UvA-DARE (Digital Academic Repository)

Modulation of immune tolerance: the role of tolerogenic dendritic cells and TNFα

Boks, M.A.

Publication date
2012
Document Version
Final published version

Link to publication

Citation for published version (APA):

Boks, M. A. (2012). Modulation of immune tolerance: the role of tolerogenic dendritic cells and
TNFα. [Thesis, externally prepared, Universiteit van Amsterdam].

General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s)
and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open
content license (like Creative Commons).

Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please
let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material
inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter
to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You
will be contacted as soon as possible.

UvA-DARE is a service provided by the library of the University of Amsterdam (https://dare.uva.nl)

Download date:13 1月 2025
 MODULATION OF IMMUNE TOLERANCE
The role of tolerogenic dendritic cells and TNFα

Martine A. Boks

MODULATION OF IMMUNE TOLERANCE • The role of tolerogenic dendritic cells and TNFα Martine A. Boks 2012
Modulation of immune tolerance

The role of tolerogenic dendritic cells and TNFα

Martine A. Boks
The work described in this thesis was performed at
the department of Immunopathology, Sanquin Blood
Supply, Amsterdam, The Netherlands.
Printing of this thesis was financially supported by
Sanquin Blood Supply Division Research, Sanquin Blood
Supply Division Reagents, CellGenix, Dutch Arthritis
Foundation and Pfizer.
ISBN: 978-94-6182-153-9
Copyright © by Martine A. Boks, 2012
Cover design, layout and printing by OffPage,
Amsterdam, The Netherlands
 Modulation of immune tolerance

The role of tolerogenic dendritic cells and TNFα

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor

aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. D.C. van den Boom
ten overstaan van een door het college voor promoties
ingestelde commissie,
in het openbaar te verdedigen in de Aula der Universiteit
op vrijdag 26 oktober 2012, te 11:00 uur

door

Martine Annemarie Boks

Geboren te Voorburg
Promotiecommissie

Promotor: Prof. dr. S.M. van Ham

Co-promotores: Dr. J.A. ten Brinke
Dr. J.J. Zwaginga

Overige leden: Prof. dr. C. van Kooten

Prof. dr. Y. van Kooyk
Prof. dr. M.L. Kapsenberg
Prof. dr. L.A. Aarden
Prof. dr. W.J. Stiekema
Dr. C.M.U. Hilkens

Faculteit der Natuurwetenschappen, Wiskunde en Informatica

Contents

Chapter 1 General introduction 7

Chapter 2 An optimized CFSE-based T-cell suppression assay to evaluate 35

the suppressive capacity of regulatory T-cells induced by human
tolerogenic dendritic cells
Scand. J. Immunol. 2010 Aug;72(2):158-68

Chapter 3 IL-10-generated tolerogenic dendritic cells are optimal for 55

functional regulatory T cell induction – A comparative study of
human clinical-applicable DC
Clin. Immunol. 2012 Mar;142(3):332-42

Chapter 4 Inhibition of TNFRII signalling by anti-TNFα primes naïve CD4+ 77

T cells towards IL-10+ cells with strong regulatory phenotype
and function
Manuscript in preparation

Chapter 5 CD4+ T cells co-expressing IL-10 and IFNγ display a regulatory 97

gene profile and dampen immune responses

Chapter 6 Summarizing discussion 119

Addendum Nederlandse samenvatting 141

Curriculum Vitae 145
Dankwoord 147
 1
CHAPTER
General introduction
 GENERAL INTRODUCTION

Immune activation
The immune system consists of a complex network of proteins and cells that

CHAPTER 1
cooperate to protect us from disease. Invading pathogens, tumour cells and other
foreign antigens are efficiently recognized and eradicated. The immune system
comprises of innate and adaptive immunity. Innate immunity depends on invariant
receptors that recognize common features of pathogens and provides a first line of
defence against a broad spectrum of pathogens. Adaptive immunity recognizes any
foreign antigen with antigen-specific receptors, expressed by B and T lymphocytes,
and provides specific immunity. In the same time, the immune system does not
respond in absence of danger, providing immunological tolerance against innocuous
antigens.

Dendritic cells
Host defence relies on the recognition of pathogens and/ or tissue injury, and
the subsequent initiation of an appropriate immune response. Dendritic cells
(DC) belong to the ‘family’ of professional antigen-presenting cells (APC), which
also include B cells and macrophages. APC have the unique ability to recognize
pathogens, transfer this information and activate lymphocytes, the effector cells of
the acquired immune system.1 DC are generated in the bone marrow and migrate
as precursor cells to potential entry sites of pathogens. Myeloid DC are positioned
beneath our body surfaces, e.g. immediately below mucosal surfaces, like the
respiratory and gastrointestinal tract, the skin, and also in the blood and lymphoid
tissues. Plasmacytoid DC are located primarily in the blood and lymphoid organs.2
DC form an essential part of the innate immune system in that they act as sentinels
by scavenging their surroundings to search for pathogens or other ‘non-self’
molecules. Their broad tissue distribution optimises the chance of encountering
pathogens.
DC express many pattern recognition receptors (PRR), like C-type lectin
receptors (CLR), Toll-like receptors (TLR) and Nod-like receptors (NLR). These
receptors recognize a wide variety of PAMPs (pathogen-associated molecular
patterns) of viruses, bacteria, fungi and parasites.3-5 These PRR induce downstream
signalling after ligation to a PAMP, and some also aid in the uptake of pathogens,
thereby enhancing the delivery of antigens to the endocytic system. Here, antigens
are processed into small fragments, i.e. peptides, and loaded onto specialized
presenting molecules; major histocompatibility complex (MHC) molecules.6 Peptide-
loaded MHC molecules are presented to and recognized by T lymphocytes, which is
necessary to induce a specific immune response.
After PRR ligation, DC undergo a maturation process, thereby developing
additional functions that enhance the ability of DC to induce an immune response.
Maturation of DC regulates antigen-processing by lowering the pH of endocytic
vacuoles, activating proteolysis and transporting peptide-MHC complexes to
the cell surface.6 In addition, maturing DC upregulate CC chemokine receptor 7
(CCR7), which drives their migration to secondary lymphoid organs such as spleen

9
 CHAPTER 1

and draining lymph nodes, where naïve T cells reside.7 Maturing DC also remodel
their surface, typically expressing many lymphocyte binding and co-stimulatory
CHAPTER 1

molecules that enable T cell activation and differentiation.1 All these properties
ensure the encounter with lymphocytes in secondary lymphoid organs and the
induction of T cell activation in answer to danger. Thus, scavenging immature DC
in the periphery are specialized in capturing and processing antigens, due to high
expression of PRR and high endocytosis rates. After antigen recognition and uptake,
DC mature and gene expression is altered, resulting in a functional change from
phagocytic cells into presenting cells.

T cell activation
DC maturation leads to expression of peptide-loaded MHC molecules and co-
stimulatory molecules, and the synthesis of specific cytokines for an appropriate
lymphocyte response. DC are primed by pathogens and provide pathogen-specific
signals to T cells which then become protective effector T cells.8 The fate of naïve
T cells is determined by three signals that are provided by mature DC.9 Signal 1
results from recognition of peptide-loaded MHC on the DC cell surface by the
T cell receptor (TCR). This determines the antigen-specificity of the response. The
peptide-presenting MHC molecules are of two types, MHC class I and MHC class II,
which stimulate cytotoxic CD8+ T cells (CTL) and helper CD4+ T cells, respectively.10
The MHC class I antigen presentation pathway is active in almost all cell types,
providing a mechanism for displaying a sample of peptides derived from proteins
that are being synthesized in the cell at any given time. Intracellular antigens,
processed in the cytosol, bind to MHC class I molecules and are recognized by
CD8+ T cells, which, once activated, can directly kill a target cell that presents
these antigens, for example virus-infected cells or tumour cells. To activate naïve
CD8+ T cells in the lymph node, DC take up cellular fragments and digest it in
the endolysosomal pathway. Via a process termed cross-presentation, antigens are
transferred from the endolysosomes into the cytosol, ensuring their presentation
by MHC class I molecules in DC.11 The MHC class II pathway is constitutively active
only in professional APC. Extracellular antigens are taken up and processed in the
endocytic pathway. Subsequent fusion to MHC class II containing vesicles allows
antigen binding to MHC class II molecules and are transported to the plasma
membrane, where they are recognized by CD4+ T cells, which, when activated, have
profound immunoregulatory effects.
A co-stimulatory signal 2 is necessary for a protective response with good T cell
activation and proliferation. Co-stimulation is evoked via ligation of co-stimulatory
molecules expressed on the DC surface to their receptors on the T cell surface,
i.e. CD80/ CD86 to CD28. CD28 is constitutively expressed on the T cell surface,
and ligation provides signals for T cell activation and survival.12 Absence of this
co-stimulation, i.e. in absence of danger and therefore absence of DC maturation,
prevents that autoreactive T cells are activated and instead become anergic. Lastly,
the type of immune response is dictated by the T cell polarizing signal 3, often
cytokines or polarizing co-stimulatory molecules.8 The types of cytokines produced

10
 GENERAL INTRODUCTION

are influenced by the DC subset and the mode of DC activation. Different DC subsets
express different types of TLR and CLR, resulting in a differential responsiveness of

CHAPTER 1
DC to pathogens and the selective production of cytokines.13 Depending on the
type of pathogen and cytokine milieu encountered DC drive the development of
different pro-inflammatory T helper (Th) cells to ensure a pathogen-class specific
immune response, while minimising tissue damage.8 All three signals are delivered
by professional APC, such as DC, after they have maturated and migrated to the
T cell area in secondary lymphoid organs where they induce an immune response.

CD4+ T helper cell polarization

DC mature in distinct ways in response to different pathogens, their products
or other environmental stimuli, e.g. cytokines, chemokines and inflammatory
mediators produced by neighbouring tissue cells in response to pathogens.8
Differential maturation of DC leads to the polarization of different Th subsets with
unique functions, resulting in a specific immune response against different types
of pathogens. Intracellular pathogens prime DC for IL-12 production. DC-derived
IL-12 primes naïve CD4+ T cells into IFNγ-producing Th1 cells.14 In addition, IL-18,
IFNα, IFNβ, and IFNγ itself (from activated Th1 cells or natural killer (NK) cells) can
induce IFNγ and lymphotoxin α (LTα) production in T cells.15,16 IFNγ is crucial in host
defence against intracellular bacteria and viruses, by promoting CD8+ cytotoxic T cell
responses. In addition, Th1 cells activate the antimicrobial activities of macrophages
to assist clearance of ingested microbes and cellular debris and to help achieve
resolution of infection (Figure 1).
Parasitic helminths prime DC to express Th2 polarizing factors. In addition,
tissue factors like TSLP, PGE2 and histamine induce Th2-polarizing DC. Although
much remains to be learned about these Th2 polarizing factors, one important
prerequisite for Th2 induction is the downregulation of IL-12 production.15,16 IL-4 is
a potent inducer of Th2 responses. The source of IL-4 however remains a matter of
debate. The Th2 population itself produces IL-4, as well as basophils, however it is
not clear whether DC produce IL-4. In addition, OX40L, IL-7, IL-25, IL-33 and TSLP
have been implicated in Th2 induction by DC. Th2 cells produce IL-4, IL-5, IL-9 and
IL-13 by which they recruit eosinophils, basophils and mast cells, and help B cells in
IgE antibody production.17 IgE binds to Fcε receptors on mast cells. Antigen binding
to this IgE triggers mast cells to degranulate, thereby initiating a local inflammatory
response and recruiting basophils and eosinophils which contribute in the defence
against parasites. In addition to parasitic helminth infections, Th2 responses are
associated with asthma and allergies.
For long, effector CD4+ Th cells were divided into Th1 and Th2 cells, depending
on the cytokines they produce. Lately, more CD4+ Th cell subsets have been
identified, such as Th17 cells.18 Th17 cells produce IL-17, IL-17F and IL-22, thereby
enhancing neutrophil responses, and clearing fungi and possibly also extracellular
bacteria. Furthermore, Th17 cells also have a role in tissue inflammation and tissue
regeneration.19 Th17 cells are induced by TGFβ, IL-6 and IL-1β, and IL-23 stabilizes
the Th17 effector phenotype.

11
 CHAPTER 1


 
CHAPTER 1



IL-12 IL-4 IL-1β IL-10 IL-6 IL-4 IL-6

IL-18 IL-7 IL-6 TGFβ IL-12 TGFβ TNFα
IFNα IL-25 IL-23 IFNα
IFNβ IL-33 TGFβ IFNβ
IFNγ TSLP

 γ
  γ  
 

Th1 cell Th2 cell Th17 cell induced Treg Tfh cell Th9 cell Th22 cell

IL-4, IL-5, IL17, IL-17F,

IFNγ, LTα IL-10, TGFβ IL-4, IL-21 IL-9 IL-22, TNFα
IL-9, IL-13 IL-22

Intracellular Parasites Bacteria/ fungi Immune B cell help Parasites Barrier function
pathogens (allergy) (autoimmunity) regulation GC formation (allergy) (skin inflammation)

Figure 1. Naïve CD4+ T helper cell polarization. Cytokines play a critical role in the
differentiation of Th subsets and in the function of Th cells. Upon TCR activation and co-
stimulation, both signals provided by APC, naïve CD4+ T cells differentiate into distinct
Th subsets in the context of combinations of cytokines. APC secrete different cytokines in
response to different pathogens. The differentiation processes of Th cells involves upregulation
of master transcriptional regulators. The master transcription factors regulate the production
of Th subset specific effector cytokines, necessary for an effective and pathogen-specific
immune response. Th; T helper cell, Treg; regulatory T cell, Tfh; follicular T helper cell.

T follicular helper (Tfh) cells have emerged as the predominant subset of CD4+
T cells responsible for regulating humoral immunity. Tfh cells provide help to B cells
by IL-4 and IL-21 production, and thereby induce high affinity antibody production,
isotype switching and differentiation of B cells into plasma cells and memory B
cells.20 Tfh cells are probably initially activated by DC within the T cell zone of
secondary lymphoid tissues. ICOS, OX40 and CD40L interactions, and IL-6, IL-12
and type I IFNs have all been implicated in providing signals for IL-21 production
and Tfh polarization, but much remains to be elucidated. Activated CD4+ Tfh cells
can then localize to B cell follicles where they interact with antigen-specific B cells.
In addition to the above described established Th cell subsets, very recently two
new subsets have been described. Th9 cells are CD4+ Th cells that secrete IL-9 and
are induced by IL-4 and TGFβ.21-23 IL-9 has long been thought to be a Th2 cytokine,
since it promotes allergic inflammation and contributes to parasite immunity, but
Th9 cells have lower expression of other Th2 cytokines. The newest Th subset is

12
 GENERAL INTRODUCTION

called Th22 and is characterized by IL-22 production. IL-22 producing cells were
first considered to be a variant of Th17 cells, but recent reports show that IL-22 can

CHAPTER 1
be produced without production of IL-17.24-26 Th22 cells can be induced by DC in
an IL-6- and TNFα-dependent manner. Th22 cells express skin-homing chemokine
receptors CCR4 and CCR10. IL-22 triggers the production of anti-microbial peptides
by epithelial cells. Th22 cells are therefore thought to play a role in early host
defence, but also in inflammatory skin diseases.27,28
In summary, effector T cells must deal with a wide variety of pathogens and the
differentiation of CD4+ Th cells into various specialized effector subsets promotes
the generation of a pathogen-class specific immune response, while minimising
tissue damage.

Plasticity of CD4+ T helper subsets

Although different Th subsets have been described, it is unclear whether these
subsets are truly unique Th subsets. The original definition of Th subsets was
based on the production of hallmark cytokines. To date, it seems that most Th
subsets can display plasticity when stimulated under different polarizing conditions.
For instance, IL-17-secreting Th cells were initially suggested to represent a new
subset because they do not make the other subset-defining cytokines IFNγ and
IL-4. It is now appreciated that Th17 cells often become IFNγ producers under IL-12
conditioning or under chronic inflammatory conditions such as EAE (experimental
autoimmune encephalomyelitis).29,30 In addition, IL-12 is capable of restoring the
IL-12-responsiveness in established effector Th2 cell clones in vitro, resulting in
high levels of IFNγ and downregulated IL-4 production.31 Also in the setting of viral
infection Th2 cells can be reprogrammed to express IFNγ.32 Tfh cells appear to be
the most fluid subset. Tfh cells can become Th1, Th2 and Th17 cells.33 It remains
debatable whether Tfh cells should be designated as a distinct subset, or simply a
state of differentiation that can be imposed upon all Th subsets.
The different Th subsets were originally described to express a unique
transcription factor, i.e. T-bet for Th1 34; Gata3 for Th2 35; Rorγt for Th17 19; and Bcl6
for Tfh 20, regulating subset-specific gene expression, including signature cytokines.
It is now clear that Th subsets can express more than one ‘master regulator’
transcription factor.36 For example, Th17 cells, when generated in the absence of
TGFβ, express Rorγt together with T-bet.37 Furthermore, Bcl6 can be transiently co-
expressed with T-bet in Th1 cells, which is regulated by IL-2.38 On a molecular level,
this switching can be explained. Histone modifications are associated with gene
expression by relaxing or condensing the chromatin structure to activate or repress
transcription. Histone methylation patterns of genes encoding master regulators are
often in a bivalent modus, with both repressive and accessible methylation marks
present.33,39,40 This can explain how other master regulators are readily induced
and how Th subsets can co-express signature cytokines, or even switch to another
subset.
Thus, it is clear that signature cytokines and transcription factors can be co-
expressed in CD4+ Th subsets, or can be changed by Th cells in specific immunologic

13
 CHAPTER 1

context, thereby regulating the fate of effector Th cells. This demonstrates the plastic
nature of CD4+ T helper cells and possibly evokes fine-tuning of T cell immune
CHAPTER 1

responses by the complex cytokine microenvironment in vivo.

Immune tolerance
besides activation of immune responses, the immune system also prevents
excessive immunity by silencing DC and T cells as well as other important players
of the immune system. Control of immune responses is essential in absence
of danger, when immune homeostasis has to be maintained, and also for the
termination of immune responses after pathogen clearance, thereby preventing
excessive immunopathology. In addition, autoimmunity is avoided by silencing
self-reactive T cells that have escaped negative selection in the thymus. DC and
immunosuppressive T cells, named regulatory T cells (Treg), play a crucial role
in maintaining immune homeostasis and preventing immunopathology as well
as autoimmunity.

IL-10
IL-10 serves an important role as anti-inflammatory cytokine in immune tolerance,
being essential for immune homeostasis and reducing immunopathology. The
principal function of IL-10 is to limit and ultimately terminate inflammatory responses.
During infection it inhibits the activity of several immune cells, e.g. T cells, B cells,
macrophages, monocytes and DC, which are required for optimal pathogen
clearance, but also contribute to tissue damage. As a consequence, IL-10 can both
impede pathogen clearance and ameliorate immunopathology.41 The impact of the
suppressive function of IL-10 was highlighted by IL-10-/- mice, which developed
severe colitis in association with commensal bacteria in the gut 42,43 and showed
other exaggerated inflammatory responses to microbial challenge. In contrast,
transgenic mice expressing IL-10 under the control of the MHCII promoter were
highly susceptible to infections, and were not able to mount effective Th1 or Th2
responses.44 IL-10 regulates the growth and/ or differentiation of DC, CD4+ and CD8+
T cells, B cells, NK cells, mast cells, granulocytes, and even non-immune cells like
keratinocytes and endothelial cells.41 IL-10 acts on DC by inhibiting differentiation
and maturation. IL-10 affects T cells by inhibiting proliferation and production of
pro-inflammatory cytokines, and by inducing Treg, as will be described below.

Regulatory T cells
Regulatory T cells are key players of immune regulation. Different types of regulatory
T cells exist. One subset develops during the normal process of T cell maturation in
the thymus, resulting in an endogenous ‘natural’ population of antigen-specific Treg
(nTreg).45 Natural Treg develop as a consequence of high-affinity antigen recognition
in the thymus. It is not known why some thymocytes escape negative selection and
differentiate into Treg cells. Natural Treg are mostly CD4+ and typically express
high levels of CD25, the IL-2 receptor. The importance of Treg in preserving self-

14
 GENERAL INTRODUCTION

tolerance was proven by mouse studies where transfer of CD4+ T cells depleted for
CD25+ Treg resulted in the development of multi-organ autoimmunity in recipient

CHAPTER 1
animals.46 Furthermore, transfer of CD4+CD25+ Treg prevented the development
of experimentally induced autoimmune diseases.47 Co-stimulation through CD28 is
required to maintain a stable pool of nTreg by promoting their self-renewal and by
supporting their survival. IL-2 may be dispensable for the thymic development of
Treg, but is crucial for their survival, homeostasis, and suppressive function in the
periphery. In addition, TGFβ and IL-10 contribute to the expansion of nTreg. Natural
CD4+CD25+ Treg specifically express the transcription factor Foxp3, which controls
their development and function.48,49 Foxp3-/- mice showed hyperactivation of T cells,
failed to give rise to CD4+CD25+ T cells and suffered severe systemic inflammation.
In addition, CTLA-4 and GITR are highly expressed by nTreg and contribute to
self-tolerance, as blocking these molecules induces autoimmune disease similar to
CD25+ T cell depletion.45 CD25+Foxp3+ nTreg suppress the activation, proliferation
and effector functions, such as cytokine production, of a wide range of immune
cells, including CD4+ and CD8+ T cells, NK cells, B cells and APC.50
The second subset of Treg are the ‘adaptive’ or ‘induced’ Treg (iTreg) which
develop in the periphery from naïve T cells under particular immunosuppressive
conditions.51 Induced Treg in human do not have a specific marker, like CD25 for
nTreg, but different iTreg subtypes keep being described, with different markers
expressed and different mechanisms of suppressive function. Induced Treg need
antigen/ TCR stimulation in order to suppress, but might not need CD28 co-
stimulation for their development and function. Suppression by iTreg can be
antigen-specific or non-specific. Furthermore, suppression can be cell-cell contact
dependent, or dependent on soluble mediators, like cytokines.51 Recently, it was
demonstrated that Treg cells co-express Foxp3 and canonical Th cell-associated
transcription factors. The emerging concept arises that these specialized Treg cell
populations control distinct Th cell subsets during a polarized Th cell-driven immune
response by expressing the relevant Th cell-associated transcription factors.52,53 The
mechanisms by which the canonical Th transcription factors control Treg activity
are unclear, but are likely to involve a combination of influences on Treg migration,
function and homeostasis.
The best defined iTregs are CD25+Foxp3+, Th3 and Tr1 cells. CD25+Foxp3+ iTreg
resemble the naturally occurring thymic-derived nTreg.51 Co-stimulatory molecules
that have been suggested to play a role in CD25+ Treg induction include GITR,
CTLA-4, ICOS and CD40L. Expression of these molecules is not unique for Treg
as they are also expressed on activated T cells. It was shown that both human and
mouse CD25- T cells turn into CD25+Foxp3+ suppressor cells in the presence of
TGFβ, IL-2 and IL-10.54 TGFβ induces Foxp3, CTLA-4 and CD25 expression and
suppressive function.51 The exact mechanism of suppression of CD25+Foxp3+ iTreg
is not clear. Cytokines TGFβ and IL-10 have been implicated in the suppressive
function, but also inhibitory receptors like CTLA-4.
Th3 cells develop after oral low dose antigen exposure and have been shown
to inhibit EAE.55 Th3 cells are defined by their production of TGFβ, and to a lesser

15
 CHAPTER 1

extent of IL-4 and IL-10.56 These cells have been shown to suppress Th1 and Th2
cells. In vitro generation of Th3 cells is enhanced by culture with TGFβ, IL-4, IL-10
CHAPTER 1

and IL-12. Th3 cells are activated in an antigen-specific fashion, but suppress in an
antigen-non-specific fashion.
Tr1 cells are induced by antigen stimulation of T cells in the presence of IL-10.57
These cells are characterized by the production of high levels of IL-10, TGFβ and
IL-5, and low or intermediate levels of IL-2, IFNγ and IL-4. It has been shown that Tr1
cells can also be generated with immature DC, IL-10 producing tolerogenic (t)DC,
rapamycin, or vitamin D3 plus dexamethasone.58-61 Co-stimulatory molecules HLA-G
and ILT4 are involved in Tr1 induction by IL-10 producing tDC.59 Tr1 cells regulate
immune responses through the secretion of immunosuppressive cytokines IL-10 and
TGFβ, and they suppress T cell responses, both in vitro and in vivo.62
Thus, iTreg develop in an immunosuppressive environment, most often in
presence of IL-10 or TGFβ, and mediate suppressive function via the same anti-
inflammatory cytokines and/ or interaction with inhibitory molecules. The unique
ability of Treg cells to control immune responses makes them central in the
prevention of immunopathology, autoimmune diseases and allergy, as well as in the
maintenance of allograft tolerance.

Tolerogenic dendritic cells

DC, although regarded traditionally as instigators of immunogenic responses, also
play a critical role in establishing T cell tolerance.63 In the absence of infection and
inflammation, DC are in an immature state and not fully differentiated to induce
immunity. Nevertheless, immature DC continuously circulate through tissues and
into lymphoid organs, capturing self antigens as well as innocuous environmental
proteins. Antigen-loaded immature DC, which lack signal 2, silence T cell reactivity
to these antigens by inducing unresponsiveness or by expanding Treg.64-67 Not
only immature DC can induce tolerance, also mature DC, matured under certain
conditions, can induce tolerance. Maturation of DC in presence of anti-inflammatory
mediators differentiates DC into tolerogenic DC with a corresponding polarizing
signal 3. These anti-inflammatory mediators can be anti-inflammatory cytokines, IL-10
and TGFβ, but also apoptotic cells, vitamin A metabolites or vitamin D metabolites68
(Figure 2). In this way, DC not only activate T cells, they also tolerize T cells to
self-antigens and harmless environmental proteins, thereby minimizing the risk of
developing autoimmune reactions. The importance of DC in maintaining immune
tolerance was shown by CD11c-/- mice, lacking CD11c+ conventional DC, which
developed spontaneous autoimmunity.69 In addition, it was shown that DC were
necessary for Treg homeostatic proliferation, maintaining the Treg compartment.70
DC that induce T cell tolerance can be found in many different organs.71
Intestinal mucosal DC have been proposed to induce immunity to infectious micro-
organisms, and tolerance to commensal bacteria and to food antigens.72 Cross-talk
between intestinal epithelial cells and DC, promotes gut homeostasis through the
induction of tDC via TSLP, retinoic acid and TGFβ.73 These gut tDC were found
to induce Foxp3+ Treg via retinoic acid and TGFβ production. DC also have an

16
 GENERAL INTRODUCTION

IL-1β Pathogens
IL-6 Pro-inflammatory
IL-12
Anti-inflammatory IL-10
TNFα mediators mediators TGFβ

CHAPTER 1
CD40 CD40
TLR CLR

CCR7 p-MHC

CD80/ CD86 p-MHC

CD80/ CD83/ CD86 ILT/ CTLA-4/ PDL

NLR
Treg Anergy
Th cell Th cell Anergy Treg
Apoptosis Treg
Treg
CTL Treg
Th cell
Treg
CTL

mature DC immature DC tolerogenic DC

High expression CCR7 High expression PRR High expression inhibitory molecules
High expression peptide-MHC High endocytosis rate Semi expression co-stimulatory
High expression costimulatory and Low expression costimulatory molecules
adhesion molecules molecules Semi-mature phenotype (CD83int)
Mature phenotype (high CD83) Immature phenotype (low CD83) Anti-inflammatory cytokines
Pro-inflammatory cytokines

Figure 2. Dendritic cell maturation. Immature DC continuously recognize and internalize

antigens, which are processed into peptides and loaded onto MHC class I and II molecules.
After antigen recognition by PRR and in presence of pro-inflammatory mediators, DC
undergo a maturation process. Maturing DC upregulate CCR7, peptide-MHC molecules
and co-stimulatory molecules. This will lead to DC migration toward draining lymph nodes
and activation of naïve T cells. Depending of the type of pathogen and cytokine milieu
encountered, DC produce specific cytokines that will drive the development of different
pathogen-specific effector T cells. In contrast, the presence of anti-inflammatory mediators
will drive DC maturation toward tolerogenic DC. Tolerogenic DC generally display a semi-
mature phenotype with intermediate expression of co-stimulatory molecules, high expression
of inhibitory molecules and produce high amounts of anti-inflammatory cytokines. Together,
this leads to anergy or apoptosis of effector T cells, or induction of Treg. TLR; Toll-like
receptor, CLR; C-type lectin receptor, NLR; Nod-like receptor, p-MHC; peptide-loaded MHC,
Th; T helper cell, CTL; cytotoxic T lymphocyte.

important role in inhalation tolerance and homeostasis in the lung.74 When DC were
specifically depleted during inhalation of harmless antigens, there was a loss of
inhalation tolerance with induction of asthma. Conversely, tolerance was induced
after intratracheal injection of DC. In addition, circulating tDC have been found in
blood, which induce Treg by ligation of inhibitory molecules HLA-G and ILT4, and
IL-10 production.59 Also Langerhans cells in the skin have been shown to induce
tolerance in absence of activation signals.75
The tolerogenic properties of DC can be mediated via soluble mediators or
via cell contact-dependent mechanisms.76 Tolerogenic DC can induce or activate
Treg by IL-10, TGFβ, or retinoic acid production.77 In addition ligation of inhibitory
molecules, like CTLA-4 78, ILT molecules 59 or PD ligand-1/2 79 can induce Treg.

17
 CHAPTER 1

Impaired co-stimulation of immature DC or production of metabolic enzymes that

regulate the abundance of essential amino acids induces anergy or deletion of
CHAPTER 1

effector T cells. Additionally, tDC can induce immune deviation by inhibiting highly
pro-inflammatory Th1 or Th17 responses.76
Thus, in absence of infection and inflammation, or under the control of anti-
inflammatory mediators, immature DC and mature tDC control immune homeostasis
to environmental and self antigens, and prevent potentially lethal tissue destruction
during chronic inflammation, by inducing T cell tolerance.

