, Moore, Y, Wan, F, and Dai, Z (2010). Cutting edge: programmed death-1 defines CD8(+)CD122(+) T cells as regulatory versus memory T cells. J Immunol 185, 803-807.
Davignon, JL, Cohen, PL, and Eisenberg, RA (1988). Rapid T cell receptor modulation accompanies lack of in vitro mitogenic responsiveness of double negative T cells to anti-CD3 monoclonal antibody in MRL/Mp-lpr mice. J Immunol 141, 1848-1854.
Deaglio, S, Dwyer, KM, Gao, W, Friedman, D, Usheva, A, Erat, A, Chen, JF, 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.
Dean, RM, Fry, T, Mackall, C, Steinberg, SM, Hakim, F, Fowler, D, Odom, J, Foley, J, Gress, R, and Bishop, MR (2008). Association of serum interleukin-7 levels with the development of acute graft-versus-host disease. J Clin Oncol 26, 5735-5741.
Derks, RA, Jankowska-Gan, E, Xu, Q, and Burlingham, WJ (2007). Dendritic cell type determines the mechanism of bystander suppression by adaptive T regulatory cells specific for the minor antigen HA-1. J Immunol 179, 3443-3451.
Di Ianni, M, Falzetti, F, Carotti, A, Terenzi, A, Castellino, F, Bonifacio, E, Del Papa, B, Zei, T, Ostini, RI, Cecchini, D, et al. (2011). Tregs prevent GVHD and promote immune reconstitution in HLA-haploidentical transplantation. Blood 117, 3921-3928.
Di Santo, JP, Kuhn, R, and Muller, W (1995). Common cytokine receptor gamma chain (gamma c)-dependent cytokines: understanding in vivo functions by gene targeting. Immunol Rev 148, 19-34.
Dijke, IE, Caliskan, K, Klepper, M, de Kuiper, R, Balk, AH, Maat, AP, Weimar, W, and Baan, CC (2009). Donor-specific immune regulation by CD8 lymphocytes expanded from rejecting human cardiac allografts. Am J Transplant 9, 397-403.
Dijke, IE, Hoeppli, RE, Ellis, T, Pearcey, J, Huang, Q, McMurchy, AN, Boer, K, Peeters, AM, Aubert, G, Larsen, I, et al. (2016). Discarded Human Thymus Is a Novel Source of Stable and Long-Lived Therapeutic Regulatory T Cells. Am J Transplant 16, 58-71.
107
Dugas, V, Beauchamp, C, Chabot-Roy, G, Hillhouse, EE, and Lesage, S (2010). Implication of the CD47 pathway in autoimmune diabetes. J Autoimmun 35, 23-32.
Dummer, W, Niethammer, AG, Baccala, R, Lawson, BR, Wagner, N, Reisfeld, RA, and Theofilopoulos, AN (2002). T cell homeostatic proliferation elicits effective antitumor autoimmunity. J Clin Invest 110, 185-192.
Duncan, B, Nazarov-Stoica, C, Surls, J, Kehl, M, Bona, C, Casares, S, and Brumeanu, TD (2010). Double negative (CD3+CD4-CD8-) TCR alphabeta splenic cells from young NOD mice provide long-lasting protection against type 1 diabetes. PLoS One 5, e11427.
Egerton, M, and Scollay, R (1990). Intrathymic selection of murine TCR alpha beta(+)CD4(-)CD8(-) thymocytes. Int Immunol 2, 157-163.
ElKassar, N, and Gress, RE (2010). An overview of IL-7 biology and its use in immunotherapy. J Immunotoxicol 7, 1-7.
Endharti, AT, Rifa, IM, Shi, Z, Fukuoka, Y, Nakahara, Y, Kawamoto, Y, Takeda, K, Isobe, K, and Suzuki, H (2005). Cutting edge: CD8(+)CD122(+) regulatory T cells produce IL-10 to suppress IFN-gamma production and proliferation of CD8(+) T cells. J Immunol 175, 7093-7097.
Ermann, J, Hoffmann, P, Edinger, M, Dutt, S, Blankenberg, FG, Higgins, JP, Negrin, RS, Fathman, CG, and Strober, S (2005). Only the CD62L(+) subpopulation of CD4(+)CD25(+) regulatory T cells protects from lethal acute GVHD. Blood 105, 2220-2226.
Eusebio, M, Kuna, P, Kraszula, L, Kupczyk, M, and Pietruczuk, M (2015). The relative values of CD8(+)CD25(+)Foxp3(bright) Treg cells correlate with selected lung function parameters in asthma. Int J Immunopathol Pharmacol 28, 218-226.
Fischer, K, Voelkl, S, Heymann, J, Przybylski, GK, Mondal, K, Laumer, M, Kunz-Schughart, L, Schmidt, CA, Andreesen, R, and Mackensen, A (2005). Isolation and characterization of human antigen-specific TCR alpha beta(+) CD4(-)CD8(-) double-negative regulatory T cells. Blood 105, 2828-2835.
Fitzgerald, DC, Zhang, GX, El-Behi, M, Fonseca-Kelly, Z, Li, H, Yu, S, Saris, CJ, Gran, B, Ciric, B, and Rostami, A (2007). Suppression of autoimmune inflammation of the central nervous system by interleukin 10 secreted by interleukin 27-stimulated T cells. Nat Immunol 8, 1372-1379.
Fontenot, JD, Gavin, MA, and Rudensky, AY (2003). Foxp3 programs the development and function of CD4(+)CD25(+) regulatory T cells. Nat Immunol 4, 330-336.
108
Ford, MS, Chen, W, Wong, S, Li, C, Vanama, R, Elford, AR, Asa, SL, Ohashi, PS, and Zhang, L (2007). Peptide-activated double-negative T cells can prevent autoimmune type-1 diabetes development. EurJ Immunol 37, 2234-2241.
Ford, MS, Young, KJ, Zhang, Z, Ohashi, PS, and Zhang, L (2002). The immune regulatory function of lymphoproliferative double negative T cells in vitro and in vivo. J Exp Med 196, 261-267.
Ford, MS, Zhang, ZX, Chen, W, and Zhang, L (2006). Double-negative T regulatory cells can develop outside the thymus and do not mature from CD8(+) T cell precursors. J Immunol 177, 2803-2809.
Forrester, JV, Xu, H, Lambe, T, and Cornall, R (2008). Immune privilege or privileged immunity? Mucosal Immunol 1, 372-381.
Fox, CJ, Hammerman, PS, and Thompson, CB (2005). Fuel feeds function: energy metabolism and the T-cell response. Nat Rev Immunol 5, 844-852.
Francisco, LM, Sage, PT, and Sharpe, AH (2010). The PD-1 pathway in tolerance and autoimmunity. Immunol Rev 236, 219-242.
Fukaura, H, Kent, SC, Pietrusewicz, MJ, Khoury, SJ, Weiner, HL, and Hafler, DA (1996). Induction of circulating myelin basic protein and proteolipid protein-specific transforming growth factor-beta1-secreting Th3 T cells by oral administration of myelin in multiple sclerosis patients. J Clin Invest 98, 70-77.
Gagliani, N, Magnani, CF, Huber, S, Gianolini, ME, Pala, M, Licona-Limon, P, Guo, B, Herbert, DR, Bulfone, A, Trentini, F, et al. (2013). Coexpression of CD49b and LAG-3 identifies human and mouse T regulatory type 1 cells. Nat Med 19, 739-746.
Galea, I, Bechmann, I, and Perry, VH (2007). What is immune privilege (not)? Trends Immunol 28, 12-18.
Gao, JF, McIntyre, MS, Juvet, SC, Diao, J, Li, X, Vanama, RB, Mak, TW, Cattral, MS, and Zhang, L (2011). Regulation of antigen-expressing dendritic cells by double negative regulatory T cells. Eur J Immunol 41, 2699-2708.
Germain, RN (2008). Special regulatory T-cell review: A rose by any other name: from suppressor T cells to Tregs, approbation to unbridled enthusiasm. Immunology 123, 20-27.
Gershon, RK (1975). A disquisition on suppressor T cells. Transplant Rev 26, 170-185.
109
Gershon, RK, and Kondo, K (1971). Infectious immunological tolerance. Immunology 21, 903-914.
Gliwinski, M, Iwaszkiewicz-Grzes, D, and Trzonkowski, P (2017). Cell-based therapies with T regulatory cells. BioDrugs.
Golovina, TN, Mikheeva, T, Brusko, TM, Blazar, BR, Bluestone, JA, and Riley, JL (2011). Retinoic acid and rapamycin differentially affect and synergistically promote the ex vivo expansion of natural human T regulatory cells. PLoS One 6, e15868.
Golshayan, D, Jiang, S, Tsang, J, Garin, MI, Mottet, C, and Lechler, RI (2007). In vitro-expanded donor alloantigen-specific CD4(+)CD25(+) regulatory T cells promote experimental transplantation tolerance. Blood 109, 827-835.
Grossman, WJ, Verbsky, JW, Barchet, W, Colonna, M, Atkinson, JP, and Ley, TJ (2004). Human T regulatory cells can use the perforin pathway to cause autologous target cell death. Immunity 21, 589-601.
Groux, H, O'Garra, A, Bigler, M, Rouleau, M, Antonenko, S, de Vries, JE, and Roncarolo, MG (1997). A CD4(+) T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389, 737-742.
Hamad, AR (2010). Analysis of gene profile, steady state proliferation and apoptosis of double-negative T cells in the periphery and gut epithelium provides new insights into the biological functions of the Fas pathway. Immunol Res 47, 134-142.
Haribhai, D, Williams, JB, Jia, S, Nickerson, D, Schmitt, EG, Edwards, B, Ziegelbauer, J, Yassai, M, Li, SH, Relland, LM, et al. (2011). A requisite role for induced regulatory T cells in tolerance based on expanding antigen receptor diversity. Immunity 35, 109-122.
Hartemann, A, Bensimon, G, Payan, CA, Jacqueminet, S, Bourron, O, Nicolas, N, Fonfrede, M, Rosenzwajg, M, Bernard, C, and Klatzmann, D (2013). Low-dose interleukin 2 in patients with type 1 diabetes: a phase 1/2 randomised, double-blind, placebo-controlled trial. Lancet Diabetes Endocrinol 1, 295-305.
He, J, Zhang, X, Wei, Y, Sun, X, Chen, Y, Deng, J, Jin, Y, Gan, Y, Hu, X, Jia, R, et al. (2016). Low-dose interleukin-2 treatment selectively modulates CD4(+) T cell subsets in patients with systemic lupus erythematosus. Nat Med 22, 991-993.
He, KM, Ma, Y, Wang, S, Min, WP, Zhong, R, Jevnikar, A, and Zhang, ZX (2007). Donor double-negative Treg promote allogeneic mixed chimerism and tolerance. Eur J Immunol 37, 3455-3466.
110
Hess, AD (2006). Modulation of graft-versus-host disease: role of regulatory T lymphocytes. Biol Blood Marrow Transplant 12, 13-21.
Hillhouse, EE, Beauchamp, C, Chabot-Roy, G, Dugas, V, and Lesage, S (2010). Interleukin-10 limits the expansion of immunoregulatory CD4(-)CD8(-) T cells in autoimmune-prone non-obese diabetic mice. Immunol Cell Biol 88, 771-780.
Hillhouse, EE, and Lesage, S (2013). A comprehensive review of the phenotype and function of antigen-specific immunoregulatory double negative T cells. JAutoimmun 40, 58-65.
Hofer, T, Krichevsky, O, and Altan-Bonnet, G (2012). Competition for IL-2 between regulatory and effector T cells to chisel immune responses. Front Immunol 3, 268.
Hogquist, KA, Baldwin, TA, and Jameson, SC (2005). Central tolerance: learning self-control in the thymus. Nat Rev Immunol 5, 772-782.
Hori, S, Nomura, T, and Sakaguchi, S (2003). Control of regulatory T cell development by the transcription factor Foxp3. Science 299, 1057-1061.