IL-10+/ IFNγ+ double positive T cells

IL-10 was initially described as Th2-type cytokine CSIF (cytokine synthesis inhibitory
factor) that inhibited cytokine synthesis in Th1 cells 80, and later on was associated
with regulatory T cell responses.60,81 It is now known that the expression of IL-10 is
not specific to Th2 cells or Treg cells but that it is a much more broadly expressed
cytokine, expressed by all Th cell subsets.82 In many chronic infections the presence
of CD4+ T cells that produce high amounts of both IL-10 and IFNγ have been
documented in experimental murine models 83,84 as well as in patients.85,86 These
IL-10/ IFNγ-producing CD4+ T cells share many features with Th1 cells, but were
also the main source of protective IL-10, playing an important regulatory role for
host protection.83,84 These T cells were identified as activated T-bet+Foxp3- Th1 cells
and were distinct from Th2 cells, nTreg cells, or other subsets of iTreg cells. Thus, it
appears that, in some cases, cells with regulatory properties could arise from fully
differentiated Th1 cells as a negative-feedback loop during chronic infections.87,88
In addition, a population of CD4+CD25-CD45RA-CD127- cells in blood was found
that co-expressed IL-10 and IFNγ. These cells were characterized as Foxp3- effector-
like cells that were recently activated and rapidly secreted IL-10 upon restimulation
by which T cell proliferation was suppressed.89 Other CD4+ T cells were shown to
produce both IL-10 and (intermediate levels of) IFNγ, next to other cytokines, and
were examined on their suppressive activity. Stimulation of CD46 (the C3 receptor
which is a co-stimulatory molecule for human T cells) induces high IL-10 production
and moderate TGFβ and IFNγ production. These cells were characterized as Tr1-like
cells that suppress the activation of effector T cells in an IL-10-dependent manner.90
In addition, CD4+LAP+ T cells have been shown to produce high amounts of IL-10,
TGFβ, IFNγ and IL-4 and effectively suppress disease activity as well as the induction
of EAE.91 Furthermore, a population of CD4+LAG3+ Treg cells has been identified.
These cells produce large amounts of IL-10 and moderate amounts of IFNγ, are
characterized as Foxp3- and induce suppression in an IL-10-dependent manner.92
Apart from Foxp3- T cells, also Foxp3-expressing T cells have been described that
co-express IL-10 and IFNγ. These cells were characterized as Th1-like regulatory cells
that express Foxp3, T-bet, IL-10 and IFNγ, and protected against the development
of airway hyperreactivity.93
Thus, it seems that T cells exist that co-produce IL-10 and IFNγ and mediate
immune suppressive functions via IL-10. It will be of interest to determine whether
the IL-10/ IFNγ-producing T cells are Treg that differentiated directly from naïve

18
 GENERAL INTRODUCTION

T cells and have gained additional IFNγ production, or that they derived from Th1
cells that have lost the expression of (most of) their effector cytokines but have

CHAPTER 1
gained expression of IL-10. In favour of the first concept, is that Foxp3+ Treg cells
in mice can lose their Foxp3 expression under certain conditions and acquire Th
cell characteristics with the capacity to produce pro-inflammatory cytokines.94,95 In
contrast, various reports support the other possibility. Cytokines produced by DC
could contribute to the imprinting of IL-10 on Th1 cells. For instance, repetitive
exposure to IL-12 could induce IL-10 on IFNγ-producing cells.96 Several recent
reports suggested that IL-27 might be an important determinant for the induction of
IL-10 on Th cells.97-99 IL-27 can induce IL-10 production in differentiated Th1, Th2 and
Th17 cells, and in nonpolarized CD4+ T cells. In addition, based on the finding that
IL-10+/ IFNγ+ Th1 cells are often found in chronic infection and inflammation 83, 84,
it is likely that IL-10 production by Th1 cells is associated with conditions of
inflammation and high antigenic stimulation. Indeed, Th1 cells that produce both
IFNγ and IL-10 can be generated in vitro by inducing T cells to proliferate with high
levels of antigen-specific or polyclonal stimulation (i.e. strong TCR triggering) in the
absence 100 or presence of IL-12.101

TNFα and anti-TNFα therapy

TNFα
Tumor necrosis factor (TNF)α is a pro-inflammatory cytokine with potent effects on
inflammation and immune responses.102 It was originally described as a cell-derived
factor causing necrosis of tumours.103 TNFα is first produced as a 26 kDa cell surface
polypeptide, called transmembrane (m)TNFα. Soluble (s)TNFα is a 17 kDa cleavage
product of mTNFα, due to activity of the TNFα-converting enzyme (TACE).104 There
are two receptors for TNFα 105: TNFRI, also known as p55 or CD120a, and TNFRII,
also referred to as p75 or CD120b. Most of the TNFα responses known occur
by activation of TNFRI, which is expressed by a wide range of cells. In contrast,
TNFRII exhibits more restricted expression, being found on certain subpopulations
of immune cells, including activated T cells.106 Both sTNFα and mTNFα bind as
homotrimers to the receptors.104 Binding of TNFα to TNFRI may result in apoptosis
via caspase cascades or result in pro-survival signals via NF-κB, which is cell- and
context-dependent.107 TNFRII signalling induces cell survival and cell proliferation
via NF-κB and MAP kinases.106 In addition, an extra level of TNFα regulation is
added by the shedding of TNFRII by Treg, which blocks the pro-inflammatory action
of TNFα.108
TNFα initiates the defence response to local injury, recruits leukocytes and
drives the production of multiple pro-inflammatory cytokines. TNFR-/- mice are
resistant to systemic toxicity upon injection of LPS or SEB (Staphylococcus aureus
enterotoxin B).109 These mice are severely impaired to clear Listeria monocytogenes
and readily succumb to infection, pointing to a crucial role for this cytokine in
protection from infection. In addition, TNFα-/- mice succumb to widespread

19
 CHAPTER 1

dissemination of Mycobacterium tuberculosis as a result of deficiencies in granuloma

formation.110 In humans, mutations in TNFRI have been found. These mutations
CHAPTER 1

cause severe localized inflammation and autoinflammatory syndromes.102 Thus,

TNFα plays an important role in protection from bacterial infection and immune
surveillance.
TNFα is mainly produced by monocytes, macrophages and dendritic cells, but
many other cells, including B cells, T cells and NK cells, have been found to release
TNFα after activation. TNFα activates monocytes, macrophages as well as dendritic
cells, granulocytes, fibroblasts and endothelial cells, and thus, enhances and
propagates local and systemic inflammatory responses.102 It has been shown that
TNFα enhances cytotoxicity of CD8+ T cells 111 and NK cells.112 TNFα induces B cell
proliferation and secretion of immunoglobulins 113,114, and co-stimulates T cells and
thereby enhance proliferation and cytokine production.115-117 Finally, TNFα seems to
be essential for thymocyte development and proliferation.118
Immunoregulatory properties were also ascribed to TNFα, predominantly in mice.
In chronic inflammatory states TNFα-/- mice exhibit more severe inflammation.119
Moreover, TNFα-/- DC were unable to induce IL-10-producing CD4+ T cells.120
Addition of TNFα recovered the impaired ability of TNFα-/- DC to induce IL-10-
producing T cells. Furthermore, TNFRII is highly expressed on human and murine
Treg.121 TNFα may therefore have co-stimulatory effects on Treg, although the
outcome of this co-stimulation in relation to Treg function is a matter of debate.
In murine models, TNFα induces expansion of Foxp3+ Treg which then either show
enhanced suppressive function 122,123 or fail to control inflammation.124 For human
Treg it was shown that TNFα reduces suppressive function.125,126

Anti-TNFα therapy
In chronic inflammation, TNFα can be involved in the disease process as a key factor
for sustenance and amplification of the inflammatory processes as observed in
several immune-mediated inflammatory diseases (IMID), such as rheumatoid arthritis
(RA), ankylosing spondylitis, psoriatic arthritis, multiple sclerosis and inflammatory
bowel diseases.127 Many of these diseases are associated with excessive levels of
TNFα. In RA, TNFα drives much of the pathophysiology in the rheumatic joints.
The introduction of therapeutic anti-TNFα agents has given a major boost to the
treatment of these diseases.128-131 Treatment of RA patients with TNFα blocking
agents results in significant clinical benefit, even in patients with advanced disease.127
Currently, 5 types of anti-TNFα agents are approved for treatment of RA and other
autoimmune diseases; anti-TNFα monoclonal antibodies infliximab, adalimumab
and golimumab, etanercept (a fusion protein of TNFRII and Fc fragment) and
certolizumab pegol (a pegylated Fab fragment of an anti-TNFα antibody).
Inhibition of TNFα affects various cell types and mediators involved in the
pathogenesis of RA.132 Treatment of RA patients with TNFα blocking agents reduces
the numbers of infiltrating granulocytes, T cells, B cells and macrophages in
affected joints by reduced chemokine expression in the synovium in a great number
of patients.133 Anti-TNFα agents also reduce pro-inflammatory cytokine production,

20
 GENERAL INTRODUCTION

i.e. IL-6, TNFα, IL-1β and IL-1α, both in serum and the synovium.134,135 Anti-TNFα
agents influence DC by inhibiting maturation, resulting in reduced co-stimulatory

CHAPTER 1
molecule expression, cytokine production and T cell proliferation.136 Several studies
show that Treg of RA patients are defective and fail to suppress inflammation.
Treatment of these patients with anti-TNFα induces or expands functional Treg 137,138,
and recovers their suppressive function.126 In addition, TNFα blocking agents cause
changes in T cell polarization profiles.139-141 In parallel to these findings in RA patients,
in vitro studies showed that neutralizing TNFα inhibits T cell proliferation and IFNγ
production 115,142 and enhances suppressive capacity by Treg.125

Tolerance-inducing therapies
In many diseases, like in autoimmune diseases or allergies, the immune system
has shifted toward an undesired status of immune activation against self or
environmental antigens. In addition, patients with organ transplantation might
suffer graft rejection due to recognition of non-self antigens/ MHC molecules.
Therapies that prevent or reduce immune activation may be a means to combat
disease or graft rejection. Current therapies, which include immunosuppressive
drugs, do not specifically target the cause of the disease or transplant rejection and,
in addition, are associated with considerable side effects. Therapies that specifically
target the immunopathogenesis are explored to combat autoimmune disease or
graft rejection.
The introduction of biological therapeutics, usually directed against a cytokine
or cell surface molecule, has led to a more specific treatment of various IMID
and improved clinical and functional outcomes.127 Besides the above described
anti-TNFα agents, other biologicals have been developed that target effector
molecules involved in immune activation. These include an IL-6 receptor monoclonal
antibody (tocilizumab) 143, a B cell-depleting monoclonal antibody (rituximab; anti-
CD20) 144 and blockade of CD4 T cell co-stimulation with abatacept (fusion molecule
of CTLA-4 to IgG Fc tail) 145; all proven beneficial in the treatment of RA. An IL-1
receptor antagonist (anakinra) has been used for treatment of RA and neonatal-
onset multisystem inflammatory diseases (NOMID).146,147 In addition, phase II clinical
trials with a T cell-depleting monoclonal antibody (otelixizumab) have proven
beneficial for the treatment of type I diabetes.148 Overall, these biologic agents are
well tolerated, however, immunogenicity, i.e. antibody formation against the drug,
affects the clinical outcome in some patients.149 Furthermore, these agents mediate
their effects in an antigen non-specific manner, resulting in generalized immune
suppression.Given the central role of Treg in immune regulation, targeting of Treg
might benefit the restoration of immune homeostasis in autoimmune diseases or
after transplantation. Treg can be targeted in vivo with immunosuppressive agents
and/ or the anti-inflammatory cytokine IL-10.61,150 In addition, Treg can be generated
or expanded in vitro by use of cytokines IL-2, IL-35 and TGFβ, dendritic cells or
immunosuppressive agents.151 Transfer of expanded Treg cells is effective in the
prevention and, in some cases, the treatment of allograft rejection or ongoing

21
 CHAPTER 1

inflammatory and autoimmune diseases in mice.152 For clinical use of in vitro

expanded Treg cells it is important to generate a pure population of functional Treg.
CHAPTER 1

In addition, it is necessary that these cells are functionally stable and maintain their
suppressive function, since Treg cells display plasticity.94,95
Tolerogenic DC may also have a strong potential as a cellular therapy to treat
autoimmune disorders, hypersensitivity diseases or to prevent undesired immune
responses against allogeneic transplants. Given the central role of DC in immunity
and tolerance, they are ideal therapeutic targets for pharmacological modulation of
immune responses in an antigen-specific manner. Tolerogenic DC induce Treg, which
in turn induce more widespread tolerance.153,154 In this respect, ex vivo generated
tDC are considered as an important therapeutic alternative to maintain, restore or
induce antigen-specific immunological tolerance.155 Evidence from animal models
has shown that the injection of ex vivo generated tDC can prevent or stop transplant
rejection, for example in skin graft 156 and heart graft models 157, or re-establish self-
tolerance in autoimmune diseases, like collagen-induced arthritis 158, diabetes 159
and experimental autoimmune encephalomyelitis.160 These positive results have
led to the first clinical trials with monocyte-derived tDC for type I diabetes 161 and
for rheumatoid arthritis (J.D. Isaacs and co-workers, Newcastle University; and
R. Thomas and co-workers, University of Queensland). Clinical trials with tDC have
not started in the field of transplantation, although monocyte-derived tDC from
rhesus macaques show promising results when infused into allogeneic recipients
showing a clear T cell hyporesponsiveness to donor allogeneic-antigens.162
Human monocytes-derived tDC can be differentiated in vitro using various
immunosuppressive and anti-inflammatory compounds. Among others, the anti-
inflammatory cytokines IL-10 and TGFβ, glucocorticoids, like dexamethasone,
anti-inflammatory/ immunosuppressive drugs, like rapamycin and aspirin, and
many other anti-inflammatory mediators, e.g. the active form of vitamin D3
(1α, 25-dihydroxyvitamin D3) or vitamin D3 analogs and calcineurin inhibitors
have all been identified to induce DC with tolerogenic properties in vitro.163-165
These immunosuppressive agents influence the development, maturation,
and consequently the function of the DC. In vitro generated tDC are generally
characterized by an immature or semi-mature phenotype, with low expression of
co-stimulatory molecules and high expression of inhibitory molecules. In addition,
these tDC produce low amounts of pro-inflammatory cytokines and high amounts
of anti-inflammatory cytokines. Thereby tDC mediate peripheral T cell tolerance by
inducing anergy or apoptosis of effector T cells, and/ or induction of Treg.155

Scope of this thesis

Regulation of immune responses is pivotal for prevention of excessive immunity
upon immune activation and for maintenance of immune homeostasis. In this
thesis, we examined the role of tolerogenic DC and TNFα in regulation of immune
responses.

22
 GENERAL INTRODUCTION

One of the most important tolerance inducing mechanisms of tDC is the induction
of regulatory T cells. For human induced Treg however, specific markers are not

CHAPTER 1
available, since the described iTreg markers only identify specific iTreg subsets or are
also expressed by activated T cells. Therefore, analysis of the functional suppressive
capacity of Treg is to date the best read out system for human Treg. In chapter 2,
we optimised an in vitro suppression assay to analyse suppression of proliferation of
activated responder T cells by tDC-induced Treg. We demonstrate that titration of
APC, which are used to stimulate proliferation of responder T cells, is important for
the optimal read out of suppression. Inclusion of a negative control condition, e.g.
T cells primed by mature immunoactivatory DC, is demonstrated to be essential to
discriminate between specific and aspecific suppression.
In chapter 3, we generated an array of clinically-applicable tDC with various
tolerance-inducing compounds; vitamin D3, IL-10, dexamethasone, TGFβ and
rapamycin. Although tDC have been studied extensively in research settings, the
described tDC methods have often not been converted into clinically applicable
protocols. We cultured DC under current good manufacturing practices (cGMP)
and compared essential functional characteristics of tDC, i.e. migratory capacity,
Treg induction and functional stability. We show that for good migratory capacity
and stable immunosuppressive phenotype and function, an additional maturation
stimulus was necessary. We identified IL-10-generated tDC as superior tolerogenic
DC in that they produced high amounts of IL-10, induced potent Treg from naïve
T cells, and inhibited memory T cell activation.
TNFα is a potent pro-inflammatory cytokine that co-stimulates T cells. However,
the precise role of TNFα in naïve CD4+ T cell priming is unclear. In chapter 4 we
demonstrate that neutralization of TNFα during naïve T cell priming induced IL-10
expression, which was mediated via inhibition of TNFα–TNFRII interaction. In
addition, anti-TNFα treated cells displayed a strong regulatory gene profile and
enhanced suppressive capacity.
A large percentage of IL-10+ T cells induced by IL-10 tDC co-express IFNγ.
IL-10 and IFNγ co-expression has been described for Tr1 cells, characterized as
IL-10+ suppressive cells, and for Th1 cells, characterized as IL-10+ T-bet+ cells. To
further define the phenotype and function of IL-10+/ IFNγ+ co-expressing T cells,
we performed whole genome gene expression analyses in chapter 5. IL-10+/ IFNγ+
T cells have a regulatory gene profile compared to IFNγ+/ IL-10- T cells. In addition,
IL-10+/ IFNγ+ T cells displayed a strong regulatory phenotype and potent suppressive
capacity, demonstrating a regulatory role for IL-10+/ IFNγ+ co-expressing T cells.
Chapter 6 provides a summary and general discussion of the results described
in this thesis.

23
 CHAPTER 1

References
1. Banchereau, J., Briere, F., Caux, via MHC class II and CD40 molecules
CHAPTER 1

C., Davoust, J., Lebecque, S., Liu, and downregulation by IL-4 and IL-10. J
Y.J., Pulendran, B., Palucka, K. Exp. Med. 1996; 184:741-746.
Immunobiology of dendritic cells. Annu. 15. de Jong, E.C., Smits, H.H., Kapsenberg,
Rev. Immunol. 2000; 18:767-811. M.L. Dendritic cell-mediated
2. Reizis, B., Bunin, A., Ghosh, H.S., Lewis, T cell polarization. Springer Semin.
K.L., Sisirak, V. Plasmacytoid Dendritic Immunopathol. 2005; 26:289-307.
Cells: Recent Progress and Open
16. Zhu, J., Yamane, H., Paul, W.E.
Questions. Annu. Rev. Immunol. 2011;
Differentiation of effector CD4 T cell
29:163-183.
populations. Annu. Rev. Immunol. 2010;
3. Kawai, T., Akira, S. The role of pattern- 28:445-489.
recognition receptors in innate
immunity: update on Toll-like receptors. 17. Oliphant, C.J., Barlow, J.L., McKenzie,
Nat Immunol 2010; 11:373-384. A.N. Insights into the initiation of type
2 immune responses. Immunology 2011;
4. Medzhitov, R. Recognition of
134:378-385.
microorganisms and activation of
the immune response. Nature 2007; 18. Langrish, C.L., Chen, Y., Blumenschein,
449:819-826. W.M., Mattson, J., Basham, B., Sedgwick,
5. Figdor, C.G., van, K.Y., Adema, G.J. C- J.D., McClanahan, T., Kastelein, R.A.,
type lectin receptors on dendritic cells Cua, D.J. IL-23 drives a pathogenic T cell
and Langerhans cells. Nat. Rev. Immunol. population that induces autoimmune
2002; 2:77-84. inflammation. J. Exp. Med. 2005;
201:233-240.
6. Trombetta, E.S., Mellman, I. Cell Biology
of Antigen Processing in Vitro and in 19. Korn, T., Bettelli, E., Oukka, M., Kuchroo,
Vivo. Annu. Rev. Immunol. 2004; 23:975- V.K. IL-17 and Th17 Cells. Annu. Rev.
1028. Immunol. 2009; 27:485-517.
7. Sallusto, F., Lanzavecchia, A. 20. Deenick, E.K., Ma, C.S., Brink, R.,
Understanding dendritic cell and T- Tangye, S.G. Regulation of T follicular
lymphocyte traffic through the analysis helper cell formation and function by
of chemokine receptor expression. antigen presenting cells. Curr. Opin.
Immunol. Rev. 2000; 177:134-140. Immunol. 2011; 23:111-118.
8. Kapsenberg, M.L. Dendritic-cell control 21. Veldhoen, M., Uyttenhove, C., van,
of pathogen-driven T-cell polarization. S.J., Helmby, H., Westendorf, A., Buer,
Nat Rev Immunol 2003; 3:984-993. J., Martin, B., Wilhelm, C., Stockinger,
9. Kalinski, P., Hilkens, C.M.U., Wierenga, B. Transforming growth factor-beta
E.A., Kapsenberg, M.L. T-cell priming ‘reprograms’ the differentiation of
by type-1and type-2 polarized dendritic T helper 2 cells and promotes an
cells: the concept of a third signal. interleukin 9-producing subset. Nat.
Immunology Today 1999; 20:561-567. Immunol. 2008; 9:1341-1346.
10. Jensen, P.E. Recent advances in antigen 22. Dardalhon, V., Awasthi, A., Kwon, H.,
processing and presentation. Nat Galileos, G., Gao, W., Sobel, R.A.,
Immunol 2007; 8:1041-1048. Mitsdoerffer, M., Strom, T.B., Elyaman,
11. Kurts, C., Robinson, B.W.S., Knolle, P.A. W., Ho, I.C., Khoury, S., Oukka, M.,
Cross-priming in health and disease. Nat Kuchroo, V.K. IL-4 inhibits TGF-beta-
Rev Immunol 2010; 10:403-414. induced Foxp3+ T cells and, together
12. Greenwald, R.J., Freeman, G.J., Sharpe, with TGF-beta, generates IL-9+ IL-10+
A.H. The B7 Family Revisited. Annu. Rev. Foxp3(-) effector T cells. Nat. Immunol.
Immunol. 2004; 23:515-548. 2008; 9:1347-1355.
13. Pulendran, B. Variegation of the Immune 23. Schmitt, E., Germann, T., Goedert,
Response with Dendritic Cells and S., Hoehn, P., Huels, C., Koelsch, S.,
Pathogen Recognition Receptors. J Kuhn, R., Muller, W., Palm, N., Rude,
Immunol 2005; 174:2457-2465. E. IL-9 production of naive CD4+
14. Koch, F., Stanzl, U., Jennewein, P., Janke, T cells depends on IL-2, is synergistically
K., Heufler, C., Kampgen, E., Romani, N., enhanced by a combination of TGF-beta
Schuler, G. High level IL-12 production and IL-4, and is inhibited by IFN-gamma.
by murine dendritic cells: upregulation J. Immunol. 1994; 153:3989-3996.

24
 GENERAL INTRODUCTION

24. Trifari, S., Kaplan, C.D., Tran, E.H., 32. Hegazy, A.N., Peine, M., Helmstetter,
Crellin, N.K., Spits, H. Identification of a C., Panse, I., Frohlich, A., Bergthaler, A.,
human helper T cell population that has Flatz, L., Pinschewer, D.D., Radbruch,

CHAPTER 1
abundant production of interleukin 22 A., Lohning, M. Interferons direct Th2
and is distinct from T(H)-17, T(H)1 and cell reprogramming to generate a stable
T(H)2 cells. Nat. Immunol. 2009; 10:864- GATA-3(+)T-bet(+) cell subset with
871. combined Th2 and Th1 cell functions.
25. Nograles, K.E., Zaba, L.C., Shemer, Immunity. 2010; 32:116-128.
A., Fuentes-Duculan, J., Cardinale, I., 33. Lu, K.T., Kanno, Y., Cannons, J.L.,
Kikuchi, T., Ramon, M., Bergman, R., Handon, R., Bible, P., Elkahloun, A.G.,
Krueger, J.G., Guttman-Yassky, E. IL- Anderson, S.M., Wei, L., Sun, H., O’Shea,
22-producing “T22” T cells account for J.J., Schwartzberg, P.L. Functional and
upregulated IL-22 in atopic dermatitis epigenetic studies reveal multistep
despite reduced IL-17-producing TH17 differentiation and plasticity of in vitro-
T cells. J. Allergy Clin. Immunol. 2009; generated and in vivo-derived follicular
123:1244-1252. T helper cells. Immunity. 2011; 35:622-
26. Duhen, T., Geiger, R., Jarrossay, D., 632.
Lanzavecchia, A., Sallusto, F. Production 34. Szabo, S.J., Kim, S.T., Costa, G.L.,
of interleukin 22 but not interleukin 17 by Zhang, X., Fathman, C.G., Glimcher,
a subset of human skin-homing memory L.H. A novel transcription factor, T-bet,
T cells. Nat. Immunol. 2009; 10:857-863. directs Th1 lineage commitment. Cell
27. Wolk, K., Witte, E., Wallace, E., 2000; 100:655-669.
Docke, W.D., Kunz, S., Asadullah, K., 35. Zheng, W., Flavell, R.A. The transcription
Volk, H.D., Sterry, W., Sabat, R. IL- factor GATA-3 is necessary and sufficient
22 regulates the expression of genes for Th2 cytokine gene expression in CD4
responsible for antimicrobial defense, T cells. Cell 1997; 89:587-596.
cellular differentiation, and mobility 36. Nakayamada, S., Takahashi, H., Kanno,
in keratinocytes: a potential role in Y., O’Shea, J.J. Helper T cell diversity
psoriasis. Eur. J. Immunol. 2006; and plasticity. Curr. Opin. Immunol.
36:1309-1323. 2012.
28. Boniface, K., Guignouard, E., Pedretti, 37. Ghoreschi, K., Laurence, A., Yang, X.P.,
N., Garcia, M., Delwail, A., Bernard, Tato, C.M., McGeachy, M.J., Konkel,
F.X., Nau, F., Guillet, G., Dagregorio, G., J.E., Ramos, H.L., Wei, L., Davidson,
Yssel, H., Lecron, J.C., Morel, F. A role for T.S., Bouladoux, N., Grainger, J.R.,
T cell-derived interleukin 22 in psoriatic Chen, Q., Kanno, Y., Watford, W.T., Sun,
skin inflammation. Clin. Exp. Immunol. H.W., Eberl, G., Shevach, E.M., Belkaid,
2007; 150:407-415. Y., Cua, D.J., Chen, W., O’Shea, J.J.
29. Hirota, K., Duarte, J.H., Veldhoen, M., Generation of pathogenic T(H)17 cells
Hornsby, E., Li, Y., Cua, D.J., Ahlfors, in the absence of TGF-beta signalling.
H., Wilhelm, C., Tolaini, M., Menzel, U., Nature 2010; 467:967-971.
Garefalaki, A., Potocnik, A.J., Stockinger, 38. Oestreich, K.J., Mohn, S.E., Weinmann,
B. Fate mapping of IL-17-producing A.S. Molecular mechanisms that control
T cells in inflammatory responses. Nat the expression and activity of Bcl-6 in
Immunol 2011; 12:255-263. T(H)1 cells to regulate flexibility with a
30. Bending, D., De la Pena, H., Veldhoen, T(FH)-like gene profile. Nat. Immunol.
M., Phillips, J.M., Uyttenhove, C., 2012.
Stockinger, B., Cooke, A. Highly purified 39. Wei, G., Wei, L., Zhu, J., Zang, C., Hu-
Th17 cells from BDC2.5NOD mice Li, J., Yao, Z., Cui, K., Kanno, Y., Roh,
convert into Th1-like cells in NOD/SCID T.Y., Watford, W.T., Schones, D.E., Peng,
recipient mice. J. Clin. Invest 2009; W., Sun, H.W., Paul, W.E., O’Shea, J.J.,
119:565-572. Zhao, K. Global mapping of H3K4me3
31. Smits, H.H., van Beelen, A.J., Hessle, and H3K27me3 reveals specificity and
C., Westland, R., de, J.E., Soeteman, E., plasticity in lineage fate determination
Wold, A., Wierenga, E.A., Kapsenberg, of differentiating CD4+ T cells. Immunity.
M.L. Commensal Gram-negative 2009; 30:155-167.
bacteria prime human dendritic cells for 40. Mukasa, R., Balasubramani, A., Lee, Y.K.,
enhanced IL-23 and IL-27 expression Whitley, S.K., Weaver, B.T., Shibata, Y.,
and enhanced Th1 development. Eur. J. Crawford, G.E., Hatton, R.D., Weaver,
Immunol 2004; 34:1371-1380. C.T. Epigenetic instability of cytokine

25
 CHAPTER 1

and transcription factor gene loci 51. Lohr, J., Knoechel, B., Abbas, A.K.
underlies plasticity of the T helper 17 cell Regulatory T cells in the periphery.
lineage. Immunity. 2010; 32:616-627. Immunol. Rev. 2006; 212:149-162.
CHAPTER 1

41. Moore, K.W., de Waal, M.R., Coffman, 52. Campbell, D.J., Koch, M.A. Phenotypical
R.L., O’Garra, A. Interleukin-10 and and functional specialization of FOXP3+
the interleukin-10 receptor. Annu. Rev. regulatory T cells. Nat. Rev. Immunol.
Immunol. 2001; 19:683-765. 2011; 11:119-130.
42. Kuhn, R., Lohler, J., Rennick, D., Rajewsky, 53. Duhen, T., Duhen, R., Lanzavecchia, A.,
K., Muller, W. Interleukin-10-deficient Sallusto, F., Campbell, D.J. Functionally
mice develop chronic enterocolitis. Cell distinct subsets of human FOXP3+ Treg
1993; 75:263-274. cells that phenotypically mirror effector
43. Sellon, R.K., Tonkonogy, S., Schultz, M., TH cells. Blood 2012; 119:4430-4440.
Dieleman, L.A., Grenther, W., Balish, 54. Wan, Y.Y., Flavell, R.A. The roles
E., Rennick, D.M., Sartor, R.B. Resident for cytokines in the generation and
enteric bacteria are necessary for maintenance of regulatory T cells.
development of spontaneous colitis and Immunol. Rev. 2006; 212:114-130.
immune system activation in interleukin- 55. Chen, Y., Kuchroo, V.K., Inobe, J., Hafler,
10-deficient mice. Infect. Immun. 1998; D.A., Weiner, H.L. Regulatory T cell clones
66:5224-5231. induced by oral tolerance: suppression
44. Groux, H., Cottrez, F., Rouleau, M., of autoimmune encephalomyelitis.
Mauze, S., Antonenko, S., Hurst, S., Science 1994; 265:1237-1240.
McNeil, T., Bigler, M., Roncarolo, M.G., 56. Weiner, H.L. Induction and mechanism
Coffman, R.L. A transgenic model to of action of transforming growth factor-
analyze the immunoregulatory role of IL- beta-secreting Th3 regulatory cells.
10 secreted by antigen-presenting cells. Immunol. Rev. 2001; 182:207-214.
J. Immunol. 1999; 162:1723-1729. 57. Levings, M.K., Sangregorio, R., Galbiati,
45. Sakaguchi, S., Ono, M., Setoguchi, F., Squadrone, S., Waal Malefyt, R.,
R., Yagi, H., Hori, S., Fehervari, Z., Roncarolo, M.G. IFN-{{alpha}} and IL-
Shimizu, J., Takahashi, T., Nomura, T. 10 Induce the Differentiation of Human
Foxp3+ CD25+ CD4+ natural regulatory Type 1 T Regulatory Cells. J Immunol
T cells in dominant self-tolerance and 2001; 166:5530-5539.
autoimmune disease. Immunol. Rev. 58. Levings, M.K., Gregori, S., Tresoldi, E.,
2006; 212:8-27. Cazzaniga, S., Bonini, C., Roncarolo,
46. Sakaguchi, S., Sakaguchi, N., Asano, M.G. Differentiation of Tr1 cells by
M., Itoh, M., Toda, M. Immunologic immature dendritic cells requires IL-
self-tolerance maintained by activated 10 but not CD25+CD4+ Tr cells. Blood
T cells expressing IL-2 receptor alpha- 2005; 105:1162-1169.
chains (CD25). Breakdown of a single 59. Gregori, S., Tomasoni, D., Pacciani, V.,
mechanism of self-tolerance causes Scirpoli, M., Battaglia, M., Magnani,
various autoimmune diseases. J Immunol C.F., Hauben, E., Roncarolo, M.G.
1995; 155:1151-1164. Differentiation of type 1 T regulatory (Tr1)
47. McHugh, R.S., Shevach, E.M. Cutting cells by tolerogenic DC-10 requires the
Edge: Depletion of CD4+CD25+ IL-10-dependent ILT4/HLA-G pathway.
regulatory T cells is necessary, but not Blood 2010; 116:935-944.
sufficient, for induction of organ-specific 60. Barrat, F.J., Cua, D.J., Boonstra, A.,
autoimmune disease. J Immunol 2002; Richards, D.F., Crain, C., Savelkoul,
168:5979-5983. H.F., de Waal-Malefyt, R., Coffman, R.L.,
48. Hori, S., Nomura, T., Sakaguchi, S. Hawrylowicz, C.M., O’Garra, A. In vitro
Control of regulatory T cell development generation of interleukin 10-producing
by the transcription factor Foxp3. regulatory CD4(+) T cells is induced by
Science 2003; 299:1057-1061. immunosuppressive drugs and inhibited
49. Fontenot, J.D., Gavin, M.A., Rudensky, by T helper type 1 (Th1)- and Th2-
A.Y. Foxp3 programs the development inducing cytokines. J. Exp. Med. 2002;
and function of CD4+CD25+ regulatory 195:603-616.
T cells. Nat. Immunol. 2003; 4:330-336. 61. Battaglia, M., Stabilini, A., Draghici, E.,
50. Sakaguchi, S., Miyara, M., Costantino, Gregori, S., Mocchetti, C., Bonifacio,
C.M., Hafler, D.A. FOXP3+ regulatory E., Roncarolo, M.G. Rapamycin and
T cells in the human immune system. Nat interleukin-10 treatment induces T
Rev Immunol 2010; 10:490-500. regulatory type 1 cells that mediate

26
 GENERAL INTRODUCTION

antigen-specific transplantation 72. Rescigno, M., Di, S.A. Dendritic cells in

tolerance. Diabetes 2006; 55:40-49. intestinal homeostasis and disease. J.
62. Roncarolo, M.G., Gregori, S., Battaglia, Clin. Invest 2009; 119:2441-2450.

CHAPTER 1
M., Bacchetta, R., Fleischhauer, K., 73. Iliev, I.D., Spadoni, I., Mileti, E.,
Levings, M.K. Interleukin-10-secreting Matteoli, G., Sonzogni, A., Sampietro,
type 1 regulatory T cells in rodents and G.M., Foschi, D., Caprioli, F., Viale, G.,
humans. Immunol. Rev. 2006; 212:28-50. Rescigno, M. Human intestinal epithelial
63. Steinman, R.M., Hawiger, D., cells promote the differentiation of
Nussenzweig, M.C. Tolerogenic tolerogenic dendritic cells. Gut 2009.
dendritic cells. Annu. Rev. Immunol. 74. de Heer, H.J., Hammad, H., Soullie,
2003; 21:685-711. T., Hijdra, D., Vos, N., Willart, M.A.,
64. Jonuleit, H., Schmitt, E., Schuler, G., Hoogsteden, H.C., Lambrecht, B.N.
Knop, J., Enk, A.H. Induction of Interleukin Essential role of lung plasmacytoid
10-producing, Nonproliferating CD4+ dendritic cells in preventing asthmatic
T Cells with Regulatory Properties by reactions to harmless inhaled antigen. J.
Repetitive Stimulation with Allogeneic Exp. Med. 2004; 200:89-98.
Immature Human Dendritic Cells. The 75. Gregori, S. Dendritic cells in networks
Journal of Experimental Medicine 2000; of immunological tolerance. Tissue
192:1213-1222. Antigens 2011; 77:89-99.
65. Dhodapkar, M.V., Steinman, R.M., 76. Cools, N., Ponsaerts, P., Van Tendeloo,
Krasovsky, J., Munz, C., Bhardwaj, N. V.F.I., Berneman, Z.N. Balancing
Antigen-specific inhibition of effector between immunity and tolerance:
T cell function in humans after injection an interplay between dendritic cells,
of immature dendritic cells. J. Exp. Med. regulatory T cells, and effector T cells. J
2001; 193:233-238. Leukoc Biol 2007; 82:1365-1374.
66. Hawiger, D., Inaba, K., Dorsett, Y., Guo, 77. Maldonado, R.A., von Andrian, U.H.
M., Mahnke, K., Rivera, M., Ravetch, How tolerogenic dendritic cells induce
J.V., Steinman, R.M., Nussenzweig, M.C. regulatory T cells. Adv. Immunol. 2010;
Dendritic Cells Induce Peripheral T Cell 108:111-165.
Unresponsiveness Under Steady State 78. Sansom, D.M., Walker, L.S. The role
Conditions In Vivo. J. Exp. Med. 2001; of CD28 and cytotoxic T-lymphocyte
194:769-780. antigen-4 (CTLA-4) in regulatory T-cell
67. Roncarolo, M.G., Levings, M.K., biology. Immunol. Rev. 2006; 212:131-
Traversari, C. Differentiation of T 148.
Regulatory Cells by Immature Dendritic 79. Keir, M.E., Butte, M.J., Freeman, G.J.,
Cells. The Journal of Experimental Sharpe, A.H. PD-1 and its ligands in
Medicine 2001; 193:5F-10. tolerance and immunity. Annu. Rev.
68. Morelli, A.E., Thomson, A.W. Tolerogenic Immunol. 2008; 26:677-704.
dendritic cells and the quest for 80. Fiorentino, D.F., Bond, M.W., Mosmann,
transplant tolerance. Nat. Rev. Immunol. T.R. Two types of mouse T helper cell. IV.
2007; 7:610-621. Th2 clones secrete a factor that inhibits
69. Ohnmacht, C., Pullner, A., King, cytokine production by Th1 clones. J.
S.B., Drexler, I., Meier, S., Brocker, T., Exp. Med. 1989; 170:2081-2095.
Voehringer, D. Constitutive ablation of 81. Vieira, P.L., Christensen, J.R., Minaee, S.,
dendritic cells breaks self-tolerance of O’Neill, E.J., Barrat, F.J., Boonstra, A.,
CD4 T cells and results in spontaneous Barthlott, T., Stockinger, B., Wraith, D.C.,
fatal autoimmunity. J. Exp. Med. 2009; O’Garra, A. IL-10-secreting regulatory
206:549-559. T cells do not express Foxp3 but
70. Suffner, J., Hochweller, K., Kuhnle, have comparable regulatory function
M.C., Li, X., Kroczek, R.A., Garbi, N., to naturally occurring CD4+CD25+
Hammerling, G.J. Dendritic cells support regulatory T cells. J. Immunol. 2004;
homeostatic expansion of Foxp3+ 172:5986-5993.
regulatory T cells in Foxp3.LuciDTR 82. Saraiva, M., O’Garra, A. The regulation
mice. J. Immunol. 2010; 184:1810-1820. of IL-10 production by immune cells. Nat
71. Belkaid, Y., Oldenhove, G. Tuning Rev Immunol 2010; 10:170-181.
microenvironments: induction of 83. Jankovic, D., Kullberg, M.C., Feng, C.G.,
regulatory T cells by dendritic cells. Goldszmid, R.S., Collazo, C.M., Wilson,
Immunity. 2008; 29:362-371. M., Wynn, T.A., Kamanaka, M., Flavell,

27
 CHAPTER 1

R.A., Sher, A. Conventional T-bet(+) protect against airway hyper-reactivity.