Hossain, MS, Takimoto, H, Ninomiya, T, Yoshida, H, Kishihara, K, Matsuzaki, G, Kimura, G, and Nomoto, K (2000). Characterization of CD4(-) CD8(-) CD3(+) T-cell receptor-alpha beta(+) T cells in murine cytomegalovirus infection. Immunology 101, 19-29.
Huang, CT, Workman, CJ, Flies, D, Pan, X, Marson, AL, Zhou, G, Hipkiss, EL, Ravi, S, Kowalski, J, Levitsky, HI, et al. (2004). Role of LAG-3 in regulatory T cells. Immunity 21, 503-513.
Huang, H, Ma, Y, Dawicki, W, Zhang, X, and Gordon, JR (2013). Comparison of induced versus natural regulatory T cells of the same TCR specificity for induction of tolerance to an environmental antigen. J Immunol 191, 1136-1143.
Inobe, J, Slavin, AJ, Komagata, Y, Chen, Y, Liu, L, and Weiner, HL (1998). IL-4 is a differentiation factor for transforming growth factor-beta secreting Th3 cells and oral administration of IL-4 enhances oral tolerance in experimental allergic encephalomyelitis. Eur J Immunol 28, 2780-2790.
Iwai, Y, Hamanishi, J, Chamoto, K, and Honjo, T (2017). Cancer immunotherapies targeting the PD-1 signaling pathway. J Biomed Sci 24, 26.
Jacobsohn, DA, and Vogelsang, GB (2007). Acute graft-versus-host disease. Orphanet J Rare Dis 2, 35.
111
Jamil, MO, and Mineishi, S (2015). State-of-the-art acute and chronic GVHD treatment. Int J Hematol 101, 452-466.
Jin, JO, Han, X, and Yu, Q (2013). Interleukin-6 induces the generation of IL-10-producing Tr1 cells and suppresses autoimmune tissue inflammation. J Autoimmun 40, 28-44.
Johansson, M, and Lycke, N (2003). A unique population of extrathymically derived alpha beta TCR(+)CD4(-)CD8(-) T cells with regulatory functions dominates the mouse female genital tract. J Immunol 170, 1659-1666.
Joller, N, Lozano, E, Burkett, PR, Patel, B, Xiao, S, Zhu, C, Xia, J, Tan, TG, Sefik, E, Yajnik, V, et al. (2014). Treg cells expressing the coinhibitory molecule TIGIT selectively inhibit proinflammatory Th1 and Th17 cell responses. Immunity 40, 569-581.
Jonuleit, H, Schmitt, E, Stassen, M, Tuettenberg, A, Knop, J, and Enk, AH (2001). Identification and functional characterization of human CD4(+)CD25(+) T cells with regulatory properties isolated from peripheral blood. J Exp Med 193, 1285-1294.
Juvet, SC, Han, M, Vanama, R, Joe, B, Kim, EY, Zhao, FL, Jeon, C, Adeyi, O, and Zhang, L (2012). Autocrine IFN-gamma controls the regulatory function of lymphoproliferative double negative T cells. PLoS One 7, e47732.
Juvet, SC, and Zhang, L (2012). Double negative regulatory T cells in transplantation and autoimmunity: recent progress and future directions. J Mol Cell Biol 4, 48-58.
Kadena, T, Matsuzaki, G, Fujise, S, Kishihara, K, Takimoto, H, Sasaki, M, Beppu, M, Nakamura, S, and Nomoto, K (1997). TCR alpha beta(+) CD4(-) CD8(-) T cells differentiate extrathymically in an lck-independent manner and participate in early response against Listeria monocytogenes infection through interferon-gamma production. Immunology 91, 511-519.
Kapp, JA (2008). Special regulatory T-cell review: Suppressors regulated but unsuppressed. Immunology 123, 28-32.
Katzman, SD, Hoyer, KK, Dooms, H, Gratz, IK, Rosenblum, MD, Paw, JS, Isakson, SH, and Abbas, AK (2011). Opposing functions of IL-2 and IL-7 in the regulation of immune responses. Cytokine 56, 116-121.
Kawai, T, and Akira, S (2010). The role of pattern-recognition receptors in innate immunity: update on Toll-like receptors. Nat Immunol 11, 373-384.
112
Kennedy-Nasser, AA, Ku, S, Castillo-Caro, P, Hazrat, Y, Wu, MF, Liu, H, Melenhorst, J, Barrett, AJ, Ito, S, Foster, A, et al. (2014). Ultra low-dose IL-2 for GVHD prophylaxis after allogeneic hematopoietic stem cell transplantation mediates expansion of regulatory T cells without diminishing antiviral and antileukemic activity. Clin Cancer Res 20, 2215-2225.
Khattri, R, Cox, T, Yasayko, SA, and Ramsdell, F (2003). An essential role for Scurfin in CD4+CD25+ T regulatory cells. Nat Immunol 4, 337-342.
Kim, GY, Ligons, DL, Hong, C, Luckey, MA, Keller, HR, Tai, X, Lucas, PJ, Gress, RE, and Park, JH (2012). An in vivo IL-7 requirement for peripheral Foxp3(+) regulatory T cell homeostasis. J Immunol 188, 5859-5866.
Kinjyo, I, Gordon, SM, Intlekofer, AM, Dowdell, K, Mooney, EC, Caricchio, R, Grupp, SA, Teachey, DT, Rao, VK, Lindsten, T, et al. (2010). Cutting edge: Lymphoproliferation caused by Fas deficiency is dependent on the transcription factor eomesodermin. J Immunol 185, 7151-7155.
Kishimoto, TK, Jutila, MA, and Butcher, EC (1990). Identification of a human peripheral lymph node homing receptor: a rapidly down-regulated adhesion molecule. Proc Natl Acad Sci U S A 87, 2244-2248.
Klatzmann, D, and Abbas, AK (2015). The promise of low-dose interleukin-2 therapy for autoimmune and inflammatory diseases. Nat Rev Immunol 15, 283-294.
Koch, U, and Radtke, F (2011). Mechanisms of T cell development and transformation. Annu Rev Cell Dev Biol 27, 539-562.
Koreth, J, Kim, HT, Jones, KT, Lange, PB, Reynolds, CG, Chammas, MJ, Dusenbury, K, Whangbo, J, Nikiforow, S, Alyea, EP, 3rd, et al. (2016). Efficacy, durability, and response predictors of low-dose interleukin-2 therapy for chronic graft-versus-host disease. Blood 128, 130-137.
Kowalczyk, A, D'Souza, CA, and Zhang, L (2014). Cell-extrinsic CTLA4-mediated regulation of dendritic cell maturation depends on STAT3. Eur J Immunol 44, 1143-1155.
Kronenberg, M, and Rudensky, A (2005). Regulation of immunity by self-reactive T cells. Nature 435, 598-604.
Kullberg, MC, Hay, V, Cheever, AW, Mamura, M, Sher, A, Letterio, JJ, Shevach, EM, and Piccirillo, CA (2005). TGF-beta1 production by CD4+ CD25+ regulatory T cells is not essential for suppression of intestinal inflammation. Eur J Immunol 35, 2886-2895.
113
Kumagai, Y, and Akira, S (2010). Identification and functions of pattern-recognition receptors. J Allergy Clin Immunol 125, 985-992.
Landolfi, MM, Van Houten, N, Russell, JQ, Scollay, R, Parnes, JR, and Budd, RC (1993). CD2-CD4-CD8- lymph node T lymphocytes in MRL lpr/lpr mice are derived from a CD2+CD4+CD8+ thymic precursor. J Immunol 151, 1086-1096.
Lanzavecchia, A, and Sallusto, F (2000). Dynamics of T lymphocyte responses: intermediates, effectors, and memory cells. Science 290, 92-97.
Lee, BP, Mansfield, E, Hsieh, SC, Hernandez-Boussard, T, Chen, W, Thomson, CW, Ford, MS, Bosinger, SE, Der, S, Zhang, ZX, et al. (2005). Expression profiling of murine double-negative regulatory T cells suggest mechanisms for prolonged cardiac allograft survival. J Immunol 174, 4535-4544.
Lerret, NM, Houlihan, JL, Kheradmand, T, Pothoven, KL, Zhang, ZJ, and Luo, X (2012). Donor-specific CD8+ Foxp3+ T cells protect skin allografts and facilitate induction of conventional CD4+ Foxp3+ regulatory T cells. Am J Transplant 12, 2335-2347.
Levings, MK, Sangregorio, R, and Roncarolo, MG (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.
Li, MO, Wan, YY, Sanjabi, S, Robertson, AK, and Flavell, RA (2006). Transforming growth factor-beta regulation of immune responses. Annu Rev Immunol 24, 99-146.
Li, XL, Menoret, S, Bezie, S, Caron, L, Chabannes, D, Hill, M, Halary, F, Angin, M, Heslan, M, Usal, C, et al. (2010). Mechanism and localization of CD8 regulatory T cells in a heart transplant model of tolerance. J Immunol 185, 823-833.
Liang, B, Workman, C, Lee, J, Chew, C, Dale, BM, 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.
Ligocki, AJ, and Niederkorn, JY (2015). Advances on Non-CD4 + Foxp3+ T Regulatory Cells: CD8+, Type 1, and Double Negative T Regulatory Cells in Organ Transplantation. Transplantation 99, 1553-1559.
Lin, SJ, Lu, CH, Yan, DC, Lee, PT, Hsiao, HS, and Kuo, ML (2014). Expansion of regulatory T cells from umbilical cord blood and adult peripheral blood CD4(+)CD25 (+) T cells. Immunol Res 60, 105-111.
114
Liston, A (2011). Immunological tolerance 50 years after the Burnet Nobel Prize. Immunol Cell Biol 89, 14-15.
Liu, T, Cong, M, Sun, G, Wang, P, Tian, Y, Shi, W, Li, X, You, H, and Zhang, D (2016). Combination of double negative T cells and anti-thymocyte serum reverses type 1 diabetes in NOD mice. J Transl Med 14, 57.
Lu, L, and Cantor, H (2008). Generation and regulation of CD8(+) regulatory T cells. Cell Mol Immunol 5, 401-406.
Lu, Y, Wang, J, Gu, J, Lu, H, Li, X, Qian, X, Liu, X, Wang, X, Zhang, F, and Lu, L (2014). Rapamycin regulates iTreg function through CD39 and Runx1 pathways. J Immunol Res 2014, 989434.
Ma, Y, He, KM, Garcia, B, Min, W, Jevnikar, A, and Zhang, ZX (2008). Adoptive transfer of double negative T regulatory cells induces B-cell death in vivo and alters rejection pattern of rat-to-mouse heart transplantation. Xenotransplantation 15, 56-63.
Magnani, CF, Alberigo, G, Bacchetta, R, Serafini, G, Andreani, M, Roncarolo, MG, and Gregori, S (2011). Killing of myeloid APCs via HLA class I, CD2 and CD226 defines a novel mechanism of suppression by human Tr1 cells. Eur J Immunol 41, 1652-1662.
Mahic, M, Henjum, K, Yaqub, S, Bjornbeth, BA, Torgersen, KM, Tasken, K, and Aandahl, EM (2008). Generation of highly suppressive adaptive CD8(+)CD25(+)FOXP3(+) regulatory T cells by continuous antigen stimulation. Eur J Immunol 38, 640-646.
Mandapathil, M, Hilldorfer, B, Szczepanski, MJ, Czystowska, M, Szajnik, M, Ren, J, Lang, S, Jackson, EK, Gorelik, E, and Whiteside, TL (2010). Generation and accumulation of immunosuppressive adenosine by human CD4+CD25highFOXP3+ regulatory T cells. J Biol Chem 285, 7176-7186.
Marek-Trzonkowska, N, Mysliwiec, M, Dobyszuk, A, Grabowska, M, Derkowska, I, Juscinska, J, Owczuk, R, Szadkowska, A, Witkowski, P, Mlynarski, W, et al. (2014). Therapy of type 1 diabetes with CD4(+)CD25(high)CD127-regulatory T cells prolongs survival of pancreatic islets - results of one year follow-up. Clin Immunol 153, 23-30.