Foxp3(-) Th1 cells are the major source of Nat. Immunol. 2004; 5:1149-1156.
host-protective regulatory IL-10 during 94. Zhou, X., Bailey-Bucktrout, S.L.,
CHAPTER 1

intracellular protozoan infection. J. Exp. Jeker, L.T., Penaranda, C., Martinez-

Med. 2007; 204:273-283. Llordella, M., Ashby, M., Nakayama, M.,
84. Anderson, C.F., Oukka, M., Kuchroo, Rosenthal, W., Bluestone, J.A. Instability
V.J., Sacks, D. CD4(+)CD25(-)Foxp3(-) of the transcription factor Foxp3 leads to
Th1 cells are the source of IL-10- the generation of pathogenic memory
mediated immune suppression in T cells in vivo. Nat. Immunol. 2009;
chronic cutaneous leishmaniasis. J. Exp. 10:1000-1007.
Med. 2007; 204:285-297. 95. Murai, M., Turovskaya, O., Kim, G.,
85. Pohl-Koppe, A., Balashov, K.E., Steere, Madan, R., Karp, C.L., Cheroutre, H.,
A.C., Logigian, E.L., Hafler, D.A. Kronenberg, M. Interleukin 10 acts on
Identification of a T cell subset capable regulatory T cells to maintain expression
of both IFN-gamma and IL-10 secretion of the transcription factor Foxp3 and
in patients with chronic Borrelia suppressive function in mice with colitis.
burgdorferi infection. J. Immunol. 1998; Nat. Immunol. 2009; 10:1178-1184.
160:1804-1810. 96. Meyaard, L., Hovenkamp, E., Otto,
86. Gerosa, F., Nisii, C., Righetti, S., S.A., Miedema, F. IL-12-induced IL-
Micciolo, R., Marchesini, M., Cazzadori, 10 production by human T cells as a
A., Trinchieri, G. CD4(+) T cell clones negative feedback for IL-12-induced
producing both interferon-gamma immune responses. J. Immunol. 1996;
and interleukin-10 predominate in 156:2776-2782.
bronchoalveolar lavages of active 97. Awasthi, A., Carrier, Y., Peron, J.P.,
pulmonary tuberculosis patients. Clin. Bettelli, E., Kamanaka, M., Flavell, R.A.,
Immunol. 1999; 92:224-234. Kuchroo, V.K., Oukka, M., Weiner, H.L.
87. Trinchieri, G. Interleukin-10 production A dominant function for interleukin 27
by effector T cells: Th1 cells show self in generating interleukin 10-producing
control. J. Exp. Med. 2007; 204:239-243. anti-inflammatory T cells. Nat. Immunol.
88. O’Garra, A., Vieira, P. TH1 cells control 2007; 8:1380-1389.
themselves by producing interleukin-10. 98. Fitzgerald, D.C., Zhang, G.X., El-
Nat Rev Immunol 2007; 7:425-428. Behi, M., Fonseca-Kelly, Z., Li, H.,
89. Haringer, B., Lozza, L., Steckel, Yu, S., Saris, C.J., Gran, B., Ciric, B.,
B., Geginat, J. Identification and Rostami, A. Suppression of autoimmune
characterization of IL-10/IFN-{gamma}- inflammation of the central nervous
producing effector-like T cells with system by interleukin 10 secreted by
regulatory function in human blood. J. interleukin 27-stimulated T cells. Nat.
Exp. Med. 2009; 206:1009-1017. Immunol. 2007; 8:1372-1379.
90. Kemper, C., Chan, A.C., Green, J.M., 99. Stumhofer, J.S., Silver, J.S., Laurence,
Brett, K.A., Murphy, K.M., Atkinson, J.P. A., Porrett, P.M., Harris, T.H., Turka,
Activation of human CD4+ cells with L.A., Ernst, M., Saris, C.J., O’Shea, J.J.,
CD3 and CD46 induces a T-regulatory Hunter, C.A. Interleukins 27 and 6 induce
cell 1 phenotype. Nature 2003; 421:388- STAT3-mediated T cell production of
392. interleukin 10. Nat. Immunol. 2007;
91. Ochi, H., Abraham, M., Ishikawa, H., 8:1363-1371.
Frenkel, D., Yang, K., Basso, A.S., Wu, H., 100. Gabrysova, L., Nicolson, K.S., Streeter,
Chen, M.L., Gandhi, R., Miller, A., Maron, H.B., Verhagen, J., Sabatos-Peyton, C.A.,
R., Weiner, H.L. Oral CD3-specific Morgan, D.J., Wraith, D.C. Negative
antibody suppresses autoimmune feedback control of the autoimmune
encephalomyelitis by inducing CD4+ response through antigen-induced
CD25- LAP+ T cells. Nat. Med. 2006; differentiation of IL-10-secreting Th1
12:627-635. cells. J. Exp. Med. 2009; 206:1755-1767.
92. Fujio, K., Okamura, T., Yamamoto, K. The 101. Saraiva, M., Christensen, J.R., Veldhoen,
Family of IL-10-secreting CD4+ T cells. M., Murphy, T.L., Murphy, K.M., O’Garra,
Adv. Immunol. 2010; 105:99-130. A. Interleukin-10 production by Th1
93. Stock, P., Akbari, O., Berry, G., Freeman, cells requires interleukin-12-induced
G.J., DeKruyff, R.H., Umetsu, D.T. STAT4 transcription factor and ERK MAP
Induction of T helper type 1-like kinase activation by high antigen dose.
regulatory cells that express Foxp3 and Immunity. 2009; 31:209-219.

28
 GENERAL INTRODUCTION

102. Aggarwal, B.B. Signalling pathways of cells by recombinant human tumor

the TNF superfamily: a double-edged necrosis factor. Cell Immunol 1989;
sword. Nat. Rev. Immunol. 2003; 3:745- 120:154-164.

CHAPTER 1
756. 112. Eisenthal, A., Rosenberg, S.A.
103. Carswell, E.A., Old, L.J., Kassel, R.L., The effect of various cytokines on
Green, S., Fiore, N., Williamson, B. An the in vitro induction of antibody-
endotoxin-induced serum factor that dependent cellular cytotoxicity in
causes necrosis of tumors. Proc. Natl. murine cells. Enhancement of IL-2-
Acad. Sci. U. S. A 1975; 72:3666-3670. induced antibody-dependent cellular
104. Horiuchi, T., Mitoma, H., Harashima, cytotoxicity activity by IL-1 and tumor
S., Tsukamoto, H., Shimoda, T. necrosis factor-alpha. J Immunol 1989;
Transmembrane TNF-alpha: structure, 142:2307-2313.
function and interaction with anti-TNF 113. Kehrl, J.H., Miller, A., Fauci, A.S. Effect of
agents. Rheumatology. (Oxford) 2010; tumor necrosis factor alpha on mitogen-
49:1215-1228. activated human B cells. J Exp. Med.
105. Tartaglia, L.A., Weber, R.F., Figari, 1987; 166:786-791.
I.S., Reynolds, C., Palladino, M.A., 114. Aversa, G., Punnonen, J., de Vries, J.E.
Jr., Goeddel, D.V. The two different The 26-kD transmembrane form of tumor
receptors for tumor necrosis factor necrosis factor alpha on activated CD4+
mediate distinct cellular responses. T cell clones provides a costimulatory
Proc. Natl. Acad. Sci. U. S. A 1991; signal for human B cell activation. J Exp.
88:9292-9296. Med. 1993; 177:1575-1585.
106. Faustman, D., Davis, M. TNF receptor 115. Aspalter, R.M., Eibl, M.M., Wolf, H.M.
2 pathway: drug target for autoimmune Regulation of TCR-mediated T cell
diseases. Nat Rev Drug Discov 2010; activation by TNF-RII. J. Leukoc. Biol.
9:482-493. 2003; 74:572-582.
107. Karin, M., Gallagher, E. TNFR signaling:
116. Scheurich, P., Thoma, B., Ucer, U.,
ubiquitin-conjugated TRAFfic signals
Pfizenmaier, K. Immunoregulatory
control stop-and-go for MAPK signaling
activity of recombinant human tumor
complexes. Immunol Rev 2009; 228:225-
necrosis factor (TNF)-alpha: induction
240.
of TNF receptors on human T cells and
108. van Mierlo, G.J., Scherer, H.U., TNF-alpha- mediated enhancement
Hameetman, M., Morgan, M.E., of T cell responses. J. Immunol. 1987;
Flierman, R., Huizinga, T.W., Toes, 138:1786-1790.
R.E. Cutting edge: TNFR-shedding by
CD4+CD25+ regulatory T cells inhibits 117. Grell, M., Becke, F.M., Wajant, H.,
the induction of inflammatory mediators. Männel, D.N., Scheurich, P. Tumor
J. Immunol. 2008; 180:2747-2751. necrosis factor (TNF) receptor type
2 mediates thymocyte proliferation
109. Pfeffer, K., Matsuyama, T., Kundig, T.M., independently of TNF receptor type 1.
Wakeham, A., Kishihara, K., Shahinian,
Eur. J. Immunol. 1998; 28:257-263.
A., Wiegmann, K., Ohashi, P.S., Kronke,
M., Mak, T.W. Mice deficient for the 118. Tartaglia, L.A., Goeddel, D.V., Reynolds,
55 kd tumor necrosis factor receptor C., Figari, I.S., Weber, R.F., Fendly, B.M.,
are resistant to endotoxic shock, yet Palladino, M.A., Jr. Stimulation of human
succumb to L. monocytogenes infection. T-cell proliferation by specific activation
Cell 1993; 73:457-467. of the 75-kDa tumor necrosis factor
110. Bean, A.G.D., Roach, D.R., Briscoe, H., receptor. J Immunol 1993; 151:4637-
France, M.P., Korner, H., Sedgwick, J.D., 4641.
Britton, W.J. Structural Deficiencies in 119. Zakharova, M., Ziegler, H.K. Paradoxical
Granuloma Formation in TNF Gene- Anti-Inflammatory Actions of TNF-
Targeted Mice Underlie the Heightened alpha: Inhibition of IL-12 and IL-23
Susceptibility to Aerosol Mycobacterium via TNF Receptor 1 in Macrophages
tuberculosis Infection, Which Is Not and Dendritic Cells. J Immunol 2005;
Compensated for by Lymphotoxin. J 175:5024-5033.
Immunol 1999; 162:3504-3511. 120. Hirata, N., Yanagawa, Y., Satoh, M.,
111. Nakano, K., Okugawa, K., Furuichi, H., Ogura, H., Ebihara, T., Noguchi, M.,
Matsui, Y., Sohmura, Y. Augmentation Matsumoto, M., Togashi, H., Seya, T.,
of the generation of cytotoxic T Onoe, K., Iwabuchi, K. Dendritic cell-
lymphocytes against syngeneic tumor derived TNF-alpha is responsible for

29
 CHAPTER 1

development of IL-10-producing CD4(+) comparison of chimeric monoclonal

T cells. Cell Immunol. 2009. antibody to tumour necrosis factor alpha
121. Chen, X., Oppenheim, J.J. Contrasting (cA2) versus placebo in rheumatoid
CHAPTER 1

effects of TNF and anti-TNF on the arthritis. Lancet 1994; 344:1105-1110.

activation of effector T cells and 130. Moreland, L.W., Baumgartner, S.W.,
regulatory T cells in autoimmunity. FEBS Schiff, M.H., Tindall, E.A., Fleischmann,
Lett. 2011. R.M., Weaver, A.L., Ettlinger, R.E.,
122. Chen, X., Bäumel, M., Männel, D.N., Cohen, S., Koopman, W.J., Mohler, K.,
Howard, O.M.Z., Oppenheim, J.J. Widmer, M.B., Blosch, C.M. Treatment of
Interaction of TNF with TNF Receptor rheumatoid arthritis with a recombinant
Type 2 Promotes Expansion and Function human tumor necrosis factor receptor
of Mouse CD4+CD25+ T Regulatory (p75)-Fc fusion protein. N. Engl. J. Med.
Cells. J Immunol 2007; 179:154-161. 1997; 337:141-147.
123. Hamano, R., Huang, J., Yoshimura, 131. van Dullemen, H.M., van Deventer,
T., Oppenheim, J.J., Chen, X. TNF S.J.H., Hommes, D.W., Bijl, H.A., Jansen,
optimally activatives regulatory T cells J., Tytgat, G.N.J., Woody, J. Treatment of
by inducing TNF receptor superfamily Crohn’s disease with anti-tumor necrosis
members TNFR2, 4-1BB and OX40. Eur. factor chimeric monoclonal antibody
J Immunol 2011; 41:2010-2020. (cA2). Gastroenterology 1995; 109:129-
124. Biton, J., Semerano, L., Delavallée, L., 135.
Lemeiter, D., Laborie, M., Grouard- 132. Wong, M., Ziring, D., Korin, Y., Desai, S.,
Vogel, G., Boissier, M.C., Bessis, N. Kim, S., Lin, J., Gjertson, D., Braun, J.,
Interplay between TNF and Regulatory Reed, E., Singh, R.R. TNF alpha blockade
T Cells in a TNF-Driven Murine Model in human diseases: Mechanisms and
of Arthritis. J Immunol 2011; 186:3899- future directions. Clin. Immunol. 2008;
3910. 126:121-136.
125. Nagar, M., Jacob-Hirsch, J., Vernitsky, 133. Taylor, P.C., Peters, A.M., Paleolog, E.,
H., Berkun, Y., Ben-Horin, S., Amariglio, Chapman, P.T., Elliott, M.J., McCloskey,
N., Bank, I., Kloog, Y., Rechavi, G., R., Feldmann, M., Maini, R.N. Reduction
Goldstein, I. TNF activates a NF- of chemokine levels and leukocyte traffic
kappa B-regulated cellular program in to joints by tumor necrosis factor alpha
human CD45RA- regulatory T cells that blockade in patients with rheumatoid
modulates their suppressive function. J. arthritis. Arthr. & Rheum. 2000; 43:38-47.
Immunol. 2010; 184:3570-3581. 134. Charles, P., Elliott, M.J., Davis, D., Potter,
126. Valencia, X., Stephens, G., Goldbach- A., Kalden, J.R., Antoni, C., Breedveld,
Mansky, R., Wilson, M., Shevach, E.M., F.C., Smolen, J.S., Eberl, G., deWoody,
Lipsky, P.E. TNF downmodulates the K., Feldmann, M., Maini, R.N. Regulation
function of human CD4+CD25hi T- of cytokines, cytokine inhibitors, and
regulatory cells. Blood 2006; 108:253- acute-phase proteins following anti-
261. TNF-alpha therapy in rheumatoid
127. van Vollenhoven, R.F. Treatment of arthritis. J. Immunol. 1999; 163:1521-
rheumatoid arthritis: state of the art 1528.
2009. Nat. Rev. Immunol. 2009; 5:531- 135. Ulfgren, A.K., Andersson, U., Engström,
541. M., Klareskog, L., Maini, R.N., Taylor,
128. Maini, R.N., Breedveld, F.C., Kalden, P.C. Systemic anti-tumor necrosis factor
J.R., Smolen, J.S., Davis, D., Macfarlane, alpha therapy in rheumatoid arthritis
J.D., Antoni, C., Leeb, B., Elliott, M.J., down-regulates synovial tumor necrosis
Woody, J.N., Schaible, T.F., Feldmann, factor alpha synthesis. Arthr. & Rheum.
M. Therapeutic efficacy of multiple 2000; 43:2391-2396.
intravenous infusions of anti-tumor 136. Baldwin, H.M., Ito-Ihara, T., Isaacs, J.D.,
necrosis factor alpha monoclonal Hilkens, C.M.U. TNF alpha blockade
antibody combined with low-dose impairs dendritic cell survival and
weekly methotrexate in rheumatoid function in rheumatoid arthritis. Ann.
arthritis. Arthritis Rheum. 1998; 41:1552- Rheum. Dis. 2010; 69:1200-1207.
1563. 137. Nadkarni, S., Mauri, C., Ehrenstein,
129. Elliott, M.J., Maini, R.N., Feldmann, M., M.R. Anti-TNF-alpha therapy induces
Kalden, J.R., Antoni, C., Smolen, J.S., a distinct regulatory T cell population
Leeb, B., Breedveld, F.C., Macfarlane, in patients with rheumatoid arthritis via
J.D., Bijl, H. Randomised double-blind TGF-beta. J Exp. Med. 2007; 204:33-39.

30
 GENERAL INTRODUCTION

138. Ehrenstein, M.R., Evans, J.G., Singh, A., evaluating primary efficacy and safety
Moore, S., Warnes, G., Isenberg, D.A., at twenty-four weeks. Arthritis Rheum.
Mauri, C. Compromised function of 2006; 54:2793-2806.

CHAPTER 1
regulatory T cells in rheumatoid arthritis 145. Kremer, J.M., Westhovens, R., Leon, M.,
and reversal by anti-TNFalpha therapy. Di, G.E., Alten, R., Steinfeld, S., Russell,
J. Exp. Med. 2004; 200:277-285. A., Dougados, M., Emery, P., Nuamah,
139. Maurice, M.M., van der Graaff, W.L., I.F., Williams, G.R., Becker, J.C.,
Leow, A., Breedveld, F.C., van Lier, R.A., Hagerty, D.T., Moreland, L.W. Treatment
Verweij, C.L. Treatment with monoclonal of rheumatoid arthritis by selective
anti-tumor necrosis factor alpha inhibition of T-cell activation with fusion
antibody results in an accumulation of protein CTLA4Ig. N. Engl. J. Med. 2003;
Th1 CD4+ T cells in the peripheral blood 349:1907-1915.
of patients with rheumatoid arthritis. 146. Jiang, Y., Genant, H.K., Watt, I.,
Arthr. & Rheum. 1999; 42:2166-2173. Cobby, M., Bresnihan, B., Aitchison, R.,
140. Schuerwegh, A.J., Van Offel, J.F., McCabe, D. A multicenter, double-blind,
Stevens, W.J., Bridts, C.H., De Clerck, dose-ranging, randomized, placebo-
L.S. Influence of therapy with chimeric controlled study of recombinant human
monoclonal tumour necrosis factor- interleukin-1 receptor antagonist in
alpha antibodies on intracellular patients with rheumatoid arthritis:
cytokine profiles of T lymphocytes radiologic progression and correlation
and monocytes in rheumatoid arthritis of Genant and Larsen scores. Arthritis
patients. Rheumatology. (Oxford) 2003; Rheum. 2000; 43:1001-1009.
42:541-548. 147. Lovell, D.J., Bowyer, S.L., Solinger,
141. Aerts, N.E., De Knop, K.J., Leysen, J., A.M. Interleukin-1 blockade by anakinra
Ebo, D.G., Bridts, C.H., Weyler, J.J., improves clinical symptoms in patients
Stevens, W.J., De Clerck, L.S. Increased with neonatal-onset multisystem
IL-17 production by peripheral T helper inflammatory disease. Arthritis Rheum.
cells after tumour necrosis factor 2005; 52:1283-1286.
blockade in rheumatoid arthritis is 148. Miller, S.A., St, O.E. Otelixizumab: a
accompanied by inhibition of migration- novel agent for the prevention of type
associated chemokine receptor 1 diabetes mellitus. Expert. Opin. Biol.
expression. Rheumatology. (Oxford) Ther. 2011; 11:1525-1532.
2010; 49:2264-2272. 149. Wolbink, G.J., Aarden, L.A., Dijkmans,
142. Goldstein, I., Ben-Horin, S., Koltakov, A., B.A. Dealing with immunogenicity of
Chermoshnuk, H., Polevoy, V., Berkun, biologicals: assessment and clinical
Y., Amariglio, N., Bank, I. Alpha1beta1 relevance. Curr. Opin. Rheumatol. 2009;
integrin+ and regulatory Foxp3+ 21:211-215.
T cells constitute two functionally distinct 150. Gregori, S., Casorati, M., Amuchastegui,
human CD4+ T cell subsets oppositely S., Smiroldo, S., Davalli, A.M., Adorini,
modulated by TNFalpha blockade. J. L. Regulatory T cells induced by 1
Immunol. 2007; 178:201-210. alpha,25-dihydroxyvitamin D3 and
143. Maini, R.N., Taylor, P.C., Szechinski, J., mycophenolate mofetil treatment
Pavelka, K., Broll, J., Balint, G., Emery, mediate transplantation tolerance. J.
P., Raemen, F., Petersen, J., Smolen, Immunol. 2001; 167:1945-1953.
J., Thomson, D., Kishimoto, T. Double- 151. Sakaguchi, S., Powrie, F., Ransohoff,
blind randomized controlled clinical trial R.M. Re-establishing immunological
of the interleukin-6 receptor antagonist, self-tolerance in autoimmune disease.
tocilizumab, in European patients Nat. Med. 2012; 18:54-58.
with rheumatoid arthritis who had an
incomplete response to methotrexate. 152. Roncarolo, M.G., Battaglia, M.
Arthritis Rheum. 2006; 54:2817-2829. Regulatory T-cell immunotherapy
for tolerance to self antigens and
144. Cohen, S.B., Emery, P., Greenwald, M.W., alloantigens in humans. Nat. Rev.
Dougados, M., Furie, R.A., Genovese, Immunol. 2007; 7:585-598.
M.C., Keystone, E.C., Loveless, J.E.,
Burmester, G.R., Cravets, M.W., Hessey, 153. Shevach, E.M. Mechanisms of Foxp3+
E.W., Shaw, T., Totoritis, M.C. Rituximab T regulatory cell-mediated suppression.
for rheumatoid arthritis refractory to anti- Immunity 2009; 30:636-645.
tumor necrosis factor therapy: Results 154. Vignali, D.A.A., Collison, L.W., Workman,
of a multicenter, randomized, double- C.J. How regulatory T cells work. Nat.
blind, placebo-controlled, phase III trial Rev. Immunol. 2008; 8:523-532.

31
 CHAPTER 1

155. Thomson, A.W., Robbins, P.D. Tolerogenic 160. Chorny, A., Gonzalez-Rey, E., Fernandez-
dendritic cells for autoimmune disease Martin, A., Pozo, D., Ganea, D., Delgado,
and transplantation. Ann. Rheum. Dis. M. Vasoactive intestinal peptide
CHAPTER 1

2008; 67:90-96. induces regulatory dendritic cells with

156. Horibe, E.K., Sacks, J., Unadkat, J., therapeutic effects on autoimmune
Raimondi, G., Wang, Z., Ikeguchi, R., disorders. Proc. Natl. Acad. Sci. U. S. A.
Marsteller, D., Ferreira, L.M., Thomson, 2005; 102:13562-13567.
A.W., Lee, W.P., Feili-Hariri, M. 161. Giannoukakis, N., Phillips, B., Finegold,
Rapamycin-conditioned, alloantigen- D., Harnaha, J., Trucco, M. Phase I
pulsed dendritic cells promote indefinite
(safety) study of autologous tolerogenic
survival of vascularized skin allografts
dendritic cells in type 1 diabetic patients.
in association with T regulatory cell
expansion. Transpl. Immunol. 2008; Diabetes Care 2011; Epub ahead of
18:307-318. print.
157. Bonham, C.A., Peng, L., Liang, X., Chen, 162. Zahorchak, A.F., Kean, L.S., Tokita,
Z., Wang, L., Ma, L., Hackstein, H., D., Turnquist, H.R., Abe, M., Finke,
Robbins, P.D., Thomson, A.W., Fung, J., Hamby, K., Rigby, M.R., Larsen,
J.J., Qian, S., Lu, L. Marked prolongation C.P., Thomson, A.W. Infusion of
of cardiac allograft survival by dendritic stably immature monocyte-derived
cells genetically engineered with NF- dendritic cells plus CTLA4Ig modulates
kappa B oligodeoxyribonucleotide alloimmune reactivity in rhesus
decoys and adenoviral vectors encoding macaques. Transplantation 2007;
CTLA4-Ig. J. Immunol. 2002; 169:3382- 84:196-206.
3391. 163. Adorini, L., Giarratana, N., Penna, G.
158. Popov, I., Li, M., Zheng, X., San, H., Pharmacological induction of tolerogenic
Zhang, X., Ichim, T., Suzuki, M., Feng, B., dendritic cells and regulatory T cells.
Vladau, C., Zhong, R., Garcia, B., Strejan, Semin. Immunol. 2004; 16:127-134.
G., Inman, R., Min, W.P. Preventing
autoimmune arthritis using antigen- 164. Hackstein, H., Thomson, A.W. Dendritic
specific immature dendritic cells: a novel cells: emerging pharmacological targets
tolerogenic vaccine. Arthritis Res. Ther. of immunosuppressive drugs. Nat. Rev.
2006; 8:R141. Immunol. 2004; 4:24-35.
159. Phillips, B.E., Giannoukakis, N., Trucco, 165. Turnquist, H.R., Thomson, A.W. Taming
M. Dendritic cell mediated therapy for the lions: manipulating dendritic cells
immunoregulation of type 1 diabetes for use as negative cellular vaccines in
mellitus. Pediatr. Endocrinol. Rev. 2008; organ transplantation. Curr. Opin. Organ
5:873-879. Transplant. 2008; 13:350-357.

32
 2
CHAPTER
An optimized CFSE-based T-cell
suppression assay to evaluate
the suppressive capacity of
regulatory T-cells induced by
human tolerogenic dendritic cells
Martine A. Boks, Jaap Jan Zwaginga,
S. Marieke van Ham and Anja ten Brinke

Scand. J. Immunol. 2010 Aug;72(2):158-68

 CHAPTER 2

Abstract
In autoimmune diseases or transplant graft rejection a therapy that will prevent or
reduce the present immune activation is highly desired. Ex vivo generated tolerogenic
dendritic cells (DCs) are considered to have a strong potential as cellular therapy
for these diseases. One of the mechanisms of immune suppression mediated by
tolerogenic DCs is the induction of regulatory T cells (Tregs). Consequently, the
efficacy of such DCs to induce Tregs will reflect their tolerogenic capacity. Because
no specific markers have been described for human induced (i)Tregs yet, the Tregs
can only be appreciated by functionality. Therefore, we have optimized an in vitro
CHAPTER 2

suppression assay to screen for human DC-induced-Treg activity. IL-10-generated

tolerogenic DCs were used to induce Tregs that were previously shown to effectively
suppress the proliferation of responder T cells stimulated with allogeneic mature DCs
(mDCs). Our results show that the suppressive capacity of IL-10 DC-induced Tregs
measured in the suppression assay increases with the iTreg dose and decreases with
higher numbers of antigen presenting cells (APCs) as T cell stimulation. Lowering
the ratio between responder T cells and stimulator mDCs present in the coculture
clearly improved the read-out of the suppression assay. Furthermore, mDC-primed
T cells in the suppression assay were shown to be an essential control condition. In
conclusion, we recommend titrations of both APCs and iTregs in the suppression
assay and to include a negative control condition with T cells primed by mDCs, to
distinguish specific and functional suppression by iTregs from possible generalised
suppressive activity.

36
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

Introduction
In many diseases, like in autoimmune diseases or transplant graft rejection, the
immune system has shifted towards an undesired status of immune activation.
Therapies that prevent induction of immune activation or reduce immune activation
are explored to combat disease or graft rejection and to increase the patient’s
wellbeing. DCs are highly specialized professional antigen presenting cells. In
recent years it has become clear that DCs not only induce immunogenic responses,
but can also initiate tolerogenic responses.1-3 In this respect, DCs play an important
role for the maintenance of immune homeostasis. The tolerogenic properties

CHAPTER 2
of DCs can be mediated via soluble factors, like cytokines, via cell contact-
dependent mechanisms, like inhibitory receptor-ligand interactions, or via impaired
costimulation. Tolerogenic DCs induce anergy or apoptosis of effector T cells and/
or expand or activate regulatory T cells.2,4,5 For these reasons, ex vivo generated
tolerogenic DCs are considered as potential cellular vaccines to treat autoimmune
disorders or to prevent undesired immune responses against allogeneic transplants.
Human tolerogenic DCs can be cultured in vitro from DC precursors using
different compounds. Often used for this purpose are the anti-inflammatory
cytokine IL-10, the active form of vitamin D­3 or analogues of vitamin D3, and anti-
inflammatory/ immunosuppressive drugs like corticosteroids, rapamycin, calcineurin
inhibitors and aspirin.6-8
For DCs that are generated for tolerance inducing therapy, their capacity to
induce Tregs is preferred more than induction of anergy or apoptosis of effector
T cells. The reason for this is that DC-induced Tregs can in turn induce a more
widespread tolerance via secretion of IL-10, TGFβ and/ or granzymes and perforin,
via expression of inhibitory molecules and via competition for growth factors.9-11 In
addition, Tregs may influence the memory response of T cells.12 A read-out system
for the reliable measurement of Treg induction by tolerogenic DCs is therefore of
great importance. In the murine system, Tregs can be identified by a combination
of markers, e.g. CD4, CD25 and FOXP3. In the human system several markers have
been described for naturally occurring Tregs, e.g. CD25hi, CD127lo, FOXP3, cytotoxic
T lymphocyte antigen (CTLA)-4 and GITR.9,10,13 CD4+CD25+ iTregs are characterized
by the same markers and resemble the naturally occurring Tregs, but are generated
in the periphery in an antigen-specific manner.14 None of these markers however, are
specific for human Tregs, as they are also expressed by activated T cells. Different
types of iTregs have been described of which T regulatory type 1 cells (Tr1) and T
helper 3 cells (Th3) are the best studied. These iTregs arise after antigen-specific
stimulation, in the presence of IL-10 or TGFβ, respectively. They are characterized
by production of IL-10 and/ or TGFβ and regulate immune responses through the
secretion of these cytokines.9,15-18 A unique phenotype that defines all the different
types of iTregs has not yet been identified. Therefore, definitive characterization of
Treg induction by tolerogenic DCs still requires an in vitro suppression assay as a
functional read-out.

37
 CHAPTER 2

Various in vitro suppression assays have been described but the setup of these
assays is often variable and poorly defined. Therefore, we set out to develop a
robust and clearly defined suppression assay. Often the assay for analysis of Treg
suppressive capacity consists of a coculture setup of a constant number of responder
T cells, which are antigen-specific or polyclonal stimulated, and a titration of Tregs.
Responder T cell proliferation is measured by [3H]-thymidine incorporation or
CFSE (5,6-carboxyfluorescein diacetate succinimidyl ester) dilution of proliferated
cells.19-21 The information obtained from [3H]-thymidine assays, however, is limited
as the proliferation of cells is only determined during the final 16 hours of culture.
CFSE dilution as a read-out mechanism allows longer monitoring, while the
CHAPTER 2

expression of surface molecules or cytokine production of proliferated cells can

be simultaneously determined.20,21 Therefore, in this study, we have optimized an
assay using the fluorescent dye CFSE. Furthermore, we have chosen to use APCs
in stead of a polyclonal stimulation in our setup to mimic the in vivo situation more
closely as Tregs can suppress T cell responses both in a direct way or an indirect
way via APCs.10,11 This APC-mediated indirect way would be missed in a suppression
assay with a polyclonal stimulus. As tolerogenic DCs to induce Tregs we used IL-10-
treated DCs.
During optimizing the suppression assay for human tolerogenic DC-induced
Tregs, we found that the number of APCs for T cell stimulation in a suppression assay
is important for optimal read-out of suppression by DC-induced Tregs. Furthermore,
we show the importance of including T cells primed by mDCs in suppression assays
as a negative control condition, because it is essential to discriminate specific from
aspecific suppression by DC-primed T cells. Longer T cell priming results in less
specificity of the assay due to aspecific suppression in the negative control condition.