Marson, A, Kretschmer, K, Frampton, GM, Jacobsen, ES, Polansky, JK, MacIsaac, KD, Levine, SS, Fraenkel, E, von Boehmer, H, and Young, RA (2007). Foxp3 occupancy and regulation of key target genes during T-cell stimulation. Nature 445, 931-935.
Martelli, MF, Di Ianni, M, Ruggeri, L, Falzetti, F, Carotti, A, Terenzi, A, Pierini, A, Massei, MS, Amico, L, Urbani, E, et al. (2014). HLA-haploidentical transplantation with
115
regulatory and conventional T-cell adoptive immunotherapy prevents acute leukemia relapse. Blood 124, 638-644.
Martina, MN, Noel, S, Saxena, A, Rabb, H, and Hamad, AR (2015). Double negative (DN) alphabeta T cells: misperception and overdue recognition. Immunol Cell Biol 93, 305-310.
Mazzucchelli, R, and Durum, SK (2007). Interleukin-7 receptor expression: intelligent design. Nat Rev Immunol 7, 144-154.
McGargill, MA, Derbinski, JM, and Hogquist, KA (2000). Receptor editing in developing T cells. Nat Immunol 1, 336-341.
McIver, Z, Serio, B, Dunbar, A, O'Keefe, CL, Powers, J, Wlodarski, M, Jin, T, Sobecks, R, Bolwell, B, and Maciejewski, JP (2008). Double-negative regulatory T cells induce allotolerance when expanded after allogeneic haematopoietic stem cell transplantation. Br J Haematol 141, 170-178.
Medawar, PB (1948). Immunity to homologous grafted skin; the fate of skin homografts transplanted to the brain, to subcutaneous tissue, and to the anterior chamber of the eye. Br J Exp Pathol 29, 58-69.
Merims, S, Li, X, Joe, B, Dokouhaki, P, Han, M, Childs, RW, Wang, ZY, Gupta, V, Minden, MD, and Zhang, L (2011). Anti-leukemia effect of ex vivo expanded DNT cells from AML patients: a potential novel autologous T-cell adoptive immunotherapy. Leukemia 25, 1415-1422.
Minagawa, R, Okano, S, Tomita, Y, Kishihara, K, Yamada, H, Nomoto, K, Shimada, M, Maehara, Y, Sugimachi, K, Yoshikai, Y, et al. (2004). The critical role of Fas-Fas ligand interaction in donor-specific transfusion-induced tolerance to H-Y antigen. Transplantation 78, 799-806.
Mogensen, TH (2009). Pathogen recognition and inflammatory signaling in innate immune defenses. Clin Microbiol Rev 22, 240-273, Table of Contents.
Moller, G (1988). Do suppressor T cells exist? Scand J Immunol 27, 247-250.
Mond, JJ, and Brunswick, M (2003). Proliferative assays for B cell function. Curr Protoc Immunol Chapter 3, Unit 3 10.
Monti, P, Scirpoli, M, Maffi, P, Ghidoli, N, De Taddeo, F, Bertuzzi, F, Piemonti, L, Falcone, M, Secchi, A, and Bonifacio, E (2008). Islet transplantation in patients with
116
autoimmune diabetes induces homeostatic cytokines that expand autoreactive memory T cells. J Clin Invest 118, 1806-1814.
Mueller, DL (2010). Mechanisms maintaining peripheral tolerance. Nat Immunol 11, 21-27.
Murugaiyan, G, Mittal, A, Lopez-Diego, R, Maier, LM, Anderson, DE, and Weiner, HL (2009). IL-27 is a key regulator of IL-10 and IL-17 production by human CD4+ T cells. J Immunol 183, 2435-2443.
Nakamura, K, Kitani, A, and Strober, W (2001). Cell contact-dependent immunosuppression by CD4(+)CD25(+) regulatory T cells is mediated by cell surface-bound transforming growth factor beta. J Exp Med 194, 629-644.
Nelson, BH (2004). IL-2, regulatory T cells, and tolerance. J Immunol 172, 3983-3988.
Netea, MG, Latz, E, Mills, KH, and O'Neill, LA (2015). Innate immune memory: a paradigm shift in understanding host defense. Nat Immunol 16, 675-679.
Netea, MG, Quintin, J, and van der Meer, JW (2011). Trained immunity: a memory for innate host defence. Cell Host Microbe 9, 355-361.
Ng, WF, Duggan, PJ, Ponchel, F, Matarese, G, Lombardi, G, Edwards, AD, Isaacs, JD, and Lechler, RI (2001). Human CD4(+)CD25(+) cells: a naturally occurring population of regulatory T cells. Blood 98, 2736-2744.
Niederkorn, JY (2006). See no evil, hear no evil, do no evil: the lessons of immune privilege. Nat Immunol 7, 354-359.
Niehues, T, Gulwani-Akolkar, B, Akolkar, PN, Tax, W, and Silver, J (1994). Unique phenotype and distinct TCR V beta repertoire in human peripheral blood alpha beta TCR+, CD4-, and CD8- double negative T cells. J Immunol 152, 1072-1081.
Ohta, A, Kini, R, Ohta, A, Subramanian, M, Madasu, M, and Sitkovsky, M (2012). The development and immunosuppressive functions of CD4(+) CD25(+) FoxP3(+) regulatory T cells are under influence of the adenosine-A2A adenosine receptor pathway. Front Immunol 3, 190.
Ohta, A, and Sitkovsky, M (2001). Role of G-protein-coupled adenosine receptors in downregulation of inflammation and protection from tissue damage. Nature 414, 916-920.
117
Ono, M, Yaguchi, H, Ohkura, N, Kitabayashi, I, Nagamura, Y, Nomura, T, Miyachi, Y, Tsukada, T, and Sakaguchi, S (2007). Foxp3 controls regulatory T-cell function by interacting with AML1/Runx1. Nature 446, 685-689.
Pandiyan, P, Zheng, L, Ishihara, S, Reed, J, and Lenardo, MJ (2007). CD4+CD25+Foxp3+ regulatory T cells induce cytokine deprivation-mediated apoptosis of effector CD4+ T cells. Nat Immunol 8, 1353-1362.
Paust, S, Lu, L, McCarty, N, and Cantor, H (2004). Engagement of B7 on effector T cells by regulatory T cells prevents autoimmune disease. Proc Natl Acad Sci U S A 101, 10398-10403.
Petrie, HT, Pearse, M, Scollay, R, and Shortman, K (1990). Development of immature thymocytes: initiation of CD3, CD4, and CD8 acquisition parallels down-regulation of the interleukin 2 receptor alpha chain. Eur J Immunol 20, 2813-2815.
Putnam, AL, Safinia, N, Medvec, A, Laszkowska, M, Wray, M, Mintz, MA, Trotta, E, Szot, GL, Liu, W, Lares, A, et al. (2013). Clinical grade manufacturing of human alloantigen-reactive regulatory T cells for use in transplantation. Am J Transplant 13, 3010-3020.
Read, S, Malmstrom, V, and Powrie, F (2000). Cytotoxic T lymphocyte-associated antigen 4 plays an essential role in the function of CD25(+)CD4(+) regulatory cells that control intestinal inflammation. J Exp Med 192, 295-302.
Reimann, J (1991). Double-negative (CD4-CD8-), TCR alpha beta-expressing, peripheral T cells. Scand J Immunol 34, 679-688.
Rifa'i, M, Kawamoto, Y, Nakashima, I, and Suzuki, H (2004). Essential roles of CD8+CD122+ regulatory T cells in the maintenance of T cell homeostasis. J Exp Med 200, 1123-1134.
Rochman, Y, Spolski, R, and Leonard, WJ (2009). New insights into the regulation of T cells by gamma(c) family cytokines. Nat Rev Immunol 9, 480-490.
Roncarolo, MG, Gregori, S, Bacchetta, R, and Battaglia, M (2014). Tr1 cells and the counter-regulation of immunity: natural mechanisms and therapeutic applications. Curr Top Microbiol Immunol 380, 39-68.
Sadlack, B, Merz, H, Schorle, H, Schimpl, A, Feller, AC, and Horak, I (1993). Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene. Cell 75, 253-261.
118
Sagoo, P, Ali, N, Garg, G, Nestle, FO, Lechler, RI, and Lombardi, G (2011). Human regulatory T cells with alloantigen specificity are more potent inhibitors of alloimmune skin graft damage than polyclonal regulatory T cells. Sci Transl Med 3, 83ra42.
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.
Sakaguchi, S, Wing, K, and Miyara, M (2007). Regulatory T cells - a brief history and perspective. Eur J Immunol 37 Suppl 1, S116-123.
Sakaguchi, S, Yamaguchi, T, Nomura, T, and Ono, M (2008). Regulatory T cells and immune tolerance. Cell 133, 775-787.
Sallusto, F, Geginat, J, and Lanzavecchia, A (2004). Central memory and effector memory T cell subsets: function, generation, and maintenance. Annu Rev Immunol 22, 745-763.
Salomon, B, Lenschow, DJ, Rhee, L, Ashourian, N, Singh, B, Sharpe, A, and Bluestone, JA (2000). B7/CD28 costimulation is essential for the homeostasis of the CD4+CD25+ immunoregulatory T cells that control autoimmune diabetes. Immunity 12, 431-440.
Schliesser, U, Streitz, M, and Sawitzki, B (2012). Tregs: application for solid-organ transplantation. Curr Opin Organ Transplant 17, 34-41.
Schmitt, EG, and Williams, CB (2013). Generation and function of induced regulatory T cells. Front Immunol 4, 152.
Schorle, H, Holtschke, T, Hunig, T, Schimpl, A, and Horak, I (1991). Development and function of T cells in mice rendered interleukin-2 deficient by gene targeting. Nature 352, 621-624.
Schwartz, RH (1990). A cell culture model for T lymphocyte clonal anergy. Science 248, 1349-1356.
Seder, RA, Marth, T, Sieve, MC, Strober, W, Letterio, JJ, Roberts, AB, and Kelsall, B (1998). Factors involved in the differentiation of TGF-beta-producing cells from naive CD4+ T cells: IL-4 and IFN-gamma have opposing effects, while TGF-beta positively regulates its own production. J Immunol 160, 5719-5728.
Serafini, G, Andreani, M, Testi, M, Battarra, M, Bontadini, A, Biral, E, Fleischhauer, K, Marktel, S, Lucarelli, G, Roncarolo, MG, et al. (2009). Type 1 regulatory T cells are
119
associated with persistent split erythroid/lymphoid chimerism after allogeneic hematopoietic stem cell transplantation for thalassemia. Haematologica 94, 1415-1426.
Shalev, I, Liu, H, Koscik, C, Bartczak, A, Javadi, M, Wong, KM, Maknojia, A, He, W, Liu, MF, Diao, J, et al. (2008). Targeted deletion of fgl2 leads to impaired regulatory T cell activity and development of autoimmune glomerulonephritis. J Immunol 180, 249-260.
Shan, J, Feng, L, Li, Y, Sun, G, Chen, X, and Chen, P (2014). The effects of rapamycin on regulatory T cells: its potential time-dependent role in inducing transplant tolerance. Immunol Lett 162, 74-86.
Shevach, EM (2006). From vanilla to 28 flavors: multiple varieties of T regulatory cells. Immunity 25, 195-201.
Shevach, EM (2009). Mechanisms of foxp3+ T regulatory cell-mediated suppression. Immunity 30, 636-645.
Shevach, EM, DiPaolo, RA, Andersson, J, Zhao, DM, Stephens, GL, and Thornton, AM (2006). The lifestyle of naturally occurring CD4+ CD25+ Foxp3+ regulatory T cells. Immunol Rev 212, 60-73.
Shirai, T, Abe, M, Yagita, H, Okumura, K, Morse, HC, 3rd, and Davidson, WF (1990). The expanded populations of CD4-CD8- T cell receptor alpha/beta+ T cells associated with the lpr and gld mutations are CD2. J Immunol 144, 3756-3761.