Materials and methods

Reagents and antibodies
CellGro DC medium, IL-4, IL-1β and TNFα were all obtained from CellGenix
(Freiburg, Germany). GM-CSF was either obtained from CellGenix or from Berlex
(Leukine) (Seattle, USA). PGE2 was obtained from Sigma-Aldrich (Steinheim,
Germany). IL-10 was obtained from PeproTech (Rock Hill, USA). For read-out assays
Iscove’s Modified Dulbecco’s Medium (IMDM, Bio Whittaker, Verviers, Belgium)
was used supplemented with 10% FCS (Bodinco, Alkmaar, The Netherlands),
20 µg/ml human transferrin, 50 µM 2-mercaptoethanol (Sigma-Aldrich), 100 U/ml
penicillin and 100 mg/ml streptomycin (Gibco, Merelbeke, Belgium). All monoclonal
antibodies (mAbs) used for flow cytometry were obtained from Becton Dickinson
(BD Biosciences, San Jose, USA).

Isolation of monocytes from healthy donors

Peripheral blood mononuclear cells (PBMCs) were isolated via separation over a
Lymphoprep gradient (d = 1.077 kg/L, Axis-Shield PoC AS, Oslo, Sweden) from
fresh aphaeresis material (Sanquin Blood Bank North West, Amsterdam, The

38
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

Netherlands), obtained from healthy volunteers upon informed consent. Monocytes

were isolated from the PBMC fraction by positive selection using CD14 microbeads
and a magnetic cell separator (MACS, Miltenyi Biotec, Bergisch Gladbach, Germany).
Alternatively, monocytes were isolated from fresh aphaeresis material by separation
using the ElutraTM cell separation system (Gambro, Lakewood, USA).

Culture of monocyte-derived dendritic cells

Monocytes were cultured at a concentration of 0.5 x 106 cells/ml in a 6 well plate
(Nunc, Roskilde, Denmark) in CellGro serum-free medium supplemented with IL-4
(800 IU/ml) and GM-CSF (1000 IU/ml). At day 6, the immature DCs (imDCs) were either

CHAPTER 2
left untreated, or matured with IL-1β (10 ng/ml), TNFα (10 ng/ml) and PGE2 (1 µg/ml)
(mature (m)DCs), or alternatively activated by a 1 hour pre-incubation of IL-10 (40 ng/ml)
followed by the maturation cocktail (IL-1β, TNFα and PGE2) (IL-10 DCs). After 2 days of
maturation, the cells were harvested, extensively washed and analysed.

Flow cytometric analysis

For phenotyping, the DCs were washed with PBS containing 0.5% bovine serum
albumin (PBA) and incubated with 50 µl mAb or appropriate isotype controls diluted
in PBA with 3 mg/ml human gamma globulin for 30 minutes in the dark at 4°C. Cells
were washed and resuspended in PBA. DAPI (Sigma-Aldrich) was added to the cells
before analysis to assess cell viability and exclude dead cells from analysis. Cells
were analysed on an LSRII flow cytometer (BD Biosciences) and analysed with FACS
Diva software (BD Biosciences).

Cytokine production of dendritic cells

DCs were harvested, washed and cultured in 96 well flat bottom plates (Nunc) at
a concentration of 1 x 104 cells/ well in culture medium. To mimic the interaction
with CD40 ligand (CD40L)-expressing T helper cells, irradiated CD40L-transfected
J558 cells were added at a concentration of 5 x 104 cells/ well. After 24 hours
of stimulation, the supernatant was harvested. The production level of IL-6, IL-10,
IL-12p70 and TNFα was determined by ELISA. For the detection of IL-6, IL-10 and
TNFα the PeliKine-compact ELISA kit was used (Sanquin Reagents, Amsterdam, The
Netherlands). For the detection of IL-12p70 a combination of BT-21 mAb (Diaclone,
Besançon, France), recognizing the p35 subunit, and C8.6 mAb (BD Biosciences),
recognizing the p40 subunit, was used in an ELISA.

Isolation of naïve and memory CD4+ T cells

Naïve CD4+CD45RA+CD45RO- T cells were purified from PBMCs or from the T cell
containing fraction of ElutraTM using CD4 T cell isolation kit II (Miltenyi Biotec),
together with mAb CD45RO-PE and anti-PE beads (Miltenyi Biotec). Naïve T cells
were used as responder T cells in an MLR and for priming to evaluate suppressive
capacity. Memory CD4+CD45RA-CD45RO+ T cells were used as responder T cells in
an MLR and in the suppression assay. Purities of T cell subsets were routinely around
92% for naïve and memory T cells.

39
 CHAPTER 2

Mixed lymphocyte reaction (MLR)

DCs were used to stimulate allogeneic memory or naïve CD4+ T cells. Varying numbers
of DCs were co-cultured in 96 well flat bottom plates (Nunc) with 5 x 104 T cells/
well in culture medium. After 5 days of culture, 0.2 µCi [3H]-thymidine (Amersham
Biosciences, Amersham, UK) was added to each well and the incorporation of
radioactivity was measured after 16 hours using a beta liquid scintillation counter
(1205 Betaplate Wallac).

Priming of T cells and cytokine production

CHAPTER 2

1 x 106 allogeneic naïve CD4+ T cells were primed in the presence of 1 x 105 DCs in
24 well plates (Nunc) for 6 days. Alternatively, naïve T cells were primed by DCs for 13
days with addition of 10 U/ml IL-2 (PeproTech) and fresh medium on day 7. DC-primed
T cells were harvested, washed and evaluated for their suppressive capacity.
Cytokines produced during the priming of T cells were measured by harvesting
the supernatant of the DC and T cell coculture after 6 days of culture. Alternatively,
5 x 104 DC-primed T cells were restimulated in 96 well flat bottom plates (Nunc) with
PMA (10 ng/ml; Sigma-Aldrich) and ionomycin (1 μg/ml; Sigma-Aldrich), or with 4
x 103 mDCs in culture medium. After 24 hours of stimulation, the supernatant was
harvested. The production of IL-10, IL-13 and IFNγ was determined by ELISA using
the PeliKine-compact ELISA kit (Sanquin Reagents).

Suppression assay
DC-primed T cells were added to a second culture consisting of responder T cells
stimulated with mDCs. Responder T cells are CD4+ memory T cells of the same
T cell donor and mDCs are from the same DC donor. To track proliferation of
responder T cells in the suppression assay, cells were labelled with 0.5 μM of the
‘green’ fluorescent dye CFSE (Invitrogen, Eugene, Oregon, USA) for 15 minutes
at RT. To discriminate between proliferated, CFSE negative, responder T cells and
non-CFSE labelled primed T cells, the primed T cells were labelled with 6 μM of
the ‘red’ fluorescent dye PKH26 (Sigma-Aldrich) for 5 minutes at RT. Primed T cells
were irradiated (30 Gy) after labelling to prevent convergence of proliferated CFSE-
and PKH26-negative cells. After labelling with the tracking dyes, 5 x 104 responder
T cells were incubated with 5 x 104 primed T cells (1:1), or at ratio’s 1:2 and 2:1.
4 x 103 mDCs were added as stimulation (1:12.5 ratio with the responder T cells),
or in varying amounts as described. Cells were cultured in 96 well round bottom
plates (Greiner Bio-One, Frickenhausen, Germany) for 6 days in culture medium.
Prior to flow cytometry, cells were stained with mAb against CD4 and DAPI. The
lymphocyte gate was established on basis of forward vs. sideward scatter plot,
including blasting cells. DAPI+ dead cells and PKH26+ cells were excluded from
analysis. Proliferation of CD4+ responder T cells was measured on an LSRII flow
cytometer (BD Biosciences) and analysed with FACS Diva software (BD Biosciences).
For blocking experiments neutralizing Abs were used at 10 μg/ml against: TGFβ
(R&D Systems, Minneapolis, USA), PD-1 (R&D Systems), IL-10 (anti-IL-10.8, Sanquin
Reagents) and CTLA-4 (Ancell, Bayport, USA). For isotype controls an irrelevant IgG1

40
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

Ab directed to FelD1 or a goat-anti-mouse Ab (Sanquin Reagents) were used. For

transwell experiments, 96-well transwell plates were used (Corning, New York, USA)
with a 0.4 μm pore size polycarbonate membrane. CFSE-labelled responder T cells
and mDCs were cultured in the lower compartment. PKH26-labelled primed T cells
were added to the upper compartment, with or without mDCs as stimulation. As a
control, PKH26-labelled primed T cells were added directly to responder T cells and
mDCs in the lower compartment.

Statistical analysis
Data are expressed as mean + SEM. The statistical significance of the data was

CHAPTER 2
analysed using a paired Student’s t test in Graphpad Prism 4.03 software.

Results
Effect of IL-10 on DC phenotype and cytokine production
Previously is shown that incubation of human monocyte-derived DCs with IL-10
promotes the generation of tolerogenic DCs which are able to induce Tregs.22-28
To generate DCs with a tolerogenic phenotype which meet the requirements
for clinical application, we generated monocyte-derived immature (im)DCs
from elutriated monocytes which were subsequently incubated with IL-10 and a
maturation cocktail, consisting of IL-1β, TNFα and PGE2 (IL-10 DCs) under serum-
free conditions. Alternatively, imDCs were treated with the maturation cocktail
only to generate immuno-activatory mature (m)DCs, or were left untreated to
keep them in their immature state. Tolerogenic DCs are characterized by low
to intermediate expression of cell surface MHC molecules and costimulatory
molecules, low production of pro-inflammatory cytokines and high production of
IL-10.4 To determine the maturation status of the DCs, we performed flow cytometric
analysis on typical DC markers (Figure 1A). IL-10 DCs clearly showed intermediate
expression of the DC maturation marker CD83, the costimulatory molecules CD80
and CD86, and the MHC class II molecule HLA-DR, compared to imDCs and mDCs.
To determine cytokine production, we stimulated the DCs with CD40L-transfected
cells to mimic the interaction with CD40L-expressing T helper cells (Figure 1B). IL-10
DCs showed a significantly higher production of IL-10 and TNFα than mDCs. The
ratio of IL-10:TNFα production was significantly higher for IL-10 DCs than for mDCs.
The IL-6 levels were not significantly different between the two conditions. For both
DC types IL-12p70 production was not detectable. These results show that, apart
from the TNFα production, the IL-10 DCs have the characteristics of semi-mature
tolerogenic DCs.

Effect of IL-10 DCs on T cells

It has been reported that IL-10 DCs have a reduced capacity to stimulate allogeneic
T cells in an MLR 22-24,26 and that the T cells rendered show a low production of IFNγ
and a high production of IL-10.23,24 To determine the effect of IL-10 DCs on the
proliferation of allogeneic T cells, DCs were cocultured with memory or naïve CD4+

41
 CHAPTER 2

A
100
mDC
IL-10 DC
% positive cells

75 imDC

50

25

0
CD80 HLA-DR CD83 CD40 CD86

Figure 1. Phenotype and cytokine

MFI±SEM CD80 HLA-DR CD83 CD40 CD86
production comparing IL-10 and
normal matured DCs with immature
DCs. ImDCs were matured with IL-10
mDC 751±88 31.367±20.199 2.130±632 416±61 3.079±1.502
CHAPTER 2

IL-10 DC 507±89 23.953±14.449 1.100±295 455±84 1.720±762 and the maturation cocktail (IL-10
imDC 157±23 22.166±13.182 86±51 239±54 403±118 DCs), with the maturation cocktail
only (mDCs) or cultivated without
maturation stimuli (imDCs). After
B IL-10 TNFα
α 2 days, the DCs were harvested
P<0.05 and washed. (A) DCs were stained
3 4 P<0.05 for different surface molecules
3
with fluorescently conjugated
2 mAbs. The percentage of positive
ng/ml
ng/ml

2
cells (average + SEM; graph) and
1
1 mean fluorescence intensity (MFI)
0 0
(average ± SEM; table) are derived
IL-10 DC mDC IL-10 DC mDC from 3 independent experiments.
(B) DCs were incubated with J558
IL-6 Ratio IL-10:TNFα
α cells expressing CD40L. After 24
P<0.05 hours of culture the supernatant
10.0 1.5 was harvested and the amount of
IL-10, TNFα and IL-6 produced was
IL-10/TNFα
α

7.5
1.0 determined by ELISA. Cytokine
ng/ml

5.0
production (mean + SEM) and the
0.5
2.5 ratio IL-10:TNFα production of 11
0.0 0.0 independent experiments is shown.
IL-10 DC mDC mDC IL-10 DC

T cells for 6 days and T cell proliferation was measured (Figure 2A). mDCs strongly
stimulated proliferation of CD4+ memory and naïve T cells. IL-10 DCs showed a
reduced stimulatory capacity compared to mDCs, although it was superior to that
of imDCs. We measured the cytokine production of naïve T cells after 6 days of
coculture with DCs (Figure 2B). Coculture of T cells with IL-10 DCs resulted in a
significant decrease in IFNγ and IL-13 production and a significant increase in
IL-10 production compared to the coculture of T cells with mDCs. Since IL-10 DCs
produce IL-10 themselves (Figure 1B), it can not be excluded that the observed
IL-10 production is DC-derived. To demonstrate enhanced IL-10 production by the
T cells cocultured with IL-10 DCs, we harvested the T cells after 6 days of coculture
and measured cytokine production after 24 hours restimulation (Figure 2C). T cells
previously cocultured with IL-10 DCs showed a significant higher IL-10 production
upon restimulation with PMA and ionomycin than the T cells primed by mDCs. IL-10
DC-primed T cells restimulated with mDCs also showed this tendency, although
it was not significant. Overall, these results show that the IL-10 DCs generated
in these experiments have a reduced capacity to stimulate allogeneic T cells and

42
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

A Allogeneic CD45RO +CD4+ memory T cells Allogeneic CD45RA +CD4+ naïve T cells
12000 20000
mDC mDC
[3H]-thymidine incorporation (CPM)

[3H]-thymidine incorporation (CPM)

IL-10 DC 17500
10000 IL-10 DC
imDC
15000
8000
12500

6000 10000

7500
4000
5000
2000
2500

0 0

CHAPTER 2
1:12.5 1:25 1:50 1:100 1:200 1:400 0:1 1:12.5 1:25 1:50 1:100 1:200 1:400 0:1
DC:T ratio DC:T ratio

B IFNγγ IL-13 IL-10

C IL-10
P<0.05 100 P<0.05
P<0.05 P<0.05
1.5 1.2 40
75
0.9 30

pg/ml
1.0
ng/ml

pg/ml
ng/ml

0.6 20 50

0.5
0.3 10 25

0.0 0.0 0 0
Tpr (IL-10 DC) Tpr (mDC) Tpr (IL-10 DC) Tpr (mDC) Tpr (IL-10 DC) Tpr (mDC) Tpr (IL-10 DC) Tpr (mDC)

Figure 2. Lymphocyte proliferation and cytokine production in allogeneic MLR. After 2

days of maturation, DCs were harvested and washed. (A) DCs were subsequently incubated
in graded doses with allogeneic CD4+CD45RA-CD45RO+ memory T cells or allogeneic
CD4+CD45RA+CD45RO- naïve T cells. Proliferation was measured on day 6 after 16 hours of
[3H]-thymidine incorporation. Results are shown as mean + SEM from triplicate assays in 2
(memory T cells) and 5 (naïve T cells) independent experiments. (B) DCs were incubated in a
1:10 ratio with allogeneic CD4+ naïve T cells. The supernatant was harvested after 6 days of
culture and the amount of IFNγ, IL-13 and IL-10 produced was determined by ELISA. Mean
cytokine production + SEM of 14 independent experiments is shown. (C) CD4+ naïve T cells
were cocultured with DCs as described in (B). After 6 days the T cells were harvested, washed
and restimulated with PMA and ionomycin or with mDCs. The supernatant was harvested
after 24 hours of culture and the amount of IL-10 was determined by ELISA. Mean cytokine
production + SEM of 14 (PMA/ ionomycin) or 7 (mDCs) independent experiments is shown.
Tpr: DC-primed T cells.

that these T cells produce less IFNγ and IL-13, and a higher amount of IL-10 upon
coculture with IL-10 DCs.

IL-10 DC-primed T cells suppress proliferation of responder T cells

One characteristic of tolerogenic DCs is their capacity to expand or activate
regulatory T cells.2,4,5 However, as specific human Treg markers are lacking, human
Tregs need to be assayed by their functionality. Most preferably this involves Treg
mediated suppression of the proliferation of activated T cells. Therefore we have
set up a suppression assay to test the functionality of human DC-induced Tregs. We
have generated potential Tregs by priming naïve T cells for 6 days with IL-10 DCs.
Primed T cells were cocultured with responder T cells and mDCs as stimulation. In
the control situation naïve T cells were primed with the immuno-activatory mDCs
as these are known to primarily induce effector T cells. Responder T cells were
labelled with CFSE, a fluorogenic compound to detect cell division. To distinguish

43
 CHAPTER 2

primed T cells from proliferated, CFSE-negative responder T cells, primed T cells

were labelled with PKH26 (see gating strategy in Figure 3A). Addition of IL-10
DC-primed T cells to the suppression assay caused a trend for suppression of the
proliferation of responder T cells (Figure 3A and B). Compared to proliferation of
mDC-stimulated responder T cells only, this however was not significant for the
cell titrations used. The introduced control condition of mDC-primed T cells did
not suppress the proliferation of responder T cells, as expected. Comparison of
IL-10 DC-primed T cells with the mDC-primed T cells, however, did show significant
suppression of proliferation of responder T cells. Thus, priming of naïve T cells with
IL-10 DCs induces Tregs with significant suppression capacity when compared to
CHAPTER 2

the negative control condition of mDC-primed T cells.

A Tresp + mDC + Tpr (IL-10 DC)

49,9%
PKH26

Count

CFSE
Figure 3. IL-10 DC-primed
Tresp Tresp Tresp + mDC
naïve T cells suppress mDC-
+ mDC + Tpr (mDC)
stimulated proliferation
of responder T cells when
1% 53,6% 75%
compared to the mDC-
primed T cells. Naïve CD4+
Count

T cells were primed by DCs

(Tpr) for 6 days. DC-primed
T cells were harvested,
washed and labelled with
CFSE PKH26. CFSE-labelled
responder CD4+ memory
T cells (Tresp) were added in a
B 1:1 ratio together with mDCs
as stimulation (Tresp:mDC =
12.5:1). (A) Gating strategy
% proliferating CD4+ Tresp

80 P<0.0005 for read-out of suppression

assays. Viable lymphocyte
gates were based on FSC
60 vs. SSC plot, including
blasting cells. DAPI+ cells
were excluded from analysis
40 (data not shown). Next,
CFSE+ cells were gated and
20 analysed for proliferation. One
representative experiment is
shown. (B) Mean proliferation
0 of responder T cells +
- +Tpr (IL-10 DC) +Tpr (mDC) SEM from 6 independent
suppression experiments is
Tresp + mDC shown.

44
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

Suppression by IL-10 DC-induced Tregs is dependent on the

amount of APCs in the suppression assay
Our results show that responder T cell proliferation is significantly suppressed by IL-10
DC-primed T cells, but only when compared to the negative control condition with
mDC-primed T cells and not when compared to the proliferation of mDC-stimulated
responder T cells. As the suppression induced by the IL-10 DC-primed T cells in
our assay was disappointing compared to their suppressive capacity described in
literature 24,25, we optimized the suppression assay. Oberg and co-workers 29 found
by using artificial APCs, in the form of anti-CD3/ anti-CD28 coated beads that fewer

CHAPTER 2
beads resulted in more suppression by natural Tregs. Therefore, we investigated
whether we could improve suppression by adjusting the number of APCs in the
suppression assay (Figure 4). With lower numbers of mDCs as stimulation, IL-10
DC-induced Tregs showed more obvious suppression of proliferation of responder
T cells. IL-10 DC-induced Tregs suppressed proliferation of responder T cells up to
44% with the lowest amount of mDCs compared to proliferation of mDC-stimulated
responder T cells. These data show that the suppressive capacity of IL-10 DC-
induced Tregs is sensitive to the number of APCs that provide T cell stimulation in
the suppression assay.

Suppression by induced Tregs is dose-dependent

In addition to the number of APCs, we also investigated the dose-dependency
of iTregs on the level of suppression in the suppression assay (Figure 5A). The
degree of suppression of responder T cell proliferation by IL-10 DC-induced Tregs
is indeed dose-dependent, whereas the control condition of mDC-primed T cells
did not show suppression of proliferation of responder T cells. IL-10 DC-induced
Tregs suppressed the proliferation of responder T cells up to 64% with the highest
amount of Tregs, compared to proliferation of mDC-stimulated responder T cells.

12.5 Tresp : 1 mDC 25 Tresp : 1 mDC 50 Tresp : 1 mDC

% proliferating CD4+ Tresp

% proliferating CD4+ Tresp

% proliferating CD4+ Tresp

60 P<0.05 60 P<0.005
80 P<0.0005

60
40 40

40
20 20
20

0 0 0
- +Tpr (IL-10 DC) +Tpr (mDC) - +Tpr (IL-10 DC) +Tpr (mDC) - +Tpr (IL-10 DC) +Tpr (mDC)

Tresp + mDC Tresp + mDC Tresp + mDC

Figure 4. IL-10 DC-induced Tregs suppress the mDC-stimulated responder T cell

proliferation more effectively in conditions with lower numbers of mDCs. Naïve CD4+
T cells were primed by DCs for 6 days. DC-primed T cells were harvested, washed and PKH26
labelled. CFSE-labelled responder CD4+ memory T cells were mixed with PKH26-labelled
primed T cells in a 1:1 ratio together with varying amounts of mDCs as stimulation (Tresp:mDC
= 12.5:1, 25:1, 50:1). Mean proliferation of responder T cells + SEM from 6 independent
suppression experiments is shown.

45
 CHAPTER 2

A 6 days priming B 13 days priming

P<0.01 P<0.005 P<0.0005
% proliferating CD4+ Tresp

% proliferating CD4+ Tresp

60 50
P<0.05 Tresp + mDC
40 + Tpr (IL-10 DC)
40 + Tpr (mDC)
30

20
20
10

0 0
1:2 1:1 2:1 1:2 1:1 2:1
CHAPTER 2

Tresp:Tpr ratio Tresp:Tpr ratio

Figure 5. The suppression capacity of IL-10 DC-induced Tregs is dose-dependent. Naïve

CD4+ T cells were primed by DCs for 6 days (A) or for 13 days with addition of fresh medium
and IL-2 (10 U/ml) on day 7 (B). DC-primed T cells were harvested, washed and PKH26 labelled.
PKH26-labelled primed T cells were added in graded doses to CFSE-labelled responder CD4+
memory T cells and mDCs (Tresp:mDC = 50:1). Mean proliferation of responder T cells + SEM
from 6 (A) and 3 (B) independent suppression experiments is shown.

These data show that the degree of suppression of responder T cell proliferation is
dependent on the concentration of iTregs in the suppression assay.
Finally, we investigated whether priming of naïve T cells with IL-10 DCs for a
longer time period would induce Tregs with a higher suppressive capacity. We
primed naïve T cells with DCs for 13 days and performed a suppression assay
under optimal conditions of low mDC stimulation and included a titration of primed
T cells (Figure 5B). Control T cells primed with mDCs for 13 days showed a trend
to suppress the proliferation of responder T cells. As a resultant, there was no
significant difference in suppression of proliferation of responder T cells between
the two conditions of primed T cells. These data show that a longer priming period
results in induction of generalised suppression in the negative control condition and
does not result in improved read-out of the suppression assay.

Suppression by induced Tregs is dependent on cell contact

Naturally occurring Tregs are described to suppress proliferation of responder T cells
via cell contact-dependent mechanisms and independent of secreted cytokines. This
in contrast to the induced Tregs Tr1 and Th3, which are typically characterized by
predominantly utilizing IL-10 and TGFβ in their respective mechanism of action.15,16
One report however, shows that IL-10 DC-induced Tregs mediate suppression in a
fashion that is cell contact-dependent between Tregs and responder T cells and not
dependent on IL-10 or TGFβ secretion.25 Here we investigated the mechanism behind
the suppressive capacity of the IL-10 DC-induced Tregs generated in this study. We
first set out to examine whether the IL-10 DC-induced Tregs mediate suppression
via cytokine secretion (Figure 6A). A neutralizing IL-10 antibody did not inhibit
the suppression of proliferation of responder T cells by IL-10 DC-induced Tregs.
Also a neutralizing TGFβ antibody did not inhibit the suppression of proliferation
of responder T cells by IL-10 DC-induced Tregs. The neutralizing antibodies were

46
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

tested for functionality and both are able to inhibit the respective cytokine (data not
shown). Next, we investigated whether cell contact is important for suppression by
IL-10 DC-induced Tregs by use of a transwell system (Figure 6B). Separating IL-10
DC-induced Tregs in the upper compartment from responder T cells and mDCs
in the lower compartment completely prevented suppression of responder T cell
proliferation. This was not due to the fact that the iTregs were not stimulated in this
experimental setup, also upon stimulation of the iTregs with mDCs in the upper
well, suppression of proliferation of responder T cells was still completely inhibited.
As cell contact seems to be important for suppression by IL-10 DC-induced Tregs,
we investigated whether the inhibitory molecules CTLA-4 or programmed death

CHAPTER 2
A Blocking cytokines B Transwell assay
75 40
% proliferating CD4+ Tresp

% proliferating CD4+ Tresp

30
50

20

25
10

0 0
- Isotype α IL-10 α TGFβ Tresp+mDC Control No stim. Stim.

C Blocking surface molecules

75
Tresp+mDC
% proliferating CD4+ Tresp

+ Tpr (IL-10 DC)

+ Tpr (mDC)

50

25

0
Isotype I α CTLA-4 Isotype II α PD-1

Figure 6. Suppression by IL-10 DC-induced Tregs is cell contact-dependent and not

cytokine-dependent. Naïve CD4+ T cell were primed by DCs for 6 days. DC-primed T cells
were harvested, washed and PKH26-labelled. PKH26-labelled primed T cells were added
1:1 to CFSE-labelled responder CD4+ memory T cells together with mDCs as stimulation
(Tresp:mDC = 50:1). (A) Blocking antibodies against IL-10, TGFβ or an isotype control were
added. The assay was performed in duplicate and mean + SEM of 3 independent experiments
is shown. (B) Suppression assay was performed in a transwell system where primed T cells
were separated from mDC-stimulated responder T cells by a microporous membrane, or not
separated (Control). Primed T cells in the upper chamber of the transwell were either not
stimulated (No stim.) or stimulated with mDCs (Stim.). The assay was performed in duplicate
and mean + SEM of 4 independent experiments is shown. (C) Blocking antibodies against
CTLA-4, PD-1 or isotype controls were added. The assay was performed in duplicate and
mean + SEM of 3 independent experiments is shown.

47
 CHAPTER 2

(PD)-1 are important for suppression by IL-10 DC-induced Tregs. Blocking CTLA-4
did not inhibit the suppression of proliferation of responder T cells by IL-10 DC-
induced Tregs. Also blocking PD-1 did not inhibit the suppression of proliferation
of responder T cells. Collectively, these results show that suppression by IL-10 DC-
induced Tregs is dependent on cell contact, or dependent on close proximity, and
not dependent on secretion of IL-10 or TGFβ.

Discussion
The induction of Tregs is an important measure of the tolerogenicity of DCs. The best
CHAPTER 2

read-out system for induction of human Tregs is a functional assay demonstrating

suppression of proliferation of stimulated responder T cells. However the setup of
suppression assays is often different and poorly defined. Therefore in this study we
optimized a suppression assay with human tolerogenic DC-induced Tregs. For our
studies we have induced Tregs from naïve T cells by IL-10-treated tolerogenic DCs.
Treatment of human monocyte-derived DCs with IL-10 results in the generation
of tolerogenic DCs.22-28 The data presented here confirm that IL-10 treatment
of DCs strongly inhibits DC maturation and T cell activation (Figure 1 and 2). In
agreement with these previous studies, our results show that IL-10 treatment
during maturation of DCs results in no detectable production of IL-12p70 and
in a higher IL-10 production than their normal matured counterparts (Figure 1B).
In contrast to a previous report 26, IL-10 DCs generated in our experiments also
have a higher TNFα production than mDCs. However, this can be explained by the
fact that McBride and co-workers 26 measured cytokine levels that were produced
in culture, while we have measured cytokine production of DCs after CD40L
restimulation. In addition, we have generated DCs under serum-free conditions,
which is a requirement for generation of clinically-applicable DCs. Furthermore, we
have used a cytokine maturation cocktail instead of LPS to mature the IL-10 DCs.
Therefore, the discrepancy in TNFα levels might be due to different experimental
designs or different cell preparations. In agreement with previous studies 22-24,26-28,
we have shown that IL-10 treatment of DCs results in a reduced expression of
the maturation marker CD83, costimulatory molecules CD80 and CD86 and the
MHC class II molecule HLA-DR, as compared with their normal mature counterparts
(Figure 1A). Furthermore, consistent with previous reports 22-24,26, we have shown
that IL-10-treated DCs have a reduced capacity to stimulate allogeneic naïve and
memory CD4+ T cells (Figure 2A).
Most assays for analysis of Treg suppressive capacity consist of a coculture setup
of Tregs with antigen-specific or polyclonal stimulated responder T cells.20,21 We
have chosen to use APCs in this case mDCs, as stimulators for the responder T cells
to mimic the in vivo situation more closely. Tregs can suppress T cell responses
both in a direct or an indirect way via the APC.10,11 In the absence of APCs the
responder T cells are the only target of suppression. In the presence of APCs
however, the target of Treg suppression can be the APC, the responder cell or both.
If Tregs induce suppression of responder T cell proliferation only via the indirect

48
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

way, this suppressive capacity would be missed when using a suppression assay
with a polyclonal stimulus. In addition, when using mDCs as T cell stimulation in a
suppression assay, the assay can potentially also be used to test the antigen-specific
suppressive capacity of the Tregs.
Our extensive analyses show that the number of mDCs used for the stimulation
of the responder T cells in the suppression assay is important for optimal read-out
of suppression by DC-induced Tregs. Less APC-mediated stimulation of responder
T cells allows more effective suppression by the Tregs (Figure 4). This is consistent
with the findings of Oberg and co-workers 29 who previously showed that upon
usage of anti-CD3/ anti-CD28 coated beads as T cell stimulation less beads resulted

CHAPTER 2
in more suppression by natural Tregs. The improved suppression we observed in the
presence of less APCs may be related to data of two studies which suggest that the
susceptibility of responder T cells for Treg-mediated suppression is determined by
the strength of activation of those responder T cells.30,31 Antons and co-workers 30
showed that increase of the strength of TCR signal can override the Treg-mediated
suppression. George and co-workers 31 addressed the question whether loss of
suppression that is seen with increasing stimulation comes from inactivation of Tregs
or whether increasing stimulation enables responder T cells to escape suppression.
Their results show that Tregs are still functional with high dose antigen stimulation
but potently stimulated responder T cells may produce sufficient levels of IL-2 to
override Treg-mediated suppression and thereby drive their own proliferation.
In addition, we have shown that the use of a negative control condition, T cells
primed by mDCs, is essential to discriminate specific suppression from generalised
suppression by DC-primed T cells. Longer priming conditions coincided with
stronger suppressive effects but also in the negative control condition (Figure 5B).
One possible reason for this is that the primed T cells are in a different, more
rested, stage of T cell activation after long priming periods, which may lead to more
generalised mechanisms of suppression upon coculture with responder T cells.
Titration of iTregs into the suppression assay showed that suppression of
proliferation of responder T cells by IL-10 DC-induced Tregs is dose-dependent
(Figure 5A). The highest amount of iTregs, that is a ratio of one responder T cell to
two iTreg cells, yielded a significant suppression of proliferation of mDC-stimulated
responder T cells. For naturally occurring Tregs often lower amounts of Tregs are
sufficient to suppress proliferation of responder T cells. The reason why we need a
relatively high amount of iTregs most likely reflects the probability that our priming
conditions did not yield a pure population of iTregs. Priming of naïve T cells by IL-10
DCs very likely results in a mixture of primed T cells with a relative enrichment of
the Treg fraction.
Our results show that suppression by IL-10 DC-induced Tregs is not dependent
on secretion of IL-10 or TGFβ (Figure 6A). Furthermore, transwell experiments
suggest that suppression by IL-10 DC-induced Tregs is dependent on cell contact
(Figure 6B). These data confirm a previous report of Steinbrink and co-workers.25 In
addition, using supernatant of IL-10 DC-induced Tregs, Steinbrink and co-workers
show that the suppression is not dependent on any soluble factor. To further address

49
 CHAPTER 2

which cell surface molecule is important for suppression, we used neutralizing

antibodies against PD-1 and CTLA-4. Our results suggest that suppression was
not mediated via these inhibitory molecules. However, Steinbrink and co-workers
reported that interaction between CTLA-4 and CD86, but not CD80, is important
for suppression by IL-10 DC-induced Tregs.25 The use of a transwell system does
not necessarily rule out the possibility that cytokines or other soluble factors are
important for suppression by Tregs.11 It might as well be that close proximity of cells,
immunological synapse formation between cells, or cell contact, is important for
efficient suppression via soluble factors. Alternatively, it is also possible that there
is no longer competition between Tregs and effector T cells for stimulation by the
CHAPTER 2

DCs, a mechanism described by Onishi and co-workers.32 Absence of competition

with Tregs then indeed results in proliferation of effector T cells. Further studies will
be needed to clarify the mechanisms responsible for the inhibitory effect of IL-10
DC-induced Tregs.
In this report we show extensive fine-tuning of a CFSE-based suppression
assay to determine the induction of human tolerogenic DC-induced Tregs. The
suppression assay uses mDC-stimulated T cell proliferation as read-out through
which it becomes broader applicable, e.g. to detect suppressive mechanisms which
are APC dependent and to determine antigen-specific suppression. Titration of
APCs in the assay is furthermore shown to be essential to define the ideal setting for
optimal suppression. In addition, the use of a negative control condition, consisting
of T cells primed by mDCs, ensures that generalised suppression phenomena can
be discerned from only specific iTreg-mediated suppression. The assay described in
this study can now be used as a tool for the comparison of different types of human
tolerogenic DCs in order to identify the optimal and best suited DC for tolerance
therapy.

Acknowledgements
This work was supported by a Sanquin PPOC grant (PPOC06-026). We thank ing.
Willy Karssing and dr. Hans Vrielink of the Sanquin Blood Bank North West Region
for leukaphaeresis and elutriation, and ing. Gijs van Schijndel and ing. Remco Visser
for help with monocyte isolations. We thank prof. dr. Ellen van der Schoot and
dr. Jan Voorberg for their critical review of the manuscript.

50
 OPTIMIZED SUPPRESSION ASSAY FOR DC-INDUCED TREG

References
1. Banchereau, J., Steinman, R.M. Dendritic Levings, M.K. Interleukin-10-secreting
cells and the control of immunity. Nature type 1 regulatory T cells in rodents and
1998; 392:245-252. humans. Immunol. Rev. 2006; 212:28-50.
2. Steinman, R.M., Hawiger, D., Nussenzweig, 16. Weiner, H.L. Induction and mechanism
M.C. Tolerogenic dendritic cells. Annu. of action of transforming growth factor-
Rev. Immunol. 2003; 21:685-711. beta-secreting Th3 regulatory cells.
3. Banchereau, J., Briere, F., Caux, C., Immunol. Rev. 2001; 182:207-214.
Davoust, J., Lebecque, S., Liu, Y.J., 17. Chen, Y., Kuchroo, V.K., Inobe, J., Hafler,
Pulendran, B., Palucka, K. Immunobiology D.A., Weiner, H.L. Regulatory T cell clones
of dendritic cells. Annu. Rev. Immunol. induced by oral tolerance: suppression of
2000; 18:767-811. autoimmune encephalomyelitis. Science

CHAPTER 2
4. Morelli, A.E., Thomson, A.W. Tolerogenic 1994; 265:1237-1240.
dendritic cells and the quest for transplant 18. Groux, H., O’Garra, A., Bigler, M.,
tolerance. Nat. Rev. Immunol. 2007; Rouleau, M., Antonenko, S., de Vries,
7:610-621. J.E., Roncarolo, M.G. A CD4+ T-cell
5. Thomson, A.W., Robbins, P.D. Tolerogenic subset inhibits antigen-specific T-cell
dendritic cells for autoimmune disease responses and prevents colitis. Nature
and transplantation. Ann. Rheum. Dis. 1997; 389:737-742.
2008; 67:90-96. 19. Quah, B.J.C., Warren, H.S., Parish, C.R.
6. Hackstein, H., Thomson, A.W. Dendritic Monitoring lymphocyte proliferation
cells: emerging pharmacological targets in vitro and in vivo with the intracellular
of immunosuppressive drugs. Nat. Rev. fluorescent dye carboxyfluorescein
Immunol. 2004; 4:24-35. diacetate succinimidyl ester. Nat.
7. Adorini, L., Giarratana, N., Penna, G. Protocols 2007; 2:2049-2056.
Pharmacological induction of tolerogenic 20. Venken, K., Thewissen, M., Hellings, N.,
dendritic cells and regulatory T cells. Somers, V., Hensen, K., Rummens, J.L.,
Semin. Immunol. 2004; 16:127-134. Stinissen, P. A CFSE based assay for
8. Turnquist, H.R., Thomson, A.W. Taming measuring CD4+CD25+ regulatory T cell
the lions: manipulating dendritic cells mediated suppression of auto-antigen
for use as negative cellular vaccines in specific and polyclonal T cell responses.
organ transplantation. Curr. Opin. Organ J. Immunol. Methods 2007; 322:1-11.
Transplant. 2008; 13:350-357. 21. Brusko, T.M., Hulme, M.A., Myhr, C.,
9. Roncarolo, M.G., Battaglia, M. Regulatory Haller, M.J., Atkinson, M.A. Assessing the
T-cell immunotherapy for tolerance to self in vitro suppressive capacity of regulatory
antigens and alloantigens in humans. Nat. T cells. Immunol. Invest. 2007; 36:607-
Rev. Immunol. 2007; 7:585-598. 628.
10. Vignali, D.A.A., Collison, L.W., Workman, 22. Steinbrink, K., Wolfl, M., Jonuleit, H.,
C.J. How regulatory T cells work. Nat. Rev. Knop, J., Enk, A.H. Induction of tolerance
Immunol. 2008; 8:523-532. by IL-10-treated dendritic cells. J.
11. Shevach, E.M. Mechanisms of Foxp3+ T Immunol. 1997; 159:4772-4780.
regulatory cell-mediated suppression. 23. Sato, K., Yamashita, N., Matsuyama, T.
Immunity 2009; 30:636-645. Human peripheral blood monocyte-
12. Carvalho-Gaspar, M., Jones, N.D., Luo, derived interleukin-10-induced semi-
S., Martin, L., Brook, M.O., Wood, K.J. mature dendritic cells induce anergic
Location and time-dependent control of CD4+ and CD8+ T cells via presentation
rejection by regulatory T cells culminates of the internalized soluble antigen and
in a failure to generate memory T cells. J. cross-presentation of the phagocytosed
Immunol. 2008; 180:6640-6648. necrotic cellular fragments. Cell. Immunol.
13. Belkaid, Y. Regulatory T cells and 2002; 215:186-194.
infection: a dangerous necessity. Nat. 24. Sato, K., Yamashita, N., Baba, M.,
Rev. Immunol. 2007; 7:875-888. Matsuyama, T. Modified myeloid dendritic
14. Lohr, J., Knoechel, B., Abbas, A.K. cells act as regulatory dendritic cells to
Regulatory T cells in the periphery. induce anergic and regulatory T cells.
Immunol. Rev. 2006; 212:149-162. Blood 2003; 101:3581-3589.
15. Roncarolo, M.G., Gregori, S., Battaglia, 25. Steinbrink, K., Graulich, E., Kubsch, S.,
M., Bacchetta, R., Fleischhauer, K., Knop, J., Enk, A.H. CD4+ and CD8+

51
 CHAPTER 2

anergic T cells induced by interleukin- 29. Oberg, H.H., Wesch, D., Lenke, J.,
10-treated human dendritic cells display Kabelitz, D. An optimized method for the
antigen-specific suppressor activity. functional analysis of human regulatory
Blood 2002; 99:2468-2476. T cells. Scand. J. Immunol. 2006; 64:353-
26. McBride, J.M., Jung, T., de Vries, J.E., 360.
Aversa, G. IL-10 alters DC function via 30. Antons, A.K., Wang, R., Kalams, S.A.,
modulation of cell surface molecules Unutmaz, D. Suppression of HIV-specific
resulting in impaired T-cell responses. and allogeneic T cell activation by human
Cell. Immunol. 2002; 215:162-172. regulatory T cells is dependent on the
27. De Smedt, T., Van Mechelen, M., De strength of signals. PLoS. One. 2008;
Becker, G., Urbain, J., Leo, O., Moser, M. 3:2952.
Effect of interleukin-10 on dendritic cell 31. George, T.C., Bilsborough, J., Viney, J.L.,
maturation and function. Eur. J. Immunol. Norment, A.M. High antigen dose and
CHAPTER 2

1997; 27:1229-1235. activated dendritic cells enable Th cells

28. Sato, K., Nagayama, H., Tadokoro, K., to escape regulatory T cell-mediated
Juji, T., Takahashi, T.A. Extracellular suppression in vitro. Eur. J. Immunol.
signal-regulated kinase, stress-activated 2003; 33:502-511.
protein kinase/c-Jun N-terminal kinase, 32. Onishi, Y., Fehervari, Z., Yamaguchi, T.,
and p38mapk are involved in IL-10- Sakaguchi, S. Foxp3+ natural regulatory
mediated selective repression of TNF- T cells preferentially form aggregates on
alpha-induced activation and maturation dendritic cells in vitro and actively inhibit
of human peripheral blood monocyte- their maturation. Proc. Natl. Acad. Sci. U.
derived dendritic cells. J. Immunol. 1999; S. A 2008; 105:10113-10118.
162:3865-3872.

52
 3
CHAPTER
IL-10-generated tolerogenic
dendritic cells are optimal for
functional regulatory T cell
induction – A comparative study
of human clinical-applicable DC
Martine A. Boks, Judith R. Kager-Groenland,
Michiel S.P. Haasjes, Jaap Jan Zwaginga,
S. Marieke van Ham and Anja ten Brinke

Clin. Immunol. 2012 Mar;142(3):332-42

 CHAPTER 3

Abstract
Tolerogenic dendritic cells (tDC) are a promising tool for specific cellular therapy to
induce immunological tolerance in transplantation and autoimmunity. To date, most
described tDC methods have not been converted into clinically applicable protocols
and systematic comparison of required functional characteristics, i.e. migration and
functional regulatory T cell (Treg) induction, is lacking. We compare clinical-grade
tDC generated with vitamin D3, IL-10, dexamethasone, TGFβ or rapamycin. For
good migratory capacity and a stable phenotype, additional maturation of tDC
was required. Maturation with a cocktail of TNFα, IL-1β and PGE2 induced optimal
migration. Importantly, all tDC showed a stable phenotype under pro-inflammatory
conditions. Especially IL-10 DC showed most powerful tolerogenic characteristics
with high IL-10 production and low T cell activation. Moreover, in a functional
suppression assay only IL-10 DC induced Treg that strongly suppressed T cell
reactivity. Thus, clinical-grade IL-10 DC show functional characteristics that make
them best suited for tolerance-inducing therapies.
CHAPTER 3

56
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

Introduction
Specific therapy to prevent or reduce immune activation is highly desired
in transplantation and autoimmunity. Current therapies, which include
immunosuppressive drugs, however, do not specifically target the cause of the
disease or transplant rejection and are associated with considerable side effects.
Ex vivo generated tolerogenic dendritic cells (tDC) are therefore an attractive
therapeutic alternative to maintain, restore or induce specific immunological
tolerance. Evidence from animal models demonstrated that injection of ex vivo
generated tDC can prevent or stop transplant rejection in skin graft 1 and heart
graft models 2,3, or re-establish self-tolerance in autoimmune diseases, like collagen-
induced arthritis 4,5, diabetes 6,7 and experimental autoimmune encephalomyelitis.8
These positive results have led to the first clinical trials with monocyte-derived tDC for
type I diabetes 9 and for rheumatoid arthritis (J.D. Isaacs and co-workers, Newcastle
University). tDC-clinical trials have not started in the field of transplantation, although
monocyte-derived tDC from rhesus macaques show promising results since tDC
infusion induces clear T cell hyporesponsiveness to donor allogeneic-antigens.10

CHAPTER 3
DC are highly specialized professional antigen-presenting cells that orchestrate
the immune response via integration of a variety of signals.11 Immunosuppressive
and anti-inflammatory compounds like IL-10 12-16, dexamethasone 17-22, 1α,25-
dihydroxyvitamin D3 20-24, rapamycin 1,22,25-27 and TGFβ 28-31 induce DC with tolerogenic
properties in vitro. tDC are generally characterized by an immature or semi-mature
phenotype, with low expression of costimulatory molecules. In addition, tDC
produce low amounts of pro-inflammatory cytokines and high amounts of anti-
inflammatory cytokines.
Although human tDC have been studied extensively in research settings, the
described tDC methods have often not been converted into clinically applicable
protocols using current Good Manufacturing Practices (cGMP). In addition,
systematic comparisons of essential functional characteristics of the various tDC
are limited. Important for induction of suppressive responses in naïve T cells is CC
chemokine receptor (CCR)7-directed migratory capacity of the ex vivo generated
tDC towards secondary lymphoid organs.32 Furthermore, as some animal studies
have shown that the tolerogenicity of DC is reversible 33-35, it is essential to
establish whether ex vivo generated tDC are stable and maintain their tolerogenic
properties when encountering pro-inflammatory signals in vivo. Finally, tDC need
to induce the right type of immune suppressive effector mechanism. tDC mediate
peripheral T cell tolerance by inducing immune deviation, anergy or apoptosis of
effector T cells, and/or induction of regulatory T cells (Treg).36-39 Key function of
tDC is the induction of functional Treg, as Treg in turn induce more widespread
tolerance.40-42
We are the first to compare the suppressive capacity of T cells primed by
different tDC by means of a functional assay in a large comparative study. The data
demonstrate that IL-10 DC induce suppressive Treg and display strong tolerogenic
potential, making them suitable for tolerance-inducing therapies.

57
 CHAPTER 3

Materials and Methods

Media and reagents
CellGro DC serum-free medium, GM-CSF, IL-4, IL-1β and TNFα were all obtained
from CellGenix (Freiburg, Germany). PGE2, dexamethasone, rapamycin, 1α,25-
dihydroxyvitamin D3 and monophosphoryl lipid A (MPLA, Toll-like receptor
(TLR)4 ligand) were purchased from Sigma-Aldrich (Steinheim, Germany). IFNγ
(Immukine) from Boehringer Ingelheim (Ingelheim am Rhein, Germany), IL-10 from
PeproTech (Rocky Hill, USA) and TGFβ from R&D Systems (Minneapolis, USA). R848
(imidazoquinoline compound; TLR7/8 ligand) was obtained from InvivoGen (San
Diego, USA) and SAC (heat-killed Staphylococcus aureus cells; TLR2 ligand) from
Calbiochem (Merck, Darmstadt, Germany). For read-out assays Iscove’s Modified
Dulbecco’s Medium (IMDM, Bio Whittaker, Verviers, Belgium) with 10% fetal calf
serum (Bodinco, Alkmaar, The Netherlands) was used as described.12

Antibodies used for flow cytometry

HLA-DR (phycoerythrin; PE), CD40 (fluorescein isothiocyanate; FITC), CD80 (FITC),
CHAPTER 3

CD83 (allophycocyanin; APC) and CD86 (APC) were purchased from Becton
Dickinson (BD Biosciences, San Jose, USA). PD-L1 (APC) and PD-L2 (PE) were
obtained from eBioscience (Vienna, Austria), and ILT3 (PE), ILT4 (APC) and CCR7
(PE) were obtained from R&D Systems. The following isotype-matched controls
were used: IgG1 (FITC), IgG1 (PE), IgG2a (PE) (Dako, Glostrup, Denmark), IgG1
(APC), IgG2a (APC) (BD Biosciences).

Isolation and culture of monocyte-derived dendritic cells

Monocytes were isolated from fresh aphaeresis material of healthy volunteers
(after informed consent) by using the ElutraTM cell separation system (Gambro,
Lakewood, USA).43 Monocytes were cultured at 1.5 x 106 cells/3 ml in 6 well plates
(Nunc, Roskilde, Denmark) in serum-free CellGro medium supplemented with IL-4
(800 IU/ml), GM-CSF (1000 IU/ml), penicillin (100 IU/ml) and streptomycin (100 μg/ml).
After 6 days, the immature DC were either left untreated (imDC) or matured (mDC)
with IL-1β (10 ng/ml), TNFα (10 ng/ml) and PGE2 (1 µg/ml) or with MPLA (2.5 μg/ml)
for 2 more days. To generate tDC, IL-10 (40 ng/ml), dexamethasone (10 nM), 1α,
25-dihydroxyvitamin D3 (100 nM), rapamycin (10 ng/ml) or TGFβ (5 ng/ml) was
added to imDC one hour before maturation.

DC migration
DC migration towards 100 ng/ml CC chemokine ligand (CCL)21 (R&D Systems)
was determined in 96 transwell plates with polycarbonate filters of 5 μm pore size
(Corning Costar, New York, USA) as described.44

DC phenotyping
DC were stained with specific mAb or appropriate isotype-matched controls as
described 12 and analyzed in presence of 4’,6-diamidino-2-phenylindole (DAPI;

58
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

Sigma-Aldrich) to exclude dead cells. Cells were measured on an LSRII flow

cytometer (BD Biosciences).

Cytokine production
DC were cultured with CD40 ligand (CD40L)-transfected J558 cells as described.12
After 24 hours, secreted IL-6, IL-10 and TNFα levels were determined by enzyme-
linked immunosorbent assay (ELISA) with PeliKine-compact ELISA kit (Sanquin
Reagents, Amsterdam, The Netherlands). For the detection of IL-23, a combination
of eBio473P19 mAb (eBioscience) and C8.6 mAb (BD Biosciences) was used.
Cytokines produced during coculture of CD4+ naïve T cells with DC were
measured after 6 days of culture. IL-10, IL-13 and IFNγ levels were determined by
ELISA using the PeliKine-compact ELISA kit (Sanquin Reagents).

Isolation of naïve and memory CD4+ T cells

CD4+CD45RA+CD45RO- naïve and CD4+CD45RA-CD45RO+ memory T cells were
purified from peripheral blood mononuclear cells or from the T cell-containing
fraction of ElutraTM using CD4 T cell isolation kit II followed by anti-CD45RO beads

CHAPTER 3
(Miltenyi Biotec, Bergisch Gladbach, Germany). Purities of T cell subsets were
routinely above 92%.

Allogeneic T cell proliferation assay

DC were used to stimulate allogeneic CD4+ memory T cells by co-culture (1:12.5) as
described.12 After 6 days, [3H]-thymidine (Amersham Biosciences, Amersham, UK)
was added and the incorporation was measured after 16 hours using a beta liquid
scintillation counter (1205 Betaplate Wallac).

DC stability
At the end of DC culture, the phenotypic and functional stability of DC was
investigated after washing and reculturing of DC in serum-free CellGro medium.
Cells were either left unstimulated or restimulated with SAC (1:1000) or R848
(10 μg/ml), both supplemented with IFNγ (1000 IU/ml). After 24 hours, restimulated
DC were analyzed for phenotype, cytokine production and T cell proliferation.

Suppression assay
DC were co-cultured (1:10) with 1 x 106 allogeneic CD4+ naïve T cells in 24 well plates
(Nunc) for 6 days. DC-primed T cells were harvested, washed and evaluated for their
suppressive capacity as described previously.12 Briefly, CD4+ memory responder
T cells (from the same T cell donor) were labeled with 0.5 μM carboxyfluorescein
diacetate succinimidyl ester (CFSE; Invitrogen, Eugene, Oregon, USA) for 15 minutes
at RT. Primed T cells were labeled with 6 μM PKH26 (Sigma-Aldrich) for 5 minutes at
RT and irradiated (30 Gray) to prevent convergence with proliferated CFSE-negative
cells. Primed T cells were incubated with 5 x 104 responder T cells (2:1, 1:1, 1:2)
and mDC (from the same DC donor; 1:50 ratio with responder T cells). Cells were
cultured in 96 well round bottom plates (Greiner Bio-One, Frickenhausen, Germany)

59
 CHAPTER 3

for 6 days in culture medium. Proliferation of responder T cells was measured on an

LSRII flow cytometer in the presence of DAPI to exclude dead cells.

Statistical analysis
Data are expressed as mean + standard error of the mean (SEM). Statistical
significance was analyzed using one-way Analysis Of Variance (ANOVA) or two-way
ANOVA (Figure 4A and C) with Dunnett’s or Bonferroni post test in Graphpad Prism
5.00 software (San Diego, USA).

Results
Effects of immunosuppressive compounds and co-maturation
on DC migratory capacity
Recently, we showed that cGMP IL-10-treated human monocyte-derived DC are
excellent tDC with good Treg-inducing capacities.12 Now, we generated different
types of tDC in a cGMP-translatable protocol using a closed system monocyte
CHAPTER 3

isolation (ElutraTM) and serum-free, cGMP-grade culture medium and cytokines for
immature DC (imDC) generation. tDC were generated using different tolerogenic
stimuli, e.g. vitamin D3, IL-10, dexamethasone, TGFβ or rapamycin, in combination
with a maturation cocktail and compared to immuno-activating mature DC (mDC;
maturation cocktail only) for migratory capacity. Addition of a maturation stimulus
was necessary to obtain migrating tDC (Figure S1), as was described before for
dexamethasone/ vitamin D3 DC.45 TGFβ DC, rapamycin DC and mDC showed equal
migration towards CCL21 (Figure 1A). Vitamin D3 DC, IL-10 DC and dexamethasone
DC migrated significantly less. tDC migrated to the same extent towards CCL19,
another CCR7 ligand (data not shown). Consistent with their migration, CCR7
expression was lower on vitamin D3 DC, IL-10 DC and dexamethasone DC (Figure 1B
and C). As vitamin D3 DC were the lowest migrators, we omitted these tDC from
further experiments.
tDC are often generated with LPS (lipopolysaccharide) as maturation stimulus.15,17,19-
21,23,24,27-29
Therefore, we compared DC matured with either the cytokine cocktail (IL-1β,
TNFα and PGE2) or with MPLA, a clinically applicable detoxified form of LPS.46 MPLA
invariably resulted in considerable lower migratory capacity than the cytokine cocktail
(Figure 1D). Consistent with the migration, CCR7 expression was lower on MPLA (t)
DC (Figure 1E). Since migration towards CCR7 ligands is necessary for de novo Treg
induction in the lymph nodes 32, we consider migration a required characteristic for
tDC. Therefore, we included the cytokine maturation cocktail in all our tDC protocols.

IL-10 DC, dexamethasone DC and TGFβ DC display phenotypic

characteristics of tolerogenic DC
To determine the phenotype of the tDC, we analyzed expression of maturation-
related markers. Compared to imDC and mDC, rapamycin DC showed a mature
phenotype and all other tDC were semi-mature as demonstrated by intermediate

60
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

A D
Cocktail DC migration DC migration

mDC mDC MPLA

Cocktail
VD3 *** IL-10
IL-10 ***
Dex
Dex ***
TGF β
TGFβ

Rapa Rapa

0 25 50 75 100 125 0 25 50 75 100 125

relative % of migrated cells % migrated cells

B E
mDC VD3 IL-10 Cocktail MPLA

70% 9% 10% 44% 12%

CHAPTER 3
Counts
Counts

Dex TGFβ Rapa IL-10 + cocktail IL-10 + MPLA

11% 44% 74% 20% 8%

CCR7 CCR7
C
CCR7
60
% positive cells

40

*
20
** ***

0
mDC VD3 IL-10 Dex TGFβ
β Rapa

Figure 1. Tolerogenic DC treated with the cytokine cocktail have migratory capacity
towards CCL21. (A) Migration of tDC matured with vitamin D3, IL-10, dexamethasone, TGFβ
or rapamycin, supplemented with IL-1β, TNFα and PGE2 compared to DC matured with the
cytokine cocktail only (mDC). Relative migration (mean + SEM) of cytokine-matured tDC
compared to mDC of 8 or 7 (vitamin D3 DC) independent experiments. Mean percentage
of migrated mDC is 75.8 ± 6.0%. No migration was detected towards medium alone (data
not shown). (B) Histograms of CCR7 expression of cytokine-matured tDC and mDC are
shown as open graphs, with a matched isotype control in grey. A representative graph of
6 independent experiments is shown. (C) CCR7 expression of cytokine-matured tDC and
mDC. The percentage of positive cells (mean + SEM) is derived from 7 or 6 (vitamin D3 DC)
independent experiments. (D) Migration of tDC and mDC matured with the cytokine cocktail
or with MPLA. A representative graph of 3 independent experiments is shown. (E) Histograms
of CCR7 expression is shown as open graphs, with a matched isotype control in grey. *p≤0.05,
**p≤0.01, ***p≤0.001 (significant difference to mDC).

61
 CHAPTER 3

A CD80 CD86 CD83 CD40 HLA-DR

mDC

imDC

IL-10
Counts

Dex

TGFβ
CHAPTER 3

Rapa

Fluorescence intensity
B CD80 CD83
CD86 CD40 +++
HLA-DR
1200 1500 2400 +++ 1500 5000
+++ +++ *
+++ +++
*** +++
4000
+++ +
800 +++ 1000 ++ 1600 1000
+++ * ** 3000
*** ***
MFI

* *** 2000
400 *** 500 800 +++ 500
++ ***
*** 1000
***
0 0 0 *** 0 0
0
TG ex

0
TG ex

a
im C
IL C

im C
IL C
Fβ

R β
0
TG ex

0
TG ex

0
TG ex

a
im C
IL C

im C
IL C

im C
IL C
R β

R β

R β
-1

-1
ap

ap
F
-1

-1

-1
ap

ap

ap
F

F
D
D

D
D
D
D

D
D

D
D
D

D
D

D
m

m
m

m
R

C
ILT3 ILT4 PD-L1 PD-L2
10000 +++ 4000 15000 800
***
+++
8000 +++ +++
3000 *** +++ 600 +
10000 * *
+++
6000
MFI

2000 400 +++

++ ++ ***
4000 ***
5000
1000 200
2000
***
0 0 0 0
im C
IL C
0
TG ex

a
Fβ

0
TG ex

0
TG ex

0
TG ex

a
im C
IL C

im C
IL C

im C
IL C
R β

R β

Fβ
ap
-1

-1

-1

-1
ap

ap

ap
F

F
D
D

D
D

D
D

D
D
D

D
m

m
R

Figure 2. Phenotype of tolerogenic DC. (A) Histograms of the different markers are
shown as open graphs, with matched isotype controls in grey. A representative graph of 13
independent experiments is shown. (B) The mean fluorescence intensity (MFI; mean + SEM)
is derived from 13 independent experiments. The MFI is corrected for the MFI of matched
isotype controls. (C) Corrected MFI (mean + SEM) from 6 independent experiments is shown.
*p≤0.05, **p≤0.01, ***p≤0.001 (significant difference to mDC (*) or to imDC (+) ).

62
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

expression of the costimulatory molecules CD80, CD86, CD40 and of the maturation
marker CD83 (Figure 2A and B). HLA-DR expression was not altered by the
immunosuppressive agents. In addition, we examined the expression of inhibitory
molecules associated with tolerogenicity.47-51 Immunoglobulin-like transcript (ILT) 3
and ILT4 were upregulated by IL-10 (Figure 2C). ILT4 expression was also slightly
upregulated by TGFβ. Programmed death ligand (PD-L)1 was highly upregulated on
all DC types compared to imDC, although by most tDC conditions less than mDC.
PD-L2 expression was either not affected or downregulated by the different DC
treatments. Remarkably, dexamethasone DC have lower expression of PD-L1 and
PD-L2 when compared to the other tDC. Altogether, these data demonstrate that
IL-10 DC, dexamethasone DC and TGFβ DC are semi-mature while rapamycin DC
have a mature phenotype.

Functional properties of tolerogenic DC

Next, we investigated secretion of pro-inflammatory cytokines and the anti-
inflammatory cytokine IL-10 upon CD40L stimulation of the tDC. TNFα and IL-6
production was similar between the different DC, except for lower secretion of

CHAPTER 3
IL-6 by rapamycin DC (Figure 3A). All tolerance-inducing agents suppressed
IL-23 production. IL-12p70 was not produced by any DC type (data not shown).
Furthermore, IL-10 was produced in 3- to 4-fold higher amounts by IL-10 DC, while
the other tDC showed similar IL-10 levels as mDC. To evaluate the ability of tDC
to induce T cell proliferation, DC were co-cultured with allogeneic CD4+ memory
T cells. Only IL-10 DC induced less T cell proliferation than mDC, and was similar to
imDC (Figure 3B). To analyze T cell polarization by the DC, we measured cytokine
production in the DC – T cell cocultures. Cocultures with IL-10 DC and dexamethasone
DC produced significant higher amounts of IL-10 (Figure 3C). Furthermore, T cells
in IL-10 DC cultures produced lower levels of IL-13 and this trend was also seen
for IFNγ. Taken together, these data show that dexamethasone DC and TGFβ DC
have a semi-mature phenotype, but resemble mDC in functional properties, while
rapamycin DC have a mature phenotype with high T cell stimulatory capacity, but
low cytokine production. IL-10 DC are semi-mature, have high IL-10 production, low
T cell stimulatory capacity with low IL-13 and IFNγ polarization.

Co-maturation is required for the stability of tolerogenic DC

A potential risk of ex vivo generated tDC is that they may switch to an activating
phenotype after encountering ‘danger’ signals in vivo. To address this issue, tDC
were treated with TLR stimuli and examined for stability. We first tested various
TLR ligands together with the pro-inflammatory cytokine IFNγ and found that R848
(TLR7/8 ligand) and SAC (TLR2 ligand) were the strongest inducers of cytokine
production by DC (Figure S2). We then used these strong stimuli to investigate
the stability of tDC compared to imDC. As expected, imDC matured upon TLR
stimulation as demonstrated by increased expression of the costimulatory molecules
CD80, CD83 and CD86 (Figure 4A). mDC were hardly affected by additional TLR
triggering, demonstrating that they were already fully mature before reactivation.

63
 CHAPTER 3

A
TNFα
α IL-6 IL-23 IL-10
4 8 6 1000
***
800
3 6
4
* 600

ng/ml

ng/ml

pg/ml
ng/ml

*
2 4
** 400
2 ***
1 2 *
200

0 0 0 0
C

C
0

TG x

TG x

0
ex

TG x

a
R β

R β

R β

R β
-1

ap

-1

ap

-1

ap
e

-1

ap
e
F

F
D

D
D

D
TG
IL

IL

IL

IL
m

m
B C
Lymphocyte proliferation IL-10 IL-13 IFNγγ
20 40 800 2000
* *
15 30 600 1500
CPM (x103)

pg/ ml

pg/ ml

pg/ ml
10 20 400 1000
CHAPTER 3

5 ** *** 10 200 500

**
0 0 0 0
mDC imDC IL-10 Dex TGFβ
β Rapa

C
0

TG x

a
TG x

TG x

a
C
0

R β
R β

R β
-1

ap
-1
e

-1
e

e
ap

ap

F
F
F
D

D
D

IL
m

m
IL

IL

Figure 3. Cytokine production and lymphocyte activation by tolerogenic DC. (A) TNFα,
IL-6, IL-23 and IL-10 production by DC after 24 hours incubation with CD40L-expressing J558
cells, as a mimic of T cell stimulation. The cultures were performed in duplicate and mean
+ SEM of 9 or 5 (IL-23) independent experiments are depicted. (B) DC were incubated with
allogeneic CD4+ memory T cells (ratio 1:12.5). Proliferation was determined on day 6 after
16 hours of [3H]-thymidine incorporation. The assay was performed in triplicate and results are
shown as mean + SEM of 8 or 4 (imDC) independent experiments. (C) DC were incubated with
CD4+ naïve T cells (ratio 1:10). Production of IL-10, IL-13 and IFNγ was determined on day 6.
The cultures were performed in duplicate and mean + SEM of 5 independent experiments
is shown. *p≤0.05, **p≤0.01, ***p≤0.001 (significant difference to mDC). CPM, counts per
minute.

Dexamethasone DC showed modest marker increase, but much less than imDC.
The other tDC showed stable marker expression upon TLR stimulation. In terms
of cytokine production, TLR-stimulated imDC produced high amounts of TNFα,
IL-6 and IL-10 (Figure 4B). All tDC produced significantly lower amounts of TNFα
and IL-6 compared to imDC. In addition, IL-10 was produced in high amounts by
the IL-10 DC. IL-12p70 production was not detected for TLR-stimulated DC (data
not shown). Importantly, the addition of the cytokine maturation cocktail to the
tolerogenic compounds was required for the stability of all tDC except rapamycin
DC, since tDC that were not cytokine-matured were as sensitive to TLR triggering as
imDC (Figure S3). The amount of T cell proliferation of tDC and mDC did not change
after TLR stimulation (Figure 4C). These data indicate that all cytokine-matured tDC
have a stable phenotype and function.

64
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

A
CD80 CD83 CD86
4 75 12
***
***
R848/IFNγ
***
SAC/IFNγ
Fold increase

3
50 8

2 *** ***
***
* *** 25 4
1 ***
**

0 0 0
imDC IL-10 Dex TGF β Rapa mDC imDC IL-10 Dex TGF β Rapa mDC imDC IL-10 Dex TGF β Rapa mDC

B
TNFα
α IL-6 IL-10
2.0 20 8 25 20 150
R848/IFNγ
ng/ml (R848/IFNγγ )

pg/ml (R848/IFNγγ )
ng/ml (R848/IFNγγ )

ng/ml (SAC/IFNγγ )

pg/ml (SAC/IFNγγ )
ng/ml (SAC/IFNγγ )

20 SAC/IFNγ
1.5 15 6 15
100
15
1.0 10 4 10
10
50
0.5 5 2 5
5 **
** **
**
0.0 0 0 0 0 0
imDC IL-10 Dex TGFβ
β Rapa mDC imDC IL-10 Dex TGFβ
β Rapa mDC imDC IL-10 Dex TGFβ
β Rapa mDC

CHAPTER 3
C Lymphocyte proliferation
25
Unstimulated
20 * R848/IFNγ
SAC/IFNγ
CPM (x103)

15

10

0
imDC IL-10 Dex TGFβ
β Rapa mDC

Figure 4. Tolerogenic DC possess stable phenotype and functions after TLR stimulation.
DC were harvested on day 8 of culture, extensively washed and recultured for one more
day with SAC or R848, both supplemented with IFNγ, or only medium. After 24 hours of
restimulation, DC were analyzed for surface molecule expression, cytokine production
and lymphocyte proliferation. (A) Marker expression as fold increase of MFI compared to
unstimulated conditions (mean + SEM) of 3 independent experiments. (B) Mean cytokine
production + SEM of 6 independent experiments. Unstimulated DC do not produce cytokines
above the detection limit. All tDC produced significantly lower amounts of TNFα and IL-6
than imDC. (C) Lymphocyte proliferation (mean + SEM) of 3 independent experiments is
shown. *p≤0.05, **p≤0.01, ***p≤0.001 (significant difference to unstimulated conditions
(A, C) or to imDC (B) ).

IL-10 DC induce strong functional Treg

To assess the tolerogenic potential of the tDC for tolerance-inducing therapy, the
induction of Treg is an essential measure. As specific markers for induced human
Treg are still lacking, we evaluated functional Treg formation by tDC in an allogeneic
suppression assay.12 CD4+ naïve T cells were primed with the various tDC or mDC.
Subsequently, primed T cells were tested for their suppressive capacity on DC-
stimulated CD4+ memory responder T cells. IL-10 DC-induced Treg significantly

65
 CHAPTER 3

suppressed DC-induced proliferation of responder T cells (Figure 5), while T cells

primed by mDC showed no suppressive capacity. T cells primed by the other tDC
showed some, but no significant suppression. These results show that IL-10 DC are
the most powerful tDC to induce Treg that potently suppress T cell reactivity.

A
Tpr (mDC) Tpr (IL-10) Tpr (Dex) Tpr (TGFβ) Tpr (Rapa)
Counts

75% 41% 48% 56% 64%

CFSE
B
CHAPTER 3

Tpr (IL-10 DC) Tpr (Dex DC)

60 50
Tresp+mDC
% CFSElow Tresp

+Tpr (tDC)
% CFSElow Tresp

40
40 +Tpr (mDC)
+ ** 30
++ ***
20
20 ***
10

0 0
2:1 1:1 1:2 2:1 1:1 1:2
Tpr:Tresp ratio Tpr:Tresp ratio

Tpr (TGFβ
β DC) Tpr (Rapa DC)
50 50
% CFSElow Tresp

% CFSElow Tresp

40 40

30 30

20 20

10 10

0 0
2:1 1:1 1:2 2:1 1:1 1:2
Tpr:Tresp ratio Tpr:Tresp ratio

Figure 5. IL-10 DC induce Treg that suppress mDC-induced responder T cell proliferation.
CD4+ naïve T cells were primed by allogeneic DC (Tpr) in a 1:10 ratio for 6 days and subsequently
tested for their suppressive capacity on responder T cells. DC-primed T cells were harvested,
washed and labeled with PKH26. Primed T cells were added in a 2:1 ratio (A) or in grading
doses (B) to CFSE-labeled responder CD4+ memory T cells (Tresp) together with mDC as
stimulation (mDC:Tresp ratio 1:50). The assay was performed in duplicate. (A) Histograms
of the different Tpr conditions showing CFSE dilution of responder T cells. A representative
graph of 9 independent experiments is shown. (B) Proliferation of responder T cells (mean +
SEM; 11 independent experiments). *p≤0.05, **p≤0.01, ***p≤0.001 (significant difference to
Tpr (mDC) (*) or to Tresp+mDC (+) ).

66
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

Discussion
In this study we set out to determine the best protocol for tolerance-inducing
potential of tDC in a simple cGMP-translatable protocol. To the best of our
knowledge, this is the first comparison of different types of clinical-grade tDC
focusing on tDC migratory capacity and suppressive capacity by tDC-primed T cells.
We isolated cells in a closed system and cultured DC under serum-free conditions
using cGMP medium and cytokines. As reported before 45, additional maturation
appeared necessary, since less than 5% of tDC could migrate without maturation.
When comparing maturation with MPLA, a non-toxic, clinically approved derivative
of LPS 46, to maturation with a cytokine cocktail (consisting of IL-1β, TNFα and
PGE2), the cytokine cocktail was clearly superior in promoting migration of tDC,
which is in line with expectations.52 Especially rapamycin DC have high migratory
capacity and high expression of CCR7, which is in agreement with a previous study
of Sordi and co-workers.53 1α,25-dihydroxyvitamin D3 treatment has been shown
to induce good tolerogenic properties in DC.20-24 We here demonstrated however,
that vitamin D3 DC hardly migrate towards CCL21. Migratory capacity towards

CHAPTER 3
CCR7 ligands is needed for tDC migration towards secondary lymphoid organs 32,
and therefore essential for de novo Treg induction. In addition to good migratory
properties, tDC should not produce IL-12p70, a potent Th1-inducing cytokine. Lack
of IL-12p70 production by tDC is pivotal to their tolerogenicity.36,54 The cytokine
cocktail prevents the production of IL-12p70 55 and is therefore highly appropriate
for the generation of migratory tDC.
We and others have previously shown that IL-10 treatment during DC maturation
induces semi-mature DC with great potential to induce Treg.12-16 Both cell contact-
dependent mechanisms 15 and IL-10 secretion 16 are important for Treg induction by
IL-10 tDC. ILT3 and ILT4 are required for inhibition of T cell proliferation 56 and Treg
induction.16,57 The IL-10 DC that we generated in our clinically applicable protocol
indeed show high ILT3 and ILT4 expression, and high IL-10 production, which likely
contributes to Treg induction.
Of particular importance in tDC treatment is the stability of DC phenotype and
function, as in vivo applied tDC may encounter strong pro-inflammatory stimulation.
Therefore, it is a prerequisite for tolerance-promoting therapy that tDC phenotype
cannot convert into an immunostimulatory phenotype after in vivo administration.58
Immature DC have tolerogenic properties as well 59,60, but they are not stable and
therefore likely to be unsafe for therapeutic application.37 Our study shows that
co-maturation of tDC is indeed necessary for stable cytokine production upon TLR
restimulation. All cytokine-matured tDC are resistant to further maturation after
stimulation with strong pro-inflammatory stimuli and maintain a stable phenotype,
cytokine production and T cell stimulatory capacity. Our results confirm and extend
previous studies 20-22, showing stable phenotype and functions for all tDC.
Treatment of human monocyte-derived DC with vitamin D3 20-24, TGFβ 28-31,
rapamycin 22,26,27, dexamethasone 17-22 or IL-10 12-16 is described to induce immune
inhibitory DC. However only in a few studies the suppressive capacity of the

67
 CHAPTER 3

DC-primed T cells was tested.14,16,19,20,28,31,61 Although it is shown that vitamin D3

DC have good Treg-inducing capacity 20 and may therefore be an interesting
candidate for tolerance induction therapy, we did not address this in this study as
discussed above. In our study priming of naïve T cells with IL-10 DC induced robust
Treg. These Treg potently suppressed T cell reactivity, as was previously shown by
ourselves and others.12,14,16,31,61 High IL-10 production by IL-10 tDC-induced Treg
suggests a Tr1 phenotype. Tr1 cells mediate suppression via IL-10 and TGFβ.16,62
However, previously we showed that IL-10 and TGFβ are not important for the
suppressive function of these Treg.12 Transwell and blocking experiments suggest
that cell contact is necessary for suppression, although not mediated via CTLA-4 or
PD-1.12 Further studies will be needed to clarify the suppressive mechanism of IL-10
tDC-induced Treg.
T cells primed by TGFβ DC, dexamethasone DC or rapamycin DC were less
capable of inducing suppression. Our results show that TGFβ DC exhibit a semi-
mature phenotype, but do not differ much in functional properties when compared
to mDC. This is compatible with the findings of Torres-Aguilar and co-workers 31,
showing that DC treated with TGFβ plus IL-10 were more similar to mDC. Li and
CHAPTER 3

co-workers showed induction of functional Treg by TGFβ DC. These were, however,
generated by addition of M-CSF instead of GM-CSF.28
Our results confirm the inhibited DC maturation by dexamethasone treatment,
but we do not see a difference in cytokine production or T cell activation when
compared to mDC, and only a tendency towards induction of suppressive T cells.
Unger and co-workers showed that T cells primed by dexamethasone DC were
suppressive after 2 rounds of stimulation 20, while we studied the suppressive
capacity of the primed T cells after 1 round of tDC stimulation, which might explain
the discrepancy between the results. Also Woltman and co-workers showed
suppression by dexamethasone DC-primed T cells.19 Discrepancies might be caused
by the culture medium or different maturation stimuli used in this study.
The immunosuppressive drug rapamycin is extensively studied as a tolerance-
inducing agent for bone-marrow derived murine DC, with good clinical outcomes
when used as cellular vaccine in organ transplantation models.1,2,25,63,64 The effect
of rapamycin on human DC is less elaborately studied. Some studies show a semi-
mature phenotype and function for rapamycin DC 26,27, while a recent study also
showed a rather mature phenotype for rapamycin DC, but with modest induction
of Foxp3+ T cells.22 The regulatory capacities of the primed T cells in these studies
were however not analyzed in functional suppression assays. Here we demonstrate
for rapamycin DC a rather mature phenotype, with low cytokine production and high
T cell stimulatory capacity. In our hands, in a functional assay determining actual
suppressive capacity of primed T cells, we do not see a significant suppression by
rapamycin DC-primed T cells.
Besides the finding that IL-10 DC-primed T cells were the only ones to induce
significant suppression, IL-10 DC produce high amounts of IL-10 after CD40L
stimulation and induce low levels of T cell activation. Our data therefore suggests
that IL-10 DC have strong potential in tolerance-inducing therapies. In conclusion,

68
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

for application of tolerance-inducing DC in transplantation or autoimmunity we

recommend the use of tolerogenic compounds in combination with TNFα, IL-1β
and PGE2 to obtain tDC with migratory capacities and a stable immunosuppressive
phenotype under pro-inflammatory stress. In addition, in our cGMP culture system
only the IL-10 DC induced reproducibly potent Treg that suppressed T cell responses.
Thus, IL-10 DC show a strong potential for further exploration as a means to restore
immunological tolerance in clinical settings.

Acknowledgements
This work was supported by a Sanquin PPOC grant (PPOC06-026). We thank Sanquin
Blood Bank North West Region for leukaphaeresis and elutriation, and ing. Gijs van
Schijndel and ing. Remco Visser for help with monocyte isolations.

CHAPTER 3

69
 CHAPTER 3

Supplementals

DC migration

- No maturation
Cocktail
VD3

IL-10

Dex

Rapa

0 20 40 60 80 100
% migrated cells

Figure S1. Tolerogenic DC cultured without maturation cocktail have no migratory

capacity towards CCL21. DC were either matured with a cytokine cocktail (TNFα, IL-1β,
CHAPTER 3

PGE2) or not matured and assessed for migratory potential after 2 days. No migration was
detected towards medium alone (data not shown). The percentage migration (mean + SEM)
of 2 independent experiments is shown.

TNFα
α IL-6 IL-10
10 25 80

8 20
60
ng/mL

ng/mL

pg/mL

6 15
40
4 10
20
2 5

0 0 0
n

n
γ

γ
N

N
io

io

io
/IF

IF

/IF

IF

/IF

IF

/IF

IF

/IF

IF

/IF

IF
at

at

at
8/

8/

8/
4/

4/

4/
C

C
A

A
ul

ul

ul
SK

SK

SK
84

84

84
PL

SA

PL

SA

PL

SA
im

im

im
R

R
3C

3C

3C
M

M
st

st

st
re

re

re
m

m
Pa

Pa

Pa
no

no

no

Figure S2. TLR stimulations of immature DC. Immature DC were harvested on day 8 of
culture, extensively washed and recultured with MPLA (TLR4 ligand; 2.5 μg/ml), R848 (TLR7/8
ligand; 10 μg/ml), SAC (TLR2 ligand; 1:1000) or Pam3CSK4 (TLR3 ligand; 5 μg/ml), all
supplemented with IFNγ (1000 IU/ml). After 24 hours of stimulation, the culture supernatant
was harvested and analyzed for TNFα, IL-6 and IL-10 production. Mean cytokine production
+ SEM of 4 independent experiments is shown.

70
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

A R848/IFNγ
TNFα
α IL-6 IL-10
2.5 15 80

2.0
60
10

ng/mL
1.5

pg/ml
ng/ml

40
1.0
5
20
0.5

0.0 0 0
mDC imDC IL-10 Dex TGFβ
β Rapa mDC imDC IL-10 Dex TGF β Rapa mDC imDC IL-10 Dex TGF β Rapa

B SAC/IFNγ
TNFα
α IL-6 IL-10
15 15 250

200
10 10
ng/mL
ng/ml

150

pg/ml
100
5 5
50

CHAPTER 3
0 0 0
mDC imDC IL-10 Dex TGF β Rapa mDC imDC IL-10 Dex TGF β Rapa mDC imDC IL-10 Dex TGFβ
β Rapa

Figure S3. Tolerogenic DC cultured without maturation cocktail have no stable cytokine
production after TLR stimulation. mDC, imDC or immature tDC (cultured without the cytokine
cocktail of TNFα, IL-1β and PGE2) were harvested on day 8 of culture, extensively washed and
recultured for one more day with R848/ IFNγ (A) or with SAC/ IFNγ (B). After 24 hours of
restimulation TNFα, IL-6 and IL-10 production was measured. Mean cytokine production +
SEM of 2 independent experiments is shown. Unstimulated DC do not produce cytokines
above the detection limit.

71
 CHAPTER 3

References
1. Horibe, E.K., Sacks, J., Unadkat, J., induces regulatory dendritic cells with
Raimondi, G., Wang, Z., Ikeguchi, R., therapeutic effects on autoimmune
Marsteller, D., Ferreira, L.M., Thomson, disorders. Proc. Natl. Acad. Sci. U. S. A.
A.W., Lee, W.P., Feili-Hariri, M. 2005; 102:13562-13567.
Rapamycin-conditioned, alloantigen- 9. Giannoukakis, N., Phillips, B., Finegold,
pulsed dendritic cells promote indefinite D., Harnaha, J., Trucco, M. Phase I
survival of vascularized skin allografts (safety) study of autologous tolerogenic
in association with T regulatory cell dendritic cells in type 1 diabetic patients.
expansion. Transpl. Immunol. 2008; Diabetes Care 2011; Epub ahead of
18:307-318. print.
2. Raimondi, G., Sumpter, T.L., Matta, 10. Zahorchak, A.F., Kean, L.S., Tokita,
B.M., Pillai, M., Corbitt, N., Vodovotz, D., Turnquist, H.R., Abe, M., Finke,
Y., Wang, Z., Thomson, A.W. Mammalian J., Hamby, K., Rigby, M.R., Larsen,
target of rapamycin inhibition and C.P., Thomson, A.W. Infusion of
alloantigen-specific regulatory T cells stably immature monocyte-derived
synergize to promote long-term graft dendritic cells plus CTLA4Ig modulates
survival in immunocompetent recipients. alloimmune reactivity in rhesus
J. Immunol. 2010; 184:624-636. macaques. Transplantation 2007;
3. Bonham, C.A., Peng, L., Liang, X., Chen, 84:196-206.
Z., Wang, L., Ma, L., Hackstein, H., 11. Banchereau, J., Steinman, R.M. Dendritic
Robbins, P.D., Thomson, A.W., Fung,
CHAPTER 3

cells and the control of immunity. Nature

J.J., Qian, S., Lu, L. Marked prolongation 1998; 392:245-252.
of cardiac allograft survival by dendritic
cells genetically engineered with NF- 12. Boks, M.A., Zwaginga, J.J., van Ham,
kappa B oligodeoxyribonucleotide S.M., ten Brinke, A. An optimized
decoys and adenoviral vectors encoding CFSE-based T-cell suppression assay
CTLA4-Ig. J. Immunol. 2002; 169:3382- to evaluate the suppressive capacity
3391. of regulatory T-cells induced by human
tolerogenic dendritic cells. Scand. J.
4. van Duivenvoorde, L.M., Han, W.G.,
Immunol. 2010; 72:158-168.
Bakker, A.M., Louis-Plence, P.,
Charbonnier, L.M., Apparailly, F., van der 13. Steinbrink, K., Wolfl, M., Jonuleit,
Voort, E.I., Jorgensen, C., Huizinga, T.W., H., Knop, J., Enk, A.H. Induction of
Toes, R.E. Immunomodulatory dendritic tolerance by IL-10-treated dendritic
cells inhibit Th1 responses and arthritis cells. J. Immunol. 1997; 159:4772-4780.
via different mechanisms. J. Immunol. 14. Sato, K., Yamashita, N., Baba, M.,
2007; 179:1506-1515. Matsuyama, T. Modified myeloid
5. Popov, I., Li, M., Zheng, X., San, H., dendritic cells act as regulatory dendritic
Zhang, X., Ichim, T., Suzuki, M., Feng, B., cells to induce anergic and regulatory
Vladau, C., Zhong, R., Garcia, B., Strejan, T cells. Blood 2003; 101:3581-3589.
G., Inman, R., Min, W.P. Preventing 15. McBride, J.M., Jung, T., de Vries, J.E.,
autoimmune arthritis using antigen- Aversa, G. IL-10 alters DC function via
specific immature dendritic cells: a novel modulation of cell surface molecules
tolerogenic vaccine. Arthritis Res. Ther. resulting in impaired T-cell responses.
2006; 8:R141. Cell. Immunol. 2002; 215:162-172.
6. Adorini, L. Tolerogenic dendritic cells 16. Gregori, S., Tomasoni, D., Pacciani, V.,
induced by vitamin D receptor ligands Scirpoli, M., Battaglia, M., Magnani,
enhance regulatory T cells inhibiting C.F., Hauben, E., Roncarolo, M.G.
autoimmune diabetes. Ann. N. Y. Acad. Differentiation of type 1 T regulatory (Tr1)
Sci. 2003; 987:258-261. cells by tolerogenic DC-10 requires the
7. Phillips, B.E., Giannoukakis, N., Trucco, IL-10-dependent ILT4/HLA-G pathway.
M. Dendritic cell mediated therapy for Blood 2010; 116:935-944.
immunoregulation of type 1 diabetes 17. Piemonti, L., Monti, P., Allavena, P.,
mellitus. Pediatr. Endocrinol. Rev. 2008; Sironi, M., Soldini, L., Leone, B.E., Socci,
5:873-879. C., Di Carlo, V. Glucocorticoids affect
8. Chorny, A., Gonzalez-Rey, E., Fernandez- human dendritic cell differentiation and
Martin, A., Pozo, D., Ganea, D., Delgado, maturation. J. Immunol. 1999; 162:6473-
M. Vasoactive intestinal peptide 6481.

72
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

18. Xia, C.Q., Peng, R., Beato, F., Clare- hyporesponsiveness independent of
Salzler, M.J. Dexamethasone induces FoxP3 induction. Immunol. Lett. 2008;
IL-10-producing monocyte-derived 120:49-56.
dendritic cells with durable immaturity. 27. Monti, P., Mercalli, A., Leone, B.E.,
Scand. J. Immunol. 2005; 62:45-54. Valerio, D.C., Allavena, P., Piemonti, L.
19. Woltman, A.M., van der Kooij, S.W., Rapamycin impairs antigen uptake of
de Fijter, J.W., van, K.C. Maturation- human dendritic cells. Transplantation
resistant dendritic cells induce 2003; 75:137-145.
hyporesponsiveness in alloreactive 28. Li, G., Abediankenari, S., Kim, Y.J.,
CD45RA+ and CD45RO+ T-cell Campbell, T.B., Ito, S., Graham-Evans,
populations. Am. J. Transplant. 2006; B., Cooper, S., Broxmeyer, H.E. TGF-
6:2580-2591. [beta] combined with M-CSF and IL-4
20. Unger, W.W., Laban, S., Kleijwegt, induces generation of immune inhibitory
F.S., van der Slik, A.R., Roep, B.O. cord blood dendritic cells capable of
Induction of Treg by monocyte-derived enhancing cytokine-induced ex vivo
DC modulated by vitamin D(3) or expansion of myeloid progenitors.
dexamethasone: Differential role for Blood 2007; 110:2872-2879.
PD-L1. Eur. J. Immunol. 2009; 39:3147- 29. Fogel-Petrovic, M., Long, J.A.,
3159. Misso, N.L., Foster, P.S., Bhoola,
21. Chamorro, S., Garcia-Vallejo, J.J., K.D., Thompson, P.J. Physiological
Unger, W.W.J., Fernandes, R.J., Bruijns, concentrations of transforming growth
S.C.M., Laban, S., Roep, B.O., ‘t Hart, factor [beta]1 selectively inhibit
B.A., van Kooyk, Y. TLR triggering on human dendritic cell function. Int.

CHAPTER 3
tolerogenic dendritic cells results in Immunopharmacol. 2007; 7:1924-1933.
TLR2 up-regulation and a reduced 30. Xiao, B.G., Zhu, W.H., Lu, C.Z. The
proinflammatory immune program. J. presence of GM-CSF and IL-4 interferes
Immunol. 2009; 183:2984-2994. with effect of TGF-[beta]1 on antigen
22. Naranjo-Gomez, M., Raich-Regue, D., presenting cells in patients with multiple
Onate, C., Grau-Lopez, L., Ramo-Tello, sclerosis and in rats with experimental
C., Pujol-Borrell, R., Martinez-Caceres, autoimmune encephalomyelitis. Cell.
E., Borras, F.E. Comparative study Immunol. 2007; 249:30-36.
of clinical grade human tolerogenic 31. Torres-Aguilar, H., Aguilar-Ruiz, S.R.,
dendritic cells. J. Transl. Med. 2011; Gonzalez-Perez, G., Munguia, R., Bajana,
9:89. S., Meraz-Rios, M.A., Sanchez-Torres, C.
23. Penna, G., Adorini, L. 1 Alpha,25- Tolerogenic dendritic cells generated
dihydroxyvitamin D3 inhibits with different immunosuppressive
differentiation, maturation, activation, cytokines induce antigen-specific
and survival of dendritic cells leading to anergy and regulatory properties in
impaired alloreactive T cell activation. J. memory CD4+ T cells. J. Immunol. 2010;
Immunol. 2000; 164:2405-2411. 184:1765-1775.
24. Piemonti, L., Monti, P., Sironi, M., 32. Forster, R., Davalos-Misslitz, A.C., Rot,
Fraticelli, P., Leone, B.E., Dal Cin, E., A. CCR7 and its ligands: balancing
Allavena, P., Di Carlo, V. Vitamin D3 immunity and tolerance. Nat. Rev.
affects differentiation, maturation, and Immunol. 2008; 8:362-371.
function of human monocyte-derived 33. Fu, F., Li, Y., Qian, S., Lu, L., Chambers,
dendritic cells. J. Immunol. 2000; F., Starzl, T.E., Fung, J.J., Thomson,
164:4443-4451. A.W. Costimulatory molecule-
25. Turnquist, H.R., Raimondi, G., Zahorchak, deficient dendritic cell progenitors
A.F., Fischer, R.T., Wang, Z., Thomson, (MHC class II+, CD80dim, CD86-
A.W. Rapamycin-conditioned dendritic ) prolong cardiac allograft survival in
cells are poor stimulators of allogeneic nonimmunosuppressed recipients.
CD4+ T cells, but enrich for antigen- Transplantation 1996; 62:659-665.
specific Foxp3+ T regulatory cells and 34. Rastellini, C., Lu, L., Ricordi, C.,
promote organ transplant tolerance. J. Starzl, T.E., Rao, A.S., Thomson, A.W.
Immunol. 2007; 178:7018-7031. Granulocyte/macrophage colony-
26. Fedoric, B., Krishnan, R. Rapamycin stimulating factor-stimulated hepatic
downregulates the inhibitory dendritic cell progenitors prolong
receptors ILT2, ILT3, ILT4 on human pancreatic islet allograft survival.
dendritic cells and yet induces T cell Transplantation 1995; 60:1366-1370.

73
 CHAPTER 3

35. Farquhar, C.A., Paterson, A.M., Cobbold, 46. Ulrich, J.T., Myers, K.R. Monophosphoryl
S.P., Garcia, R.H., Fairchild, P.J., lipid A as an adjuvant. Past experiences
Yates, S.F., Adams, E., Saunders, N.J., and new directions. Pharm. Biotechnol.
Waldmann, H., Nolan, K.F. Tolerogenicity 1995; 6:495-524.
is not an absolute property of a dendritic 47. Vlad, G., Chang, C.C., Colovai, A.I.,
cell. Eur. J. Immunol. 2010; 40:1728- Berloco, P., Cortesini, R., Suciu-Foca,
1737. N. Immunoglobulin-like transcript 3:
36. Steinman, R.M., Hawiger, D., A crucial regulator of dendritic cell
Nussenzweig, M.C. Tolerogenic function. Hum. Immunol. 2009; 70:340-
dendritic cells. Annu. Rev. Immunol. 344.
2003; 21:685-711. 48. Chui, C.S., Li, D. Role of immunolglobulin-
37. Thomson, A.W., Robbins, P.D. Tolerogenic like transcript family receptors and their
dendritic cells for autoimmune disease ligands in suppressor T-cell-induced
and transplantation. Ann. Rheum. Dis. dendritic cell tolerization. Hum.
2008; 67:90-96. Immunol. 2009; 70:686-691.
38. Jonuleit, H., Schmitt, E., Steinbrink, K., 49. Fife, B.T., Pauken, K.E., Eagar, T.N.,
Enk, A.H. Dendritic cells as a tool to Obu, T., Wu, J., Tang, Q., Azuma,
induce anergic and regulatory T cells. M., Krummel, M.F., Bluestone, J.A.
Trends Immunol. 2001; 22:394-400. Interactions between PD-1 and PD-
39. Yamazaki, S., Inaba, K., Tarbell, K.V., L1 promote tolerance by blocking the
Steinman, R.M. Dendritic cells expand TCR-induced stop signal. Nat. Immunol.
antigen-specific Foxp3+CD25+CD4+ 2009; 10:1185-1192.
CHAPTER 3

regulatory T cells including suppressors 50. Francisco, L.M., Salinas, V.H., Brown, K.E.,
of alloreactivity. Immunol. Rev. 2006; Vanguri, V.K., Freeman, G.J., Kuchroo,
212:314-329. V.K., Sharpe, A.H. PD-L1 regulates
40. Roncarolo, M.G., Battaglia, M. the development, maintenance, and
Regulatory T-cell immunotherapy function of induced regulatory T cells. J.
for tolerance to self antigens and Exp. Med. 2009; 206:3015-3029.
alloantigens in humans. Nat. Rev. 51. Keir, M.E., Butte, M.J., Freeman, G.J.,
Immunol. 2007; 7:585-598. Sharpe, A.H. PD-1 and its ligands in
41. Vignali, D.A.A., Collison, L.W., Workman, tolerance and immunity. Annu. Rev.
C.J. How regulatory T cells work. Nat. Immunol. 2008; 26:677-704.
Rev. Immunol. 2008; 8:523-532. 52. Jonuleit, H., Kuhn, U., Muller, G.,
42. Shevach, E.M. Mechanisms of Foxp3+ Steinbrink, K., Paragnik, L., Schmitt, E.,
T regulatory cell-mediated suppression. Knop, J., Enk, A.H. Pro-inflammatory
Immunity 2009; 30:636-645. cytokines and prostaglandins induce
43. ten Brinke, A., Karsten, M.L., Dieker, M.C., maturation of potent immunostimulatory
Zwaginga, J.J., Vrielink, H., Marieke van dendritic cells under fetal calf serum-
Ham, S. Generation of dendritic cells for free conditions. Eur. J. Immunol. 1997;
immunotherapy is minimally impaired 27:3135-3142.
by granulocytes in the monocyte 53. Sordi, V., Bianchi, G., Buracchi, C.,
preparation. Immunobiology 2006; Mercalli, A., Marchesi, F., D’Amico,
211:633-640. G., Yang, C.H., Luini, W., Vecchi, A.,
44. ten Brinke, A., Karsten, M.L., Dieker, Mantovani, A., Allavena, P., Piemonti, L.
M.C., Zwaginga, J.J., van Ham, Differential effects of immunosuppressive
S.M. The clinical grade maturation drugs on chemokine receptor CCR7 in
cocktail monophosphoryl lipid A plus human monocyte-derived dendritic cells:
IFN[gamma] generates monocyte- selective upregulation by rapamycin.
derived dendritic cells with the capacity Transplantation 2006; 82:826-834.
to migrate and induce Th1 polarization. 54. Morelli, A.E., Thomson, A.W. Tolerogenic
Vaccine 2007; 25:7145-7152. dendritic cells and the quest for
45. Anderson, A.E., Swan, D.J., Sayers, B.L., transplant tolerance. Nat. Rev. Immunol.
Harry, R.A., Patterson, A.M., von Delwig, 2007; 7:610-621.
A., Robinson, J.H., Isaacs, J.D., Hilkens, 55. Kalinski, P., Vieira, P.L., Schuitemaker,
C.M.U. LPS activation is required J.H.N., de Jong, E.C., Kapsenberg, M.L.
for migratory activity and antigen Prostaglandin E2 is a selective inducer of
presentation by tolerogenic dendritic interleukin-12 p40 (IL-12p40) production
cells. J. Leukoc. Biol. 2009; 85:243-250. and an inhibitor of bioactive IL-12p70

74
 COMPARATIVE STUDY OF HUMAN CLINICAL-APPLICABLE DC

heterodimer. Blood 2001; 97:3466- 60. Dhodapkar, M.V., Steinman, R.M.,

3469. Krasovsky, J., Munz, C., Bhardwaj, N.
56. Beinhauer, B.G., McBride, J.M., Graf, Antigen-specific inhibition of effector
P., Pursch, E., Bongers, M., Rogy, M., T cell function in humans after injection
Korthauer, U., de Vries, J.E., Aversa, of immature dendritic cells. J. Exp. Med.
G., Jung, T. Interleukin 10 regulates 2001; 193:233-238.
cell surface and soluble LIR-2 (CD85d) 61. Steinbrink, K., Graulich, E., Kubsch, S.,
expression on dendritic cells resulting in Knop, J., Enk, A.H. CD4+ and CD8+
T cell hyporesponsiveness in vitro. Eur. J. anergic T cells induced by interleukin-
Immunol. 2004; 34:74-80. 10-treated human dendritic cells display
57. Manavalan, J.S., Rossi, P.C., Vlad, antigen-specific suppressor activity.
G., Piazza, F., Yarilina, A., Cortesini, Blood 2002; 99:2468-2476.
R., Mancini, D., Suciu-Foca, N. High 62. Roncarolo, M.G., Gregori, S., Battaglia,
expression of ILT3 and ILT4 is a general M., Bacchetta, R., Fleischhauer, K.,
feature of tolerogenic dendritic cells. Levings, M.K. Interleukin-10-secreting
Transpl. Immunol. 2003; 11:245-258. type 1 regulatory T cells in rodents and
58. Voigtlander, C., Rossner, S., Cierpka, humans. Immunol. Rev. 2006; 212:28-50.
E., Theiner, G., Wiethe, C., Menges, 63. Taner, T., Hackstein, H., Wang, Z., Morelli,
M., Schuler, G., Lutz, M.B. Dendritic A.E., Thomson, A.W. Rapamycin-treated,
cells matured with TNF can be further alloantigen-pulsed host dendritic cells
activated in vitro and after subcutaneous induce ag-specific T cell regulation and
injection in vivo which converts their prolong graft survival. Am. J. Transplant.
tolerogenicity into immunogenicity. J. 2005; 5:228-236.

CHAPTER 3
Immunother. 2006; 29:407-415. 64. Hackstein, H., Taner, T., Zahorchak, A.F.,
59. Levings, M.K., Gregori, S., Tresoldi, E., Morelli, A.E., Logar, A.J., Gessner, A.,
Cazzaniga, S., Bonini, C., Roncarolo, Thomson, A.W. Rapamycin inhibits IL-
M.G. Differentiation of Tr1 cells by 4-induced dendritic cell maturation in
immature dendritic cells requires IL- vitro and dendritic cell mobilization and
10 but not CD25+CD4+ Tr cells. Blood function in vivo. Blood 2003; 101:4457-
2005; 105:1162-1169. 4463.

75
 4
CHAPTER
Inhibition of TNFRII signalling
by anti-TNFα primes naïve CD4 +
T cells towards IL-10 + cells with
strong regulatory phenotype
and function
Martine A. Boks, Judith R. Kager-Groenland,
S. Marieke van Ham and Anja ten Brinke

Manuscript in preparation
 CHAPTER 4

Abstract
TNFα is a potent pro-inflammatory cytokine playing a pivotal role in several
autoimmune diseases. Neutralizing TNFα inhibits T cell proliferation and IFNγ
production, and enhances suppressive capacity of regulatory T cells (Treg). Little is
known about the effects of TNFα blocking agents on naïve T cell differentiation. Here,
we report that neutralizing TNFα during priming of naïve CD4+ T cells by dendritic
cells (DC) favours development of IL-10+ T helper (Th) cells at the expense of IFNγ-
induction. TNFα inhibits IL-10 via TNFRII, which becomes expressed after naïve
T cell activation. While initial CD4+ T cell activation was not affected, neutralization
of TNFα negatively affected later stages of T cell priming by counteracting full
T cell activation and increasing cell death. Whole genome gene expression analysis
revealed a regulatory gene profile of anti-TNFα-treated T cells. Indeed, neutralizing
TNFα during naïve T cell priming enhanced the suppressive function of the anti-
TNFα-treated T cells. Taken together, inhibition of TNFα–TNFRII interaction affects
late stage effector T cell development and shifts the balance of Th cell differentiation
towards immune regulation, which may be beneficial in TNFα blocking therapies.
CHAPTER 4

78
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

Introduction
Tumor necrosis factor (TNF)α is a pro-inflammatory cytokine with potent effects
on inflammation and immune responses. It initiates the defence response to local
injury, recruits leukocytes and drives the production of multiple pro-inflammatory
cytokines.1 In chronic inflammation, TNFα can be a key factor for sustenance and
amplification of the inflammatory process in tissues as observed in several immune-
mediated inflammatory diseases (IMID) such as rheumatoid arthritis (RA), ankylosing
spondylitis, psoriatic arthritis, multiple sclerosis and inflammatory bowel diseases.2-4
Treatment of patients with TNFα blocking agents results in significant clinical
benefit. Currently, 5 types of anti-TNFα agents have been approved for clinical
use; the anti-TNFα monoclonal antibodies infliximab, adalimumab and golimumab;
etanercept (a fusion protein of TNFR and Fc fragment); and certolizumab pegol
(a pegylated Fab fragment of an anti-TNFα antibody).
Inhibition of TNFα affects various cell types and mediators involved in the
pathogenesis of RA.4 Treatment of RA patients with TNFα blocking agents reduces
the numbers of infiltrating granulocytes, T cells, B cells and macrophages in affected
joints by reduced chemokine expression in the synovium.5 Anti-TNFα agents also
reduce pro-inflammatory cytokine production, i.e. IL-6, TNFα, IL-1β and IL-1α, both
in serum and the synovium.6,7
In addition to these localised effects, anti-TNFα therapy affects the T cell arm of
the adaptive immune system in autoimmunity. Both in vitro and in RA patients it was
shown that anti-TNFα agents reduce T cell activation and proliferation by inhibiting
dendritic cell (DC) maturation, resulting in reduced co-stimulatory molecule
expression and cytokine production.8-10 Several studies show that regulatory T cells

CHAPTER 4
(Treg) of RA patients are defective and fail to suppress inflammation. Treatment
of these patients with TNFα blocking agents induces and/ or expands Treg, and
recovers their suppressive function.11-13 In addition, TNFα blocking agents cause
changes in T cell polarization profiles in these patients.14-16 As these data were
obtained within an autoimmune background, the direct effect of TNFα neutralization
on T cell function remains unclear. In vitro experiments with cells from healthy
donors showed that TNFα co-stimulates T cells and enhances proliferation and
IFNγ production.17-19 In addition, TNFα inhibits the suppressive capacity of Treg.12,20
In line with these studies, neutralizing TNFα inhibits T cell proliferation and IFNγ
production 17,21, and enhances suppression by Treg.20 However, it was also shown
that neutralizing TNFα during priming of naïve CD4+ T cells reduced Treg induction
and suppressive function.22 Thus, the precise role of TNFα in T cell differentiation
and Treg function is unclear.
In this study we show that neutralization of TNFα during naïve CD4+ T cell priming
induced T helper (Th) cell polarization towards IL-10 expressing cells, via inhibition
of TNFα–TNFRII interaction. Anti-TNFα inhibited sustained T cell activation after
TNFRII upregulation. Micro array analysis revealed induction of a regulatory gene
profile, which corresponded with enhanced suppressive function of the T cells.
Taken together, this implicates that inhibition of TNFRII signalling by anti-TNFα

79
 CHAPTER 4

therapeutics counteracts late stage T cell priming and shifts CD4+ effector T cell
function towards downmodulation of immune responses.

Materials and Methods

Antibodies and reagents
CellGro DC serum-free medium, GM-CSF, IL-4, IL-1β and TNFα were all obtained from
CellGenix (Freiburg, Germany). PGE2 was purchased from Sigma-Aldrich (Steinheim,
Germany) and IL-10 from PeproTech (Rocky Hill, USA). Penicillin and streptomycin
were obtained from Gibco (Merelbeke, Beldium). For co-cultures Iscove’s Modified
Dulbecco’s Medium (IMDM, Bio Whittaker, Verviers, Belgium) with 10% fetal calf
serum (Bodinco, Alkmaar, The Netherlands) was used as described.23 Adalimumab
(Humira; Abbott), infliximab (Remicade; Schering-Plough) and etanercept (Enbrel;
Wyeth Pharmaceuticals) were used as anti-TNFα agents at 10 μg/ml. As isotype
controls, an irrelevant monoclonal antibody directed against FelD1 (cat allergen)
or the F(ab)2 fragments hereof were used (Sanquin Reagents, Amsterdam, The
Netherlands).
Fluorochrome-labelled antibodies against 4-1BB, CD25, HLA-DR, OX40, TNFRII,
IFNγ and IL-4 were purchased from Becton Dickinson (BD Biosciences, San Jose,
USA). Anti-IL-10 was obtained from Diaclone (via Sanquin Reagents), anti-CD69 from
Sanquin Reagents, and anti-GITR from R&D Systems (Minneapolis, USA). Isotype-
matched controls from Dako (Glostrup, Denmark) or BD Biosciences were used.

Isolation and culture of monocyte-derived dendritic cells

CHAPTER 4

Monocytes were isolated from fresh aphaeresis material of healthy volunteers (after
informed consent) by using the ElutraTM cell separation system (Gambro, Lakewood,
USA). Monocytes were cultured at 1.5 x 106 cells/3 ml in 6 well plates (Nunc, Roskilde,
Denmark) in serum-free CellGro medium supplemented with IL-4 (800 IU/ml),
GM-CSF (1000 IU/ml), penicillin (100 IU/ml) and streptomycin (100 μg/ml). After 6
days, the immature DC were matured with IL-10 (40 ng/ml; 1 hour pre-incubation)
together with IL-1β (10 ng/ml), TNFα (10 ng/ml) and PGE2 (1 µg/ml) for 2 more
days to generate IL-10 tolerogenic (t)DC, as described previously.24 Alternatively,
DC were matured with IL-1β, TNFα and PGE2 without IL-10 to generate mature
immunoactivatory (m)DC.

T cell-dendritic cell co-culture and intracellular cytokine staining

Naïve CD4+CD45RA+CD45RO- T cells were isolated as described previously.24
Allogeneic IL-10 tDC were co-cultured with 1 x 105 CD4+ naïve T cells (1:5) or with
anti-CD3/ anti-CD28 stimulating antibodies (1 μg/ml and 0.05 μg/ml, respectively;
Sanquin Reagents) for 13 to 15 days. Anti-TNFα (adalimumab F(ab)2, unless indicated
otherwise), isotype control F(ab)2 (both 10 μg/ml) or TNFα (10 ng/ml) were added at
start of co-culture, or at indicated time points. Antibodies against TNFRI or TNFRII
(clones 16803 and 22221; R&D Systems) were used at 10 μg/ml. After co-culture,

80
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

T cells were harvested and restimulated with PMA and ionomycin (100 ng/ml and
1 μg/ml, respectively; Sigma) for 5 hours in the presence of Brefeldin A (10 μg/ml;
Sigma). The production of IL-10, IFNγ and IL-4 was detected by intracellular FACS
staining by first fixating the cells with a 4% paraformaldehyde solution and
subsequently staining intracellularly with appropriate antibodies in a saponin
solution (0.5% saponin and 1% BSA in PBS). Cytokine expression was measured on
an LSRII flow cytometer and analyzed with FACS Diva software (BD Biosciences).
For phenotyping, DC-primed T cells were harvested on indicated time points
and incubated with specific monoclonal antibodies or appropriate isotype-matched
controls for 30 min. DAPI (4’,6-diamidino-2-phenylindole; Sigma-Aldrich) was
added to the cells before analysis to assess cell viability and exclude dead cells from
analysis. Cells were measured on an LSRII flow cytometer. For proliferation analysis,
CD4+ naïve T cells were labelled with 0.5 μM CFSE (Invitrogen, Eugene, Oregon,
USA) for 15 minutes at RT prior to co-culture.
Cytokines produced during co-culture were measured after 13 days of culture, or
on indicated time points. IL-10, IFNγ and TNFα levels were determined by enzyme-
linked immunosorbent assay (ELISA) using the PeliKine-compact ELISA kit (Sanquin
Reagents).

Micro array analysis

Naïve CD4+ T cells were CFSE labelled and co-cultured for 13 days with IL-10 tDC
in the presence or absence of anti-TNFα F(ab)2. After 13 days, the CFSElow T cells
were FACS sorted on a FACS Aria (BD Biosciences). Samples were generated from
three independent donors. The Human Sentrix-6 beadchip (Illumina) was used for

CHAPTER 4
the whole genome gene expression analysis. The data was log2 transformed and
normalized between arrays using robust spline normalization method. Triplicate
samples of anti-TNFα-treated T cells were compared to non-treated T cells by
LIMMA analysis. Genes were further filtered and analyzed using log2 fold change
calculations. The entire dataset described here will be deposited in the Gene
Expression Omnibus database.

Suppression assay
1 x 106 allogeneic CD4+ naïve T cells were co-cultured with IL-10 tDC (1:10) in
24 well plates (Nunc) for 6 days in presence of anti-TNFα or TNFα. DC-primed
T cells were harvested, washed and evaluated for their suppressive capacity as
described previously.23 Briefly, CD4+ memory responder T cells (from the same T cell
donor) were labelled with 0.5 μM CFSE for 15 minutes at RT, and primed T cells
were labelled with 6 μM PKH26 (Sigma-Aldrich) for 5 minutes at RT and irradiated
(30 Gray) to prevent convergence with proliferated CFSE-negative cells. Primed
T cells were incubated with 5 x 104 responder T cells (2:1, 1:1, 1:2) and mDC (from
the same DC donor; 1:50 ratio with responder T cells). Cells were cultured in 96
well round bottom plates (Greiner Bio-One, Frickenhausen, Germany) for 6 days in
culture medium. Proliferation of responder T cells was measured on an LSRII flow
cytometer in the presence of DAPI to exclude dead cells.

81
 CHAPTER 4

Statistical analysis
Data are expressed as mean + standard error of the mean (SEM). Statistical
significance was analyzed using two-way Analysis Of Variance (ANOVA) (Figure 1),
one-way ANOVA (Figure 1, 3, 6) with Dunnett’s or Bonferroni post test, or paired t
test (Figure 4) in Graphpad Prism 5.01 software (San Diego, USA).

Results
TNFα neutralization during priming of naïve CD4+ T cells induces
IL-10+ T cells
Anti-TNFα biologicals are widely used to treat various autoimmune diseases. In order to
investigate the role of TNFα in T cell polarization, we blocked TNFα with adalimumab,
a human monoclonal antibody against TNFα, during priming of naïve CD4+ T cells. We
have previously shown that IL-10-generated tolerogenic (t)DC are excellent inducers
of regulatory T cells (Treg) that strongly suppress T cell reactivity.23,24 As IL-10 tDC
produce high amounts of the pro-inflammatory cytokine TNFα, we used these DC
to investigate the effects of anti-TNFα on naïve T cell priming. Anti-TNFα increased
IL-10 and decreased IFNγ production, indicating polarization of T cells towards IL-10
production (Figure 1A). Since cytokines can be consumed during culture and because
IL-10 tDC produce IL-10 themselves 23,24, we next investigated T cell specific cytokine
expression using intracellular cytokine staining. Anti-TNFα significantly increased
the percentage of IL-10+ T cells and decreased IFNγ+ T cells, without affecting IL-4+
T cell polarization (Figure 1B and C). Addition of exogenous TNFα showed a reverse
CHAPTER 4

tendency towards a decrease in IL-10+ and IL-4+ T cells (Figure 1B and D).
Neutralization of TNFα may affect maturation and function of DC 9,10, whereby
T cell activation or polarization may be affected. Although we did not observe an
effect of TNFα neutralization on IL-10 tDC co-stimulatory molecule expression or
cytokine production (data not shown), we investigated whether anti-TNFα had
similar effects on DC-independent CD4+ T cell polarization. Naïve T cells were
primed with anti-CD3/ anti-CD28 stimulating antibodies, or with IL-10 tDC, in the
presence or absence of anti-TNFα. When T cells were stimulated by antibodies,
anti-TNFα also increased the percentage of IL-10+ T cells (Figure 2A). The magnitude
of IL-10 induction was comparable to that observed in the IL-10 tDC-primed T cells
(Figure 2B), although the tolerogenic DC induced higher basal levels of IL-10+
T cells (Figure 2A). These results demonstrate that TNFα neutralization directly
affects T cell polarization. Thus, blocking TNFα during priming of naïve CD4+ T cells
polarizes T cells towards IL-10 production and away from IFNγ production.

IL-10+ T cell priming is mediated by inhibiting the interaction

between TNFα and TNFRII on T cells
TNFα-producing cells express transmembrane (m)TNFα which can act both as a ligand
to cross-link TNF receptors and as a receptor to transduce signals into the cells.25,26
Adalimumab, infliximab and etanercept bind with similar affinities to soluble (s)

82
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

A IL-10 IFNγγ
200 500
* **
150 400

pg/ml 300

pg/ml
100
200
50
100

0 0
-

Fα

α
F

F
TN

TN

TN
TN
ti-

ti -
an

an
B - anti-TNFα isotype ctrl TNFα
IL-10

C IL-10 IFNγγ IL-4

15 ** ** 50 20
% IL-10 positive T cells

% IFNγγ positive T cells

% IL-4 positive T cells

*** ***
40
15
10
30
10
20
5

CHAPTER 4
5
10

0 0 0
-

rl

rl

rl
-
α

α
Fα
ct

ct

ct
F

F
TN
TN

TN
e
e

e
yp
yp

yp
ti-
ti-

ti -
ot
an
ot

ot
an

an
is
is

is

D IL-10 IFNγγ IL-4

8 50 15
% IL-10 positive T cells

% IFNγγ positive T cells

% IL-4 positive T cells

40
6
10
30
4
20
5
2
10

0 0 0
- TNFα
α - TNFα
α - TNFα
α

Figure 1. Anti-TNFα polarizes T cells towards IL-10 and away from IFNγ. Naïve CD4+
T cells were co-cultured with IL-10 tDC in the presence or absence of anti-TNFα F(ab)2, an
isotype control F(ab)2 or exogenous TNFα. (A) IL-10 and IFNγ production was measured in
the supernatant after 2 weeks of co-culture. Mean + SEM of 8 independent experiments
are depicted. (B) T cells were primed by IL-10 tDC for 2 weeks, subsequently restimulated
with PMA and ionomycin and stained intracellularly for IL-10. A representative dotplot of 11
independent experiments is shown. (C, D) Percentage of IL-10, IFNγ and IL-4 positive T cells
determined by intracellular staining of restimulated primed T cells. Mean + SEM of 14 (C) or
11 (D) independent experiments. *p≤0.05, **p≤0.01, ***p≤0.001

83
 CHAPTER 4

A α CD3α
α CD28 IL-10 tDC B

Relative % IL-10 positive T cells

2.0 30 3
% IL-10 positive T cells

% IL-10 positive T cells

1.5
20 2

1.0

10 1
0.5

0.0 0 0
-
Fα rl -
Fα rl
ct ct

28

C
tD
N N

D
i-T pe i-T pe

0
t y t y

-1
3α
t t
an o an o

IL
is is

D
αC
C
25 anti-TNFα
α
% IL-10 positive T cells

20

15 Figure 2. Anti-TNFα induces IL-10+

T cell polarization through inhibition
10 of TNFRII signalling. (A) Naïve CD4+
T cells were co-cultured for 2 weeks with
5 IL-10 tDC or with anti-CD3/anti-CD28
stimulating antibodies in presence of
0
anti-TNFα F(ab)2 or isotype control F(ab)2.
T cells were restimulated and stained for
ab

l
-

t
)2

)2

r
ep

ct
ab

ab
im

intracellular IL-10 expression. Percentage

pe
F(

F(
er
x
fli

y
an
ab

rl

ot

IL-10 positive T cells of a representative

ct
in

et
um

is
pe

experiment out of 3 independent

im

ty
al

experiments is shown. (B) Relative

is
ad

D induction of IL-10 positive T cells (mean

20 + SEM) of 3 independent experiments
% IL-10 positive T cells

is shown. (C) T cells were primed by

15 IL-10 tDC in presence of adalimumab
CHAPTER 4

F(ab)2, infliximab, etanercept, or istoype

10
controls. Mean + SEM of 3 independent
experiments. (D) T cells were primed
by IL-10 tDC in presence of exogenous
5 TNFα, anti-TNFRI, anti-TNFRII, or isotype
control. Mean + SEM of 3 independent
0 experiments is shown.
- II I rl
Fα FR FR ct
TN N N pe
ti-
T it -T y
an ot
an is

TNFα 27 and mTNFα.27-29 In contrast, only the monoclonal antibodies adalimumab

and infliximab induce outside-to-inside signalling via mTNFα.26,28 To study the role
of mTNFα in IL-10 induction in T cells, we cultured naïve T cells with IL-10 tDC
and the different anti-TNFα agents. All three TNFα inhibitors induced IL-10+ T cells
to the same extent (Figure 2C), implying that anti-TNFα agents induced IL-10 in
T cells by neutralization of TNFα and interference with TNFR signalling rather than
by signalling via mTNFα.
To investigate whether TNFα influences T cell polarization via TNFRI or TNFRII
signalling, we co-cultured naïve T cells with IL-10 tDC and TNFα or a stimulating
TNFRII antibody. Triggering TNFRII with TNFα or the stimulating antibody decreased

84
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

the IL-10+ T cells to the same extent (Figure 2D). An anti-TNFRI antibody did not
affect IL-10+ T cell polarization. Taken together, our data suggest that inhibition of
TNFRII signalling by TNFα during T cell priming induces IL-10 producing T cells.

TNFRII signalling counteracts IL-10+ T cell priming

To investigate in which phase of T cell priming TNFα affects T cell polarization,
we cultured naïve T cells with IL-10 tDC and added anti-TNFα at different time
points during culture. Addition of anti-TNFα up to day 5 of co-culture increased
IL-10+ T cell polarization (Figure 3A). Addition of anti-TNFα at later time points did
not affect IL-10+ T cell polarization. Analysis of TNFα production at the different
time points showed that TNFα was produced in detectable amounts from day 3
of culture (Figure 3B). Flow cytometric analysis of TNFRII expression showed that
TNFRII became expressed on primed T cells from day 5 onwards (Figure 3C). TNFRII
expression was not altered by presence or absence of TNFα (Figure 3C). These data
show that in order to induce IL-10 polarization in T cells, TNFRII signalling must be
prevented by neutralizing TNFα as soon as the receptor becomes upregulated.

TNFα neutralization reduces T cell activation status by affecting

late stage CD4+ T cell priming
As TNFRII signalling can only affect T cell polarization in the later phases of
T cell priming, we investigated the effects of anti-TNFα on T cell proliferation and
activation during the different phases of CD4+ T cell priming. Naïve T cells were
CFSE labelled and co-cultured with IL-10 tDC in presence or absence of anti-TNFα.
Whereas anti-TNFα did not affect T cell proliferation during the first 7 days of naïve

CHAPTER 4
CD4+ T cell priming (Figure 4A), the number of T cells in the proliferating gate was
strongly reduced at later stages (Figure 4B). As anti-TNFa did not affect the rate of
T cell proliferation (Figure 4C and D) this showed that blocking TNFα reduced CD4+
T cell survival during the second phase of T cell priming.
Next, we analyzed the activation status of proliferated CFSElow T cells at day 7 of
co-culture by measuring expression of classical activation markers CD69, CD25 and
HLA-DR, and other cell surface molecules highly expressed after T cell activation,
i.e. 4-1BB, GITR and OX40. Whereas early T cell activation marker CD69 was not
affected by anti-TNFα, expression of the other markers, that become upregulated at
later stages of T cell priming, was reduced when TNFα was neutralized (Figure 4E).
Taken together, these data demonstrate that neutralization of TNFα affects the later
stages of CD4+ T cell priming, which corresponds to the upregulation of TNFRII.

Neutralizing TNFα induces a regulatory gene profile in T cells

To further investigate the effects of TNFα neutralization on T cells, we performed
whole genome gene expression analysis on T cells that were primed by IL-10 tDC in
presence or absence of anti-TNFα. Since neutralization of TNFα during CD4+ T cell
priming resulted in increased IL-10+ T cells, reduced IFNγ+ T cells and a reduced
activation status of the T cells, we analyzed the gene expression of various Treg
signature genes.30-33 Many Treg-associated gene transcripts were highly upregulated

85
 CHAPTER 4

A B
% IL-10 positive T cells 8 300

6 ** **

α (pg/ml)
** 200

TNFα
100
2

0 0
- 0 2 5 7 9 12 2 3 4 5 7 9 13
Day of anti-TNFα
α addition Time (days)

C
250
-
anti-TNFα
TNFα
200
TNFRII (relative MFI)

150

100

50

0
CHAPTER 4

0 2 5 7 9 13
Time (days)

Figure 3. Neutralizing TNFα before significant TNFRII signalling increases IL-10+ T cells.
Naïve CD4+ T cells were co-cultured for 2 weeks with IL-10 tDC in the presence or absence of
anti-TNFα F(ab)2 or exogenous TNFα. (A) Anti-TNFα was added on different days after start of
co-culture. At day 13, T cells were restimulated and analyzed for intracellular IL-10 expression.
Mean + SEM of 3 independent experiments are depicted. (B) Endogenous TNFα production
was measured in the supernatant harvested at different days of co-culture. Mean + SEM of 3
independent experiments. (C) Relative mean fluorescence intensity (MFI) of TNFRII expression
on primed T cells during co-culture. MFI (mean + SEM) relative to expression on day 13 of
T cells primed by IL-10 tDC only is shown for 3 independent experiments. **p≤0.01

after neutralization of TNFα, i.e. galectin-3 (LGALS3), LAG-3, IL-10, legumain (LGMN),
GARP (LRRC32), CTLA-4 and PD-1 (PDCD1) (Figure 5). Foxp3, TGFβ and CD127
(IL-7R) expression were not affected (Figure 5). CD25 (IL-2Rα) and GITR (TNFRSF18)
are transiently expressed by activated T cells, whereas high sustained expression is
associated with a Treg phenotype. CD25 gene expression was not affected. GITR
gene expression was downmodulated by anti-TNFα, in line with the GITR protein
levels detected on day 7 (Figure 5 and 4E). These data show that neutralization of
TNFα induces an overall regulatory gene profile in primed CD4+ T cells.

86
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

A B Day 13
60 100
- ***

% CFSElow T cells

% CFSElow T cells
anti-TNFα
40 80

20 60

0 40
3 4 5 7 13 - anti-TNFα
α
Time (days)

C D
800

MFI CFSElow T cells

600

400
CFSElow:
- 78.5% 200
anti-TNFα 57.9%
Isotype ctrl 77%
0
- anti-TNFα
α

E
CD69 CD25 HLA-DR
2500 4000 18000

2000
3000
*
12000
1500
MFI

MFI

MFI

2000

CHAPTER 4
1000
6000
1000
500

0 0 0
- anti-TNFα
α - anti-TNFα
α - anti-TNFα
α

4-1BB GITR OX40

400 2500 3000

2000
300
2000
1500 **
MFI

MFI
MFI

200 * *
1000
1000
100
500

0 0 0
- anti-TNFα
α - anti-TNFα
α - anti-TNFα
α

Figure 4. Anti-TNFα affects late stage of CD4+ T cell priming. Naïve CD4+ T cells were CFSE
labelled and co-cultured for 2 weeks with IL-10 tDC in the presence or absence of anti-TNFα
F(ab)2. T cell proliferation (% CFSElow T cells) on day 13 of co-culture is shown, unless indicated
otherwise. (A) T cell proliferation during co-culture. A representative experiment out of 3
independent experiments is shown. (B) T cell proliferation of 10 independent experiments.
(C) Histograms of CFSE profiles of primed T cells of a representative experiment out of 10
independent experiments is shown. (D) CFSE mean fluorescence intensity (MFI) of proliferated
(CFSElow) T cells. Mean + SEM of 9 independent experiments. (E) Expression of activation
markers CD69, CD25, HLA-DR, 4-1BB, GITR and OX40 on CFSElow primed T cells on day 7 of
co-culture. MFI (mean + SEM) of 5 or 4 (OX40) independent experiments is depicted. *p≤0.05,
**p≤0.01, ***p≤0.001

87
 CHAPTER 4

LGALS3 LAG3 IL10 LGMN

1000 1000 10000
100000
probe intensity

probe intensity

probe intensity
probe intensity
100 1000
100 10000
10 100

FC: 2.91 FC: 2.87 FC: 2.61 FC: 2.54

10 1000 1 10
- anti-TNFα
α - anti-TNFα
α - anti-TNFα
α - anti-TNFα
α

LRRC32 CTLA4 PDCD1 FOXP3

100 1000 100
10000

probe intensity
probe intensity

probe intensity
probe intensity

1000 100 10

FC: 2.12 FC: 1.94 FC: 1.85 FC: 1.21

10 100 10 1
- anti-TNFα
α - anti-TNFα
α - anti-TNFα
α - anti-TNFα
α

IL2RA TGFB1 IL7R TNFRSF18

1000 100 1000
10000
probe intensity

probe intensity
probe intensity

probe intensity
10 1000

FC: 1.04 FC: -1.05 FC: -1.31 FC: -2.57

100 1 100 100
- anti-TNFα
α - anti-TNFα
α - anti-TNFα
α - anti-TNFα
α

Figure 5. Neutralizing TNFα induces a regulatory gene transcript profile in CD4+ primed
T cells. Naïve CD4+ T cells were CFSE labelled and co-cultured for 2 weeks with IL-10 tDC
in the presence or absence of anti-TNFα F(ab)2. A) After 13 days, the CFSElow T cells were
FACS sorted. Micro arrays were performed on samples of three independent donors. With
LIMMA analysis a mean log2 fold change of the log2 probe intensities was calculated for each
gene. Genes were subsequently filtered on fold change. Micro array transcript intensities of
three independent donors are plotted for selected gene transcripts. Numbers represent the
mean fold change (FC) in gene expression in anti-TNFα-treated T cells versus non-treated
T cells. LGALS3; galectin-3, LGMN; legumain, LRRC32; GARP, PDCD1; PD-1, IL2RA; CD25,
CHAPTER 4

IL7R; CD127, TNFRSF18; GITR.

Neutralizing TNFα enhances suppressive function of primed

T cells
Because neutralization of TNFα induces IL-10 expression and a regulatory gene
profile, we next investigated how this affects the suppressive capacity of IL-10 tDC-
primed T cells. Naïve T cells were primed by IL-10 tDC to induce Treg in absence
or presence of anti-TNFα and exogenous TNFα. Primed T cells were subsequently
tested for their suppressive capacity on mDC-stimulated CD4+ memory T cells.
The IL-10 tDC-primed T cells potently suppressed DC-induced proliferation
of responder T cells (Figure 6), consistent with our previous findings.23,24 T cell
priming in presence of anti-TNFα resulted in more suppression of responder T cell
proliferation. Priming T cells in presence of exogenous TNFα did not affect the
suppressive capacity of IL-10 tDC-primed T cells. Thus, neutralization of the pro-
inflammatory cytokine TNFα drives naïve CD4+ T cell priming away from inflammation
and towards a regulatory phenotype, which results in a T cell population with an
enhanced suppressive capacity.

88
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

80
Tresp + mDC

% CFSElow Tresp
Tpr (IL-10 DC + anti-TNFα)
60
Tpr (IL-10 DC)
Tpr (IL-10 DC + TNFα)
40
*
20

0
2:1 1:1
Tpr:Tresp ratio

Figure 6. Neutralizing TNFα during CD4+ T cell priming with tDC increases the suppressive
capacity. Naïve CD4+ T cells were primed by IL-10 tDC (Tpr) in the presence or absence of
anti-TNFα F(ab)2 or exogenous TNFα for 6 days and subsequently tested for their suppressive
capacity on responder T cell proliferation. IL-10 tDC-primed T cells were labelled with PKH26
and added in 2 ratios to CFSE-labelled responder memory CD4+ T cells (Tresp) together with
mDC as stimulation. Proliferation of responder T cells (mean % of CFSElow Tresp + SEM) of 4
independent experiments is depicted. *p≤0.05

Discussion
The pro-inflammatory cytokine TNFα is a major player in several autoimmune
diseases, contributing to immune responses and tissue inflammation. TNFα
blocking agents have proven very successful in the treatment of patients with
autoimmune diseases.2,3 Anti-TNFα therapy affects various cell types and seems to
induce a more regulated T cell response, however, the effect of TNFα neutralization

CHAPTER 4
on naïve T cell activation and polarization is largely unknown. We here show that
neutralizing TNFα during CD4+ T cell priming polarized naïve T cells towards IL-10
production and reduced IFNγ polarization. This was mediated via absence of TNFRII
signalling. Anti-TNFα inhibited T cell activation status of the proliferated population.
Furthermore, T cell survival was inhibited in the later stage of T cell priming when
TNFα was blocked, which correlated with the induction of TNFRII expression on
primed T cells. In addition, we demonstrate that neutralization of TNFα induced a
regulatory gene profile resulting in improved suppressive function of T cells. Taken
together, priming naïve CD4+ T cells in absence of TNFα affects the polarization,
activation and survival of T cells, resulting in a more regulated immune response.
Previously it was observed in vitro that TNFα enhances IFNγ production by
T cells 17 and anti-TNFα reduces IFNγ production.17,21 In contrast, anti-TNFα therapy
in RA patients resulted in enhanced IFNγ levels by peripheral blood T cells.14-16
We here observed a reduction in IFNγ producing cells upon TNFα blockade.
Little is known about the effects of neutralizing TNFα on IL-10 induction in T cells.
Ehrenstein and co-workers 13 have previously demonstrated that treatment of RA
patients with infliximab increases the IL-10 levels in PBMC, but it remained unclear
which cells produced IL-10. We show that neutralizing TNFα during priming of naïve
CD4+ T cells increased the number of IL-10+ T cells.

89
 CHAPTER 4

The increase in IL-10+ T cells by anti-TNFα agents might be explained either

by blockage of TNFR signalling or by stimulation of mTNFα. TNFα-producing
cells express mTNFα before cleavage into sTNFα.25,34 Stimulation of mTNFα can
induce outside-to-inside signalling and lead to cytokine production, E-Selectin
expression, calcium release and apoptosis.26,35,36 Only the monoclonal antibodies
adalimumab and infliximab, and not etanercept, can induce this reverse signalling via
mTNFα 26,28,37, which is most likely mediated via cross-linking of mTNFα. Etanercept
can only bind one TNFα trimer and is hereby not able to cross-link mTNFα.26,38 We
found no difference in IL-10 polarization between all three anti-TNFα agents. In
addition, adalimumab Fab fragments showed equal increase in IL-10+ T cells as
adalimumab F(ab)2 fragments (data not shown). Taken together, this demonstrates
that it is inhibition of TNFα–TNFR interaction rather than reverse signalling via cross-
linking mTNFα that induces IL-10 in T cells.
With a stimulating anti-TNFRII antibody we showed that inhibition of IL-10
in T cells is TNFRII-mediated. This concurs with previous reports on murine
T cells 19,39,40 and human total CD4+ T cells 17 showing that the co-stimulatory effect
of TNFα on T cells is mediated via TNFRII. In addition, our data is in line with the
report from Aspalter and co-workers 17 in which TNFRII stimulation reduced IL-10
production. Furthermore, we found that the effect of anti-TNFα on IL-10 induction
is decreased upon full expression of TNFRII on the T cells, indicating that TNFα
must be neutralized before TNFRII signalling efficiently takes place in T cells. Thus,
absence of the pro-inflammatory cytokine TNFα and thereby absence of TNFRII
signalling induces IL-10 in T cells.
We observed normal initiation of T cell activation, but decreased T cell survival
CHAPTER 4

when TNFα was inhibited. Reduced survival in late stage priming coincided with the
kinetics of upregulation of TNFRII during the later stages of T cell priming, indicating
that TNFα is supporting survival of primed T cells during their stabilization phase.
T cell responses start with a proliferation phase of growth and differentiation of
naïve T cells into effector T cells. A contraction (or transition) phase occurs in a later
phase, in which a large population of effector T cells transits to a smaller population
of memory T cells.41 TNFα showed to be necessary for CD8+ T cell survival and
transition to memory stage.42 Our data indicate that TNFα has a similar role during
CD4+ T cell priming. Furthermore, our preliminary data show that anti-TNFα treatment
downmodulated IL-2 transcription in primed CD4+ T cells (data not shown). IL-2
signalling in CD4+ T cells enhances growth and differentiation into effector T cells
during the later phases of T cell priming, and is necessary for long-term survival of
memory T cells after antigen withdrawal.43,44 Inhibited IL-2 production by anti-TNFα-
treated T cells might form the mechanism underlying reduced T cell survival upon
TNFα blockage.
Neutralization of TNFα reduced the expression of CD25, 4-1BB, GITR, HLA-DR
and OX40, in line with previous indications.12,17,18,20 We extended these observations
and showed that the expression of these molecules is reduced within the proliferated
T cell population, indicating that the T cells that have entered cell cycle, and are
therefore activated, have a lower activation status. Early activation marker CD69 is

90
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

not affected, reflecting normal initial activation of the T cells. Together, this shows
that absence of TNFα reduces T cell activation by both affecting survival as well as
the activation status of the remainder of activated cells.
Whole genome gene expression analysis showed that anti-TNFα induces a
regulatory gene profile during naïve CD4+ T cell priming with increased expression of
galectin-3, LAG-3, IL-10, legumain, GARP, CTLA-4 and PD-1, which are all associated
with Treg suppressive function.30,32,45,46 In contrast to a previous study with infliximab 21,
we observed no increase in CD25 or Foxp3 gene transcript expression, indicating that
we have not expanded CD25+Foxp3+ nTreg or induced CD25+Foxp3+ iTreg.
In line with the regulatory gene profile, neutralizing TNFα during T cell priming
with tDC enhances the suppressive capacity in an in vitro suppression assay. This
finding is supported by studies in RA patients where anti-TNFα treatment induces
and/ or recruits functional Treg.11,13 In contrast, an in vitro study showed that priming
T cells with tDC in presence of adalimumab reduced Treg induction and suppressive
function.22 However, in this study vitamin D3 tDC were used, which express high
levels of mTNFα. In addition, interaction between mTNFα and TNFRII was important
for Treg induction in this study, as etanercept did not inhibit Treg induction by these
tDC, while we observed induction of IL-10+ T cells with both adalimumab and
etanercept. Although it was previously shown that TNFα inhibits Treg suppressive
function 12,20, we observed no effect of exogenous TNFα during Treg induction,
which can be explained by the experimental set-up. We investigated the effect of
TNFα and anti-TNFα on the induction of Treg, which we subsequently tested for
suppressive function in a suppression assay, whereas previous studies assessed the
effect of TNFα during a functional suppression assay.

CHAPTER 4
Neutralizing TNFα during priming of naïve CD4+ T cells induces a regulatory gene
profile and suppressive function. In RA patients preferentially activated Th1 and Th17
cells are present 47,48, and Treg fail to suppress inflammation.11-13 As we illustrate here,
the continuous presence of TNFα inhibits Treg induction from naïve T cells. Without
sufficient immune regulation to downregulate inflammation, the effector Th cells
can significantly contribute to the chronic inflammation. Anti-TNFα can inhibit this
autoreactive loop of chronic inflammation by inducing Treg. In this study we showed
that absence of TNFα enhances the IL-10 production, induces a regulatory gene
profile and induces functional Treg. Altogether, the induction of Treg together with
less T cell activation and less IFNγ production upon TNFα blockade might all play a
role in the therapeutic efficacy of TNFα inhibitors in autoimmune disease.

Acknowledgements
This work was supported by a Sanquin PPOC grant (PPOC06-026). We thank
Sanquin Blood Bank North West Region for leukaphaeresis and elutriation and Gijs
van Schijndel and Suzanne Lissenberg-Thunnissen for monocyte isolations. We
thank Jacques Neefjes, Petra Paul and Arno Velds from The Netherlands Cancer
Institute for help with analysis of the micro array data. We thank Jaap Jan Zwaginga
for critical review of the manuscript.

91
 CHAPTER 4

References
1. Aggarwal, B.B. Signalling pathways of 10. Baldwin, H.M., Ito-Ihara, T., Isaacs, J.D.,
the TNF superfamily: a double-edged Hilkens, C.M.U. TNF alpha blockade
sword. Nat. Rev. Immunol. 2003; 3:745- impairs dendritic cell survival and
756. function in rheumatoid arthritis. Ann.
2. Wong, M., Ziring, D., Korin, Y., Desai, S., Rheum. Dis. 2010; 69:1200-1207.
Kim, S., Lin, J., Gjertson, D., Braun, J., 11. Nadkarni, S., Mauri, C., Ehrenstein,
Reed, E., Singh, R.R. TNF alpha blockade M.R. Anti-TNF-alpha therapy induces
in human diseases: Mechanisms and a distinct regulatory T cell population
future directions. Clin. Immunol. 2008; in patients with rheumatoid arthritis via
126:121-136. TGF-beta. J Exp. Med. 2007; 204:33-39.
3. van Vollenhoven, R.F. Treatment of 12. Valencia, X., Stephens, G., Goldbach-
rheumatoid arthritis: state of the art Mansky, R., Wilson, M., Shevach, E.M.,
2009. Nat. Rev. Immunol. 2009; 5:531- Lipsky, P.E. TNF downmodulates the
541. function of human CD4+CD25hi T-
4. Feldmann, M., Steinman, L. Design of regulatory cells. Blood 2006; 108:253-
effective immunotherapy for human 261.
autoimmunity. Nature 2005; 435:612- 13. Ehrenstein, M.R., Evans, J.G., Singh, A.,
619. Moore, S., Warnes, G., Isenberg, D.A.,
5. Taylor, P.C., Peters, A.M., Paleolog, E., Mauri, C. Compromised function of
Chapman, P.T., Elliott, M.J., McCloskey, regulatory T cells in rheumatoid arthritis
R., Feldmann, M., Maini, R.N. Reduction and reversal by anti-TNFalpha therapy.
of chemokine levels and leukocyte traffic J. Exp. Med. 2004; 200:277-285.
to joints by tumor necrosis factor alpha 14. Aerts, N.E., De Knop, K.J., Leysen, J.,
blockade in patients with rheumatoid Ebo, D.G., Bridts, C.H., Weyler, J.J.,
arthritis. Arthr. & Rheum. 2000; 43:38-47. Stevens, W.J., De Clerck, L.S. Increased
6. Charles, P., Elliott, M.J., Davis, D., Potter, IL-17 production by peripheral T helper
A., Kalden, J.R., Antoni, C., Breedveld, cells after tumour necrosis factor
F.C., Smolen, J.S., Eberl, G., deWoody, blockade in rheumatoid arthritis is
K., Feldmann, M., Maini, R.N. Regulation accompanied by inhibition of migration-
of cytokines, cytokine inhibitors, and associated chemokine receptor
CHAPTER 4

acute-phase proteins following anti- expression. Rheumatology. (Oxford)

TNF-alpha therapy in rheumatoid 2010; 49:2264-2272.
arthritis. J. Immunol. 1999; 163:1521- 15. Schuerwegh, A.J., Van Offel, J.F.,
1528. Stevens, W.J., Bridts, C.H., De Clerck,
7. Ulfgren, A.K., Andersson, U., Engström, L.S. Influence of therapy with chimeric
M., Klareskog, L., Maini, R.N., Taylor, monoclonal tumour necrosis factor-
P.C. Systemic anti-tumor necrosis factor alpha antibodies on intracellular
alpha therapy in rheumatoid arthritis cytokine profiles of T lymphocytes
down-regulates synovial tumor necrosis and monocytes in rheumatoid arthritis
factor alpha synthesis. Arthr. & Rheum. patients. Rheumatology. (Oxford) 2003;
2000; 43:2391-2396. 42:541-548.
8. Balanescu, A., Radu, E., Nat, R., Regalia, 16. Maurice, M.M., van der Graaff, W.L.,
T., Bojinca, V., Ionescu, R., Balanescu, Leow, A., Breedveld, F.C., van Lier, R.A.,
S., Savu, C., Predeteanu, D. Early and Verweij, C.L. Treatment with monoclonal
late effect of infliximab on circulating anti-tumor necrosis factor alpha
dendritic cells phenotype in rheumatoid antibody results in an accumulation of
arthritis patients. Int. J. Clin. Pharmacol. Th1 CD4+ T cells in the peripheral blood
Res. 2005; 25:9-18. of patients with rheumatoid arthritis.
9. van Lieshout, A.W., Barrera, P., Smeets, Arthr. & Rheum. 1999; 42:2166-2173.
R.L., Pesman, G.J., van Riel, P.L., van den 17. Aspalter, R.M., Eibl, M.M., Wolf, H.M.
Berg, W.B., Radstake, T.R. Inhibition of Regulation of TCR-mediated T cell
TNF alpha during maturation of dendritic activation by TNF-RII. J. Leukoc. Biol.
cells results in the development of semi- 2003; 74:572-582.
mature cells: a potential mechanism 18. Scheurich, P., Thoma, B., Ucer, U.,
for the beneficial effects of TNF alpha Pfizenmaier, K. Immunoregulatory
blockade in rheumatoid arthritis. Ann. activity of recombinant human tumor
Rheum. Dis. 2005; 64:408-414. necrosis factor (TNF)-alpha: induction

92
 ANTI-TNFα INDUCES T CELL TOLERANCE VIA INHIBITION OF TNFRII SIGNALLING

of TNF receptors on human T cells and to-inside signals through transmembrane

TNF-alpha- mediated enhancement TNF-alpha. Gastroenterology 2005;
of T cell responses. J. Immunol. 1987; 128:376-392.
138:1786-1790. 27. Kaymakcalan, Z., Sakorafas, P., Bose, S.,
19. Grell, M., Becke, F.M., Wajant, H., Scesney, S., Xiong, L., Hanzatian, D.K.,
Männel, D.N., Scheurich, P. Tumor Salfeld, J., Sasso, E.H. Comparisons of
necrosis factor (TNF) receptor type affinities, avidities, and complement
2 mediates thymocyte proliferation activation of adalimumab, infliximab,
independently of TNF receptor type 1. and etanercept in binding to soluble and
Eur. J. Immunol. 1998; 28:257-263. membrane tumor necrosis factor. Clin.
20. Nagar, M., Jacob-Hirsch, J., Vernitsky, Immunol. 2009; 131:308-316.
H., Berkun, Y., Ben-Horin, S., Amariglio, 28. Mitoma, H., Horiuchi, T., Tsukamoto, H.,
N., Bank, I., Kloog, Y., Rechavi, G., Tamimoto, Y., Kimoto, Y., Uchino, A.,
Goldstein, I. TNF activates a NF- To, K., Harashima, S., Hatta, N., Harada,
kappa B-regulated cellular program in M. Mechanisms for cytotoxic effects of
human CD45RA- regulatory T cells that anti-tumor necrosis factor agents on
modulates their suppressive function. J. transmembrane tumor necrosis factor
Immunol. 2010; 184:3570-3581. alpha-expressing cells: comparison
21. Goldstein, I., Ben-Horin, S., Koltakov, A., among infliximab, etanercept, and
Chermoshnuk, H., Polevoy, V., Berkun, adalimumab. Arthr. & Rheum. 2008;
Y., Amariglio, N., Bank, I. Alpha1beta1 58:1248-1257.
integrin+ and regulatory Foxp3+ 29. Arora, T., Padaki, R., Liu, L., Hamburger,
T cells constitute two functionally distinct A.E., Ellison, A.R., Stevens, S.R., Louie,
human CD4+ T cell subsets oppositely J.S., Kohno, T. Differences in binding
modulated by TNFalpha blockade. J. and effector functions between classes
Immunol. 2007; 178:201-210. of TNF antagonists. Cytokine 2009;
22. Kleijwegt, F.S., Laban, S., Duinkerken, 45:124-131.
G., Joosten, A.M., Zaldumbide, A., 30. Probst-Kepper, M., Geffers, R., Kroger,
Nikolic, T., Roep, B.O. Critical role for A., Viegas, N., Erck, C., Hecht, H.J.,
TNF in the induction of human antigen- Lunsdorf, H., Roubin, R., Moharregh-
specific regulatory T cells by tolerogenic Khiabani, D., Wagner, K., Ocklenburg,
dendritic cells. J. Immunol. 2010; F., Jeron, A., Garritsen, H., Arstila, T.P.,
185:1412-1418.

CHAPTER 4
Kekalainen, E., Balling, R., Hauser, H.,
23. Boks, M.A., Zwaginga, J.J., van Ham, Buer, J., Weiss, S. GARP: a key receptor
S.M., ten Brinke, A. An optimized controlling FOXP3 in human regulatory
CFSE-based T-cell suppression assay T cells. J. Cell Mol. Med. 2009; 13:3343-
to evaluate the suppressive capacity 3357.
of regulatory T-cells induced by human
31. Corthay, A. How do regulatory T cells
tolerogenic dendritic cells. Scand. J.
work? Scand. J. Immunol. 2009; 70:326-
Immunol. 2010; 72:158-168.
336.
24. Boks, M.A., Kager-Groenland, J.R.,
Haasjes, M.S., Zwaginga, J.J., van Ham, 32. Keir, M.E., Butte, M.J., Freeman, G.J.,
S.M., ten Brinke, A. IL-10-generated Sharpe, A.H. PD-1 and its ligands in
tolerogenic dendritic cells are optimal tolerance and immunity. Annu. Rev.
for functional regulatory T cell induction Immunol. 2008; 26:677-704.
- A comparative study of human clinical- 33. Shevach, E.M. Mechanisms of Foxp3+
applicable DC. Clin. Immunol. 2012; T regulatory cell-mediated suppression.
142:332-342. Immunity 2009; 30:636-645.
25. Horiuchi, T., Mitoma, H., Harashima, 34. Bazzoni, F., Beutler, B. The tumor necrosis
S., Tsukamoto, H., Shimoda, T. factor ligand and receptor families. N.
Transmembrane TNF-alpha: structure, Engl. J Med. 1996; 334:1717-1725.
function and interaction with anti-TNF 35. Watts, A.D., Hunt, N.H., Wanigasekara,
agents. Rheumatology. (Oxford) 2010; Y., Bloomfield, G., Wallach, D.,
49:1215-1228. Roufogalis, B.D., Chaudhri, G. A casein
26. Mitoma, H., Horiuchi, T., Hatta, N., kinase I motif present in the cytoplasmic
Tsukamoto, H., Harashima, S., Kikuchi, domain of members of the tumour
Y., Otsuka, J., Okamura, S., Fujita, S., necrosis factor ligand family is implicated
Harada, M. Infliximab induces potent in ‘reverse signalling’. EMBO J. 1999;
anti-inflammatory responses by outside- 18:2119-2126.

93
 CHAPTER 4

36. Harashima, S., Horiuchi, T., Hatta, N., during pathogen challenge. Immunol.
Morita, C., Higuchi, M., Sawabe, T., Rev. 2010; 236:110-124.
Tsukamoto, H., Tahira, T., Hayashi, K., 42. Shi, M., Ye, Z., Umeshappa, K.S., Moyana,
Fujita, S., Niho, Y. Outside-to-inside T., Xiang, J. Alpha tumor necrosis factor
signal through the membrane TNF-alpha contributes to CD8(+) T cell survival in
induces E-selectin (CD62E) expression the transition phase. Biochem. Biophys.
on activated human CD4+ T cells. J. Res. Commun. 2007; 360:702-707.
Immunol. 2001; 166:130-136. 43. Dooms, H., Kahn, E., Knoechel, B.,
37. Van den Brande, J.M., Braat, H., van Abbas, A.K. IL-2 induces a competitive
den Brink, G.R., Versteeg, H.H., Bauer, survival advantage in T lymphocytes. J
C.A., Hoedemaeker, I., van, M.C., Immunol 2004; 172:5973-5979.
Hommes, D.W., Peppelenbosch, M.P., 44. Malek, T.R., Yu, A., Scibelli, P.,
van Deventer, S.J. Infliximab but not Lichtenheld, M.G., Codias, E.K. Broad
etanercept induces apoptosis in lamina programming by IL-2 receptor signaling
propria T-lymphocytes from patients for extended growth to multiple
with Crohn’s disease. Gastroenterology cytokines and functional maturation of
2003; 124:1774-1785. antigen-activated T cells. J Immunol
38. Scallon, B., Cai, A., Solowski, N., 2001; 166:1675-1683.
Rosenberg, A., Song, X.Y., Shealy, D., 45. Liang, B., Workman, C., Lee, J., Chew,
Wagner, C. Binding and functional C., Dale, B.M., Colonna, L., Flores, M.,
comparisons of two types of tumor Li, N., Schweighoffer, E., Greenberg,
necrosis factor antagonists. J. Pharmacol. S., Tybulewicz, V., Vignali, D., Clynes,
Exp. Ther. 2002; 301:418-426. R. Regulatory T cells inhibit dendritic
39. Tartaglia, L.A., Weber, R.F., Figari, cells by lymphocyte activation gene-3
I.S., Reynolds, C., Palladino, M.A., engagement of MHC class II. J. Immunol.
Jr., Goeddel, D.V. The two different 2008; 180:5916-5926.
receptors for tumor necrosis factor 46. Misra, N., Bayry, J., Lacroix-Desmazes,
mediate distinct cellular responses. S., Kazatchkine, M.D., Kaveri, S.V.
Proc. Natl. Acad. Sci. U. S. A 1991; Cutting edge: human CD4+CD25+
88:9292-9296. T cells restrain the maturation and
40. Vandenabeele, P., Declercq, W., antigen-presenting function of dendritic
Vercammen, D., Van de Craen, M., cells. J. Immunol. 2004; 172:4676-4680.
CHAPTER 4

Grooten, J., Loetscher, H., Brockhaus, 47. Skapenko, A., Leipe, J., Lipsky, P.E.,
M., Lesslauer, W., Fiers, W. Functional Schulze-Koops, H. The role of the T cell
characterization of the human tumor in autoimmune inflammation. Arthr. Res.
necrosis factor receptor p75 in a Ther. 2005; 7 Suppl 2:S4-14.
transfected rat/mouse T cell hybridoma. 48. van den Berg, W.B., Miossec, P. IL-17 as a
J. Exp. Med. 1992; 176:1015-1024. future therapeutic target for rheumatoid
41. McKinstry, K.K., Strutt, T.M., Swain, S.L. arthritis. Nat. Rev. Rheumatol. 2009;
Regulation of CD4+ T-cell contraction 5:549-553.

94
 5
CHAPTER
CD4 + T cells co-expressing IL-10
and IFNγ display a regulatory
gene profile and dampen
immune responses
Martine A. Boks, Judith R. Kager-Groenland,
S. Marieke van Ham and Anja ten Brinke
 CHAPTER 5

Abstract
The anti-inflammatory cytokine interleukin-10 (IL-10) limits inflammatory responses
by inhibiting the activity of several immune cells. Regulatory T cells (Treg) are
key IL-10 producing cells. In recent years IL-10 has also been reported to be co-
expressed by pro-inflammatory effector CD4+ T helper (Th) subsets, including Th1
cells. In spite of many descriptions of CD4+ T cells co-expressing the seemingly
contradictory cytokines IL-10 and IFNγ, the extended phenotype and function of
this intriguing population remains poorly defined. Here we investigated IL-10/ IFNγ
co-expressing CD4+ T cells and compared this population to IL-10 and IFNγ single
positive T cells. With whole genome gene expression analysis we demonstrated a
strong regulatory gene profile and a suppressed Th1 gene profile. Protein analysis
confirmed a regulatory phenotype for IL-10+/ IFNγ+ T cells, with specific enhanced
expression of GARP and PD-1 by this population. IL-10+/ IFNγ+ T cells displayed
potent suppressive capacity on responder T cell proliferation. Thus, IL-10/ IFNγ
co-expressing CD4+ T cells are suppressive cells that may contribute to host tissue
protection by downmodulating pro-inflammatory immune responses.
CHAPTER 5

98
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

Introduction
The principal function of the anti-inflammatory cytokine interleukin-10 (IL-10) is to
limit and ultimately terminate inflammatory responses. In infection, IL-10 inhibits the
activity of several immune cells, e.g. T cells, B cells, natural killer (NK) cells, mast cells,
granulocytes, macrophages, monocytes and dendritic cells (DC).1,2 These cell types
are required for optimal pathogen clearance, but at the same time may contribute to
tissue damage. IL-10 is thought to play a crucial role in reducing pathology caused
by excessive and prolonged inflammation and in prevention of autoimmunity. The
impact of the suppressive function of IL-10 was highlighted by IL-10-deficient mice,
which developed severe colitis in association with commensal bacteria in the gut
and showed other exaggerated inflammatory responses to microbial challenge.3,4 In
contrast, transgenic mice with IL-10 over-expressing antigen presenting cells (APC)
were highly susceptible to infections, and were not able to mount effective CD4+ T
helper (Th) 1 or Th2 cell responses.5
IL-10 regulates immune responses by acting on several key players of the
immune system. IL-10 inhibits expression of MHC class II and co-stimulatory
molecules CD80 and CD86 on APC (DC, monocytes, macrophages), and limits
production of pro-inflammatory cytokines and chemokines by APC.1 Consequently,
IL-10 inhibits the development of pro-inflammatory Th cell responses by APC.6 IL-10
can also act directly on CD4+ Th cells by inhibiting proliferation and production of
pro-inflammatory cytokines.7,8 In addition, IL-10 enhances the differentiation of IL-
10-secreting regulatory T cells (Treg), thus providing a positive regulatory feedback
loop for its induction.9-11
IL-10 was initially described as Th2-type cytokine CSIF (cytokine synthesis
inhibitory factor) that inhibited cytokine synthesis in Th1 cells 12, but later studies
demonstrated that the production of IL-10 was associated with regulatory T cell
responses.9,10,13,14 Tr1 cells are inducible Treg induced upon antigen-specific priming
of naïve T cells in the presence of IL-10.9,15 Specialized IL-10 producing DC, such
as immature DC 16,17 and tolerogenic (t)DC 18,19, play a key role in this process.
Specific markers for Tr1 cells have not been found; their cytokine production profile
is their key feature.11 Tr1 cells produce large amounts of IL-10, TGFβ and IL-5, and
CHAPTER 5
intermediate amounts of IFNγ. Tr1 cells suppress effector T cell proliferation through
secretion of IL-10 and TGFβ.11,20
It is now known that the expression of IL-10 is not specific to Th2 cells or
regulatory T cells, but that it is a much more broadly expressed cytokine. IL-10 can
be expressed by many cells of the adaptive immune system, including Th1 and Th17
cell subsets.21 There are indications that IL-10 production by these subsets serves
to protect against excessive immunopathology. Specifically in chronic infections, in
human and in experimental animal models, CD4+ T cells that produce high amounts
of both IL-10 and IFNγ are present.22-26 These cells share many features with Th1
cells, but also play an important regulatory role for host protection, as they are the
main source of protective IL-10.24,25 In a Toxoplasma gondii mouse model, IL-10
was rapidly, but transiently, induced after in vitro activation of IFNγ+ Th1 cells.24

99
 CHAPTER 5

The transient IL-10 expression was important for host protection, whereas the
instability of IL-10 synthesis was suggested to be a mechanism to prevent sustained
suppression of effector functions. This finding is reminiscent of recent findings in
Staphyloccus aureus-specific Th17 cells that transiently secrete IL-10 and at the
same time transiently downmodulate IL-17 secretion upon restimulation.27 The
reciprocal IL-10/ IL-17 regulation was observed over repeated cycles of restimulation,
suggesting that continuous antigen stimulation during chronic infection exerts
tissue protective effects by mediating transient IL-10 production. Thus, it appears
that cells with regulatory properties could arise from fully differentiated Th cells,
possibly as a (transient) negative feedback loop.
In spite of many descriptions of CD4+ T cells co-expressing IL-10 and IFNγ,
the extended phenotype and function of this intriguing population remains poorly
defined. In this study we induced IL-10+/ IFNγ+ T cells through priming of naïve
CD4+ T cells with tolerogenic (t)DC. Whole genome gene expression analysis
and protein expression analysis demonstrated that the IL-10/ IFNγ co-expressing
T cells exhibited a strong regulatory phenotype, with enhanced expression of
GARP and PD-1. In addition, IL-10+/ IFNγ+ T cells displayed potent suppressive
capacity on responder T cell proliferation, demonstrating that this cell type serves
to downmodulate immune responses.

Materials and Methods

Antibodies and reagents
CellGro DC serum-free medium, GM-CSF, IL-4, IL-1β and TNFα were all obtained
from CellGenix (Freiburg, Germany). IL-10 and IL-2 were purchased from
PeproTech (Rocky Hill, USA). PGE2, PMA, ionomycin and brefeldin A were all
obtained from Sigma-Aldrich (Steinheim, Germany). Penicillin and streptomycin
were obtained from Gibco (Merelbeke, Beldium). For co-cultures Iscove’s Modified
Dulbecco’s Medium (IMDM, Bio Whittaker, Verviers, Belgium) with 10% fetal calf
serum (Bodinco, Alkmaar, The Netherlands) was used as described.28 Adalimumab
(Humira; Abbott) F(ab)2 fragments were used as TNFα blocking agent at 10 μg/ml.
CHAPTER 5

As isotype control, F(ab)2 fragments of an irrelevant monoclonal antibody directed

against FelD1 (cat allergen) were used (Sanquin Reagents, Amsterdam, The
Netherlands).
The following fluorochrome-labelled monoclonal antibodies were used: CTLA-4,
Galectin-3, GARP, IFNγ and IL-4 from Becton Dickinson (BD Biosciences, San Jose,
USA). IL-10 was obtained from Diaclone (via Sanquin Reagents), PD-1 from R&D
Systems (Minneapolis, USA), IL-17 from eBioscience (Vienna, Austria), and LAG-3
from LSBio (LifeSpan Biosciences; Seattle, USA). The following antibodies were used
with an appropriate secondary antibody: Galectin-10 (R&D) and Legumain (Santa
Cruz Biotechnology; Heidelberg, Germany). Isotype-matched controls from Dako
(Glostrup, Denmark), Sanquin Reagents or BD Biosciences were used. Secondary
rabbit-anti-mouse antibody from Dako was used.

100
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

Isolation and culture of monocyte-derived dendritic cells

Monocytes were isolated from fresh aphaeresis material of healthy volunteers (after
informed consent) by using the ElutraTM cell separation system (Gambro, Lakewood,
USA). Monocytes were cultured at 1.5 x 106 cells/3 ml in 6 well plates (Nunc, Roskilde,
Denmark) in serum-free CellGro medium supplemented with IL-4 (800 IU/ml),
GM-CSF (1000 IU/ml), penicillin (100 IU/ml) and streptomycin (100 μg/ml). After 6
days, the immature DC were matured with IL-10 (40 ng/ml; 1 hour pre-incubation)
together with IL-1β (10 ng/ml), TNFα (10 ng/ml) and PGE2 (1 µg/ml) for 2 more
days to generate IL-10 tolerogenic (t)DC, as described previously.29 Alternatively,
DC were matured with IL-1β, TNFα and PGE2 without IL-10 to generate mature
immunoactivatory (m)DC.

T cell-dendritic cell co-culture and intracellular cytokine staining

Naïve CD4+CD45RA+CD45RO- T cells were isolated as described previously.29
Allogeneic IL-10 tDC were co-cultured with 1 x 105 CD4+ naïve T cells (1:5) in 96-wells
flat-bottom plates (Nunc) for 13–15 days. Anti-TNFα or isotype control F(ab)2 fragments
(10 μg/ml) were added at start of co-culture. Fresh medium plus recombinant human
IL-2 (10 U/ml) was added at day 7 of co-culture, and the cells were expanded for
the next 6-8 days. Subsequently, T cells were restimulated with PMA and ionomycin
(10 ng/ml and 1 μg/ml, respectively) for 5 hours in the presence of brefeldin A
(10 μg/ml). Production of IL-4, IL-10, IL-17 and IFNγ was detected by intracellular
FACS staining by first fixating the cells with a 4% paraformaldehyde solution and
subsequently staining intracellularly with appropriate antibodies in a saponin solution
(0.5% saponin and 1% BSA in PBS). Cytokine expression was measured on an LSRII
flow cytometer and analysed with FACS Diva software (BD Biosciences).
For phenotyping, T cells were incubated with specific monoclonal antibodies
or appropriate isotype-matched controls and analyzed by first gating IL-10+/ IFNγ-,
IL-10+/ IFNγ+ and the IL-10-/ IFNγ+ T cell populations and subsequently analyzing
expression of the specific markers within these populations.

Isolation of cytokine-expressing T cells

Allogeneic IL-10 tDC were co-cultured with 1 x 106 allogeneic CD4+ naïve T cells CHAPTER 5
(1:10) in 24 well plates (Nunc) for 13 days in presence of anti-TNFα F(ab)2 fragments
(10 μg/ml). Fresh medium plus IL-2 (10 U/ml) was added at day 7 of co-culture, and
the cells were expanded for the next 6 days. Subsequently, T cells were restimulated
with PMA and ionomycin (10 ng/ml and 1 μg/ml, respectively) for 5 hours to boost
cytokine production. DC-primed T cells were isolated based on IL-10 and IFNγ
expression using a cytokine secretion assay (Miltenyi Biotec, Bergisch Gladbach,
Germany) and subsequently FACS sorted on a FACS Aria (BD Biosciences).

Micro array analysis

Naïve CD4+ T cells were co-cultured for 13 days with IL-10 tDC in the presence of
anti-TNFα F(ab)2. After 13 days, IL-10+ and IFNγ+ T cells were isolated. Samples
were generated from three independent donors. The Human Sentrix-6 beadchip

101
 CHAPTER 5

(Illumina) was used for the whole genome gene expression analysis. The data was
log2 transformed and normalized between arrays using robust spline normalization
method. Triplicate samples of IL-10+ sorted cells were compared to IFNγ+/ IL-10-
cells by LIMMA analysis. Genes were further filtered and analyzed using log2 fold
change calculations. The entire dataset described here will be deposited in the
Gene Expression Omnibus database.

Suppression assay
Suppression assays were performed as described previously.28 Briefly, CD4+ memory
responder T cells (from the same T cell donor as primed T cells) were labelled with
0.5 μM CFSE (Invitrogen, Eugene, Oregon, USA) for 15 minutes at RT. Primed T cells,
either directly assessed or sorted based on cytokine expression, were labelled with
6 μM PKH26 (Sigma-Aldrich) for 5 minutes at RT and irradiated (60 Gray) to prevent
convergence with proliferated CFSE-negative cells. Primed T cells were incubated
with 5 x 104 responder T cells (1:2) and mature DC (from the same DC donor;
1:50 ratio with responder T cells). Cells were cultured in 96 well round bottom
plates (Greiner Bio-One, Frickenhausen, Germany) for 6 days in culture medium.
Experiments were performed in triplicate. Proliferation of responder T cells was
measured on an LSRII flow cytometer (BD Biosciences) in the presence of DAPI to
exclude dead cells.

Statistical analysis
Data are expressed as mean + standard error of the mean (SEM). Statistical
significance was analyzed using one-way Analysis Of Variance (ANOVA) with
Bonferroni post test in Graphpad Prism 5.01 software (San Diego, USA).

Results
IL-10+ T cells co-express IFNγ
Previously, we demonstrated that priming of naïve CD4+ T cells with IL-10 tolerogenic
(t)DC in the presence of an anti-TNFα monoclonal antibody induced IL-10-
CHAPTER 5

expressing T cells.30 Remarkably, the majority (> 66%) of this IL-10+­ T cell population
co-expressed IFNγ (Figure 1A). A smaller proportion of the IL-10+ T cell population
co-expressed IL-4. No co-expression with IL-17 was observed. Although priming
of naïve T cells in absence of TNFα increased the total IL-10+ T cell population
(Figure 1B), none of the different subpopulations was preferentially increased
(Figure 1C). Thus, IL-10+ T cells induced from naïve CD4+ T cells by tDC are driven
towards an effector Th cell that secretes both IFNγ and IL-10.

IL-10+/ IFNγ+ T cells display a regulatory gene profile

Since the majority of the IL-10+ T cells produced IFNγ, we investigated whether
these cells have a Treg or Th1 signature. We performed whole genome gene
expression analysis using isolated subsets from different donors. To obtain IL-10+

102
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

A
IL-10

IFNγ IL-4 IL-17

B C
** ***
15 120
IL10+
relative % positive cells
within IL-10 population
% IL-10 positive cells

IL-10+ /IL-4+
10 80 IL-10+ /IFNγ+
IL-10+ /IFNγ+ /IL-4+

5 40

0 0
- + anti-TNFα
α + isotype ctrl - + anti-TNFα
α + isotype ctrl

Figure 1. IL-10 tolerogenic DC differentiate naïve CD4+ T cells toward IL-10 expressing
cells that co-express IFNγ. Naïve CD4+ T cells were co-cultured with IL-10 tDC in the presence
of anti-TNFα F(ab)2. After 2 weeks, primed T cells were restimulated with PMA and ionomycin
and stained intracellularly for IL-10, IFNγ, IL-4 and IL-17. (A) A representative dotplot of 14
independent experiments is shown. (B, C) T cells were primed by IL-10 tDC in the presence
or absence of anti-TNFα F(ab)2 or an isotype control F(ab)2. Percentage of total IL-10 positive
T cells (B) or relative percentage of IL-10 subpopulations (C) as determined by intracellular
staining of restimulated primed T cells. Mean + SEM of 14 independent experiments is shown.
The percentage of IL-10+ T cells that co-expressed IFNγ was 71.3 ± 12.9%, 70.0 ± 12.9% and
66.4 ± 12.6% for T cells primed by tDC only, with additional anti-TNFα or with an isotype
control, respectively. **p≤0.01, ***p≤0.001

and IFNγ+ T cells, we isolated these cells based on cytokine expression with a CHAPTER 5
cytokine secretion assay and FACS sorting. For this, naïve CD4+ T cells were co-
cultured with IL-10 tDC in presence of anti-TNFα. After 2 weeks of priming, the IL-10+
T cell population and the IFNγ+/ IL-10- T cell population were isolated. In Figure 2
the cytokine secretion assay is depicted in comparison with an intracellular FACS
staining. The cytokine secretion assay showed the same percentage of cytokine
expressing T cells as determined by an intracellular staining, demonstrating that this
can be used to identify and isolate the different effector T cell populations.
We compared the gene expression of IFNγ+/ IL-10- T cells to total IL-10+ T cells,
which for the greater part co-expressed IFNγ. Micro array-based analysis of the
mRNA transcripts showed that transcription of 38 genes was upregulated ≥ 4-fold
in IL-10+ T cells compared with that in IFNγ+/ IL-10- T cells (Table I). Transcription of
8 genes was downregulated ≥ 4-fold in IL-10+ T cells compared with that in IFNγ+/

103
 CHAPTER 5

Table I – List of genes upregulated 4-fold or more in IL-10+ T cells compared to

IFNγ+/ IL-10- T cells

Gene symbol Genbank ID Gene FC* log FC**

IL10 NM_000572.2 Interleukin 10 (IL10) 22,15 4,47
TUBB2B NM_178012.3 Tubulin, beta 2B (TUBB2B) 16,78 4,07
Ribonuclease, RNase A family, 2 (liver,
RNASE2 NM_002934.2 12,56 3,65
eosinophil-derived neurotoxin) (RNASE2)
INDO NM_002164.3 Indoleamine-pyrrole 2,3 dioxygenase (INDO) 11,99 3,58
CLC NM_001828.4 Charcot-Leyden crystal protein (CLC) 11,91 3,57
IDO1 NM_002164.4 Indoleamine 2,3-dioxygenase 1 (IDO1) 11,87 3,57
Myristoylated alanine-rich protein kinase C
MARCKS NM_002356.5 10,97 3,46
substrate (MARCKS)
Proteoglycan 2, bone marrow (natural killer
PRG2 NM_002728.4 cell activator, eosinophil granule major basic 9,80 3,29
protein) (PRG2)
PRG3 NM_006093.2 Proteoglycan 3 (PRG3) 8,04 3,01
Plasminogen activator, urokinase receptor
PLAUR NM_001005376.1 7,22 2,85
(PLAUR), transcript variant 2
CCL17 NM_002987.2 Chemokine (C-C motif) ligand 17 (CCL17) 6,97 2,80
CCL22 NM_002990.3 Chemokine (C-C motif) ligand 22 (CCL22) 6,93 2,79
GALR2 NM_003857.2 Galanin receptor 2 (GALR2) 6,15 2,62
LGMN NM_001008530.1 Legumain (LGMN), transcript variant 2 6,11 2,61
CST3 NM