Simonetta, F, Gestermann, N, Bloquet, S, and Bourgeois, C (2014). Interleukin-7 Optimizes FOXP3+CD4+ Regulatory T Cells Reactivity to Interleukin-2 by Modulating CD25 Expression. PLoS One 9, e113314.
Sindhi, R, Manavalan, JS, Magill, A, Suciu-Foca, N, and Zeevi, A (2005). Reduced immunosuppression in pediatric liver-intestine transplant recipients with CD8+CD28- T-suppressor cells. Hum Immunol 66, 252-257.
Singer, BD, King, LS, and D'Alessio, FR (2014). Regulatory T Cells as Immunotherapy. Front Immunol 5, 46.
Singh, K, Kozyr, N, Stempora, L, Kirk, AD, Larsen, CP, Blazar, BR, and Kean, LS (2012). Regulatory T cells exhibit decreased proliferation but enhanced suppression after pulsing with sirolimus. Am J Transplant 12, 1441-1457.
Smith, KA (1988). Interleukin-2: inception, impact, and implications. Science 240, 1169-1176.
120
Sojka, DK, and Fowell, DJ (2011). Regulatory T cells inhibit acute IFN-gamma synthesis without blocking T-helper cell type 1 (Th1) differentiation via a compartmentalized requirement for IL-10. Proc Natl Acad Sci U S A 108, 18336-18341.
Starr, TK, Jameson, SC, and Hogquist, KA (2003). Positive and negative selection of T cells. Annu Rev Immunol 21, 139-176.
Strober, S, Dejbachsh-Jones, S, Van Vlasselaer, P, Duwe, G, Salimi, S, and Allison, JP (1989). Cloned natural suppressor cell lines express the CD3+CD4-CD8- surface phenotype and the alpha, beta heterodimer of the T cell antigen receptor. J Immunol 143, 1118-1122.
Stumhofer, JS, Silver, JS, Laurence, A, Porrett, PM, Harris, TH, Turka, LA, Ernst, M, Saris, CJ, O'Shea, JJ, and Hunter, CA (2007). Interleukins 27 and 6 induce STAT3-mediated T cell production of interleukin 10. Nat Immunol 8, 1363-1371.
Surh, CD, and Sprent, J (2008). Homeostasis of naive and memory T cells. Immunity 29, 848-862.
Suzuki, H, Kundig, TM, Furlonger, C, Wakeham, A, Timms, E, Matsuyama, T, Schmits, R, Simard, JJ, Ohashi, PS, Griesser, H, et al. (1995). Deregulated T cell activation and autoimmunity in mice lacking interleukin-2 receptor beta. Science 268, 1472-1476.
Takahashi, T, Tagami, T, Yamazaki, S, Uede, T, Shimizu, J, Sakaguchi, N, Mak, TW, 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.
Tang, Q, Bluestone, JA, and Kang, SM (2012). CD4(+)Foxp3(+) regulatory T cell therapy in transplantation. J Mol Cell Biol 4, 11-21.
Thomson, CW, Mossoba, ME, Siatskas, C, Chen, W, Sung, A, Medin, JA, and Zhang, L (2007). Lentivirally transduced recipient-derived dendritic cells serve to ex vivo expand functional FcRgamma-sufficient double-negative regulatory T cells. MolTher 15, 818-824.
Thornton, AM, Donovan, EE, Piccirillo, CA, and Shevach, EM (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.
Thornton, AM, and Shevach, EM (1998). CD4(+)CD25(+) immunoregulatory T cells suppress polyclonal T cell activation in vitro by inhibiting interleukin 2 production. J Exp Med 188, 287-296.
121
Tran, GT, Hodgkinson, SJ, Carter, NM, Verma, ND, Plain, KM, Boyd, R, Robinson, CM, Nomura, M, Killingsworth, M, and Hall, BM (2012). IL-5 promotes induction of antigen-specific CD4+CD25+ T regulatory cells that suppress autoimmunity. Blood 119, 4441-4450.
Trzonkowski, P, Bacchetta, R, Battaglia, M, Berglund, D, Bohnenkamp, HR, ten Brinke, A, Bushell, A, Cools, N, Geissler, EK, Gregori, S, et al. (2015). Hurdles in therapy with regulatory T cells. Sci Transl Med 7, 304ps318.
Trzonkowski, P, Bieniaszewska, M, Juscinska, J, Dobyszuk, A, Krzystyniak, A, Marek, N, Mysliwska, J, and Hellmann, A (2009). First-in-man clinical results of the treatment of patients with graft-versus-host disease with human ex vivo expanded CD4(+)CD25(+)CD127(-) T regulatory cells. Clin Immunol 133, 22-26.
Veerapathran, A, Pidala, J, Beato, F, Betts, B, Kim, J, Turner, JG, Hellerstein, MK, Yu, XZ, Janssen, W, and Anasetti, C (2013). Human regulatory T cells against minor histocompatibility antigens: ex vivo expansion for prevention of graft-versus-host disease. Blood 122, 2251-2261.
Venken, K, Thewissen, M, Hellings, N, Somers, V, Hensen, K, Rummens, JL, and Stinissen, P (2007). A CFSE based assay for measuring CD4+CD25+ regulatory T cell mediated suppression of auto-antigen specific and polyclonal T cell responses. J Immunol Methods 322, 1-11.
Voelkl, S, Gary, R, and Mackensen, A (2011). Characterization of the immunoregulatory function of human TCR-alphabeta+ CD4- CD8- double-negative T cells. Eur J Immunol 41, 739-748.
Vuddamalay, Y, Attia, M, Vicente, R, Pomie, C, Enault, G, Leobon, B, Joffre, O, Romagnoli, P, and van Meerwijk, JP (2016). Mouse and human CD8 CD28 regulatory T lymphocytes differentiate in the thymus. Immunology.
Wang, F, Huang, CY, and Kanagawa, O (1998). Rapid deletion of rearranged T cell antigen receptor (TCR) Valpha-Jalpha segment by secondary rearrangement in the thymus: role of continuous rearrangement of TCR alpha chain gene and positive selection in the T cell repertoire formation. Proc Natl Acad Sci U S A 95, 11834-11839.
Wei, S, Kryczek, I, Zou, L, Daniel, B, Cheng, P, Mottram, P, Curiel, T, Lange, A, and Zou, W (2005). Plasmacytoid dendritic cells induce CD8+ regulatory T cells in human ovarian carcinoma. Cancer Res 65, 5020-5026.
Weiner, HL (2001a). Induction and mechanism of action of transforming growth factor-beta-secreting Th3 regulatory cells. Immunol Rev 182, 207-214.
122
Weiner, HL (2001b). The mucosal milieu creates tolerogenic dendritic cells and T(R)1 and T(H)3 regulatory cells. Nat Immunol 2, 671-672.
Wells, AD (2009). New insights into the molecular basis of T cell anergy: anergy factors, avoidance sensors, and epigenetic imprinting. J Immunol 182, 7331-7341.
Wen, Z, Shimojima, Y, Shirai, T, Li, Y, Ju, J, Yang, Z, Tian, L, Goronzy, JJ, and Weyand, CM (2016). NADPH oxidase deficiency underlies dysfunction of aged CD8+ Tregs. J Clin Invest.
Willerford, DM, Chen, J, Ferry, JA, Davidson, L, Ma, A, and Alt, FW (1995). Interleukin-2 receptor alpha chain regulates the size and content of the peripheral lymphoid compartment. Immunity 3, 521-530.
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.
Wu, S, Jin, L, Vence, L, and Radvanyi, LG (2010). Development and application of 'phosphoflow' as a tool for immunomonitoring. Expert Rev Vaccines 9, 631-643.
Wu, Y, Borde, M, Heissmeyer, V, Feuerer, M, Lapan, AD, Stroud, JC, Bates, DL, Guo, L, Han, A, Ziegler, SF, et al. (2006). FOXP3 controls regulatory T cell function through cooperation with NFAT. Cell 126, 375-387.
Xie, Z, Chang, C, and Zhou, Z (2014). Molecular mechanisms in autoimmune type 1 diabetes: a critical review. Clin Rev Allergy Immunol 47, 174-192.
Ye, H, Chang, Y, Zhao, X, and Huang, X (2011). Characterization of CD3+CD4-CD8- (double negative) T cells reconstitution in patients following hematopoietic stem-cell transplantation. Transpl Immunol 25, 180-186.
Young, KJ, DuTemple, B, Phillips, MJ, and Zhang, L (2003a). Inhibition of graft-versus-host disease by double-negative regulatory T cells. J Immunol 171, 134-141.
Young, KJ, DuTemple, B, Zhang, Z, Levy, G, and Zhang, L (2001). CD4(-)CD8(-) regulatory T cells implicated in preventing graft-versus-host and promoting graft-versus-leukemia responses. Transplant Proc 33, 1762-1763.
Young, KJ, Kay, LS, Phillips, MJ, and Zhang, L (2003b). Antitumor activity mediated by double-negative T cells. Cancer Res 63, 8014-8021.
123
Young, KJ, Yang, L, Phillips, MJ, and Zhang, L (2002). Donor-lymphocyte infusion induces transplantation tolerance by activating systemic and graft-infiltrating double-negative regulatory T cells. Blood 100, 3408-3414.
Yu, X, Harden, K, Gonzalez, LC, Francesco, M, Chiang, E, Irving, B, Tom, I, Ivelja, S, Refino, CJ, Clark, H, et al. (2009). The surface protein TIGIT suppresses T cell activation by promoting the generation of mature immunoregulatory dendritic cells. Nat Immunol 10, 48-57.
Zarek, PE, Huang, CT, Lutz, ER, Kowalski, J, Horton, MR, Linden, J, Drake, CG, and Powell, JD (2008). A2A receptor signaling promotes peripheral tolerance by inducing T-cell anergy and the generation of adaptive regulatory T cells. Blood 111, 251-259.
Zhang, D, Yang, W, Degauque, N, Tian, Y, Mikita, A, and Zheng, XX (2007). New differentiation pathway for double-negative regulatory T cells that regulates the magnitude of immune responses. Blood 109, 4071-4079.
Zhang, L, Bertucci, AM, Ramsey-Goldman, R, Burt, RK, and Datta, SK (2009). Regulatory T cell (Treg) subsets return in patients with refractory lupus following stem cell transplantation, and TGF-beta-producing CD8+ Treg cells are associated with immunological remission of lupus. J Immunol 183, 6346-6358.
Zhang, Q, Chikina, M, Szymczak-Workman, AL, Horne, W, Kolls, JK, Vignali, KM, Normolle, D, Bettini, M, Workman, CJ, and Vignali, DAA (2017). LAG3 limits regulatory T cell proliferation and function in autoimmune diabetes. Sci Immunol 2.
Zhang, ZX, Ma, Y, Wang, H, Arp, J, Jiang, J, Huang, X, He, KM, Garcia, B, Madrenas, J, and Zhong, R (2006). Double-negative T cells, activated by xenoantigen, lyse autologous B and T cells using a perforin/granzyme-dependent, Fas-Fas ligand-independent pathway. J Immunol 177, 6920-6929.
Zhang, ZX, Yang, L, Young, KJ, DuTemple, B, and Zhang, L (2000). Identification of a previously unknown antigen-specific regulatory T cell and its mechanism of suppression. Nat Med 6, 782-789.
Zhang, ZX, Young, K, and Zhang, L (2001). CD3+CD4-CD8- alphabeta-TCR+ T cell as immune regulatory cell. J Mol Med (Berl) 79, 419-427.
Zhao, DM, Thornton, AM, DiPaolo, RJ, and Shevach, EM (2006). Activated CD4+CD25+ T cells selectively kill B lymphocytes. Blood 107, 3925-3932.
124
Zheng, J, Liu, Y, Lau, YL, and Tu, W (2010). CD40-activated B cells are more potent than immature dendritic cells to induce and expand CD4(+) regulatory T cells. Cell MolImmunol 7, 44-50.
Zheng, Y, Josefowicz, SZ, Kas, A, Chu, TT, Gavin, MA, and Rudensky, AY (2007). Genome-wide analysis of Foxp3 target genes in developing and mature regulatory T cells. Nature 445, 936-940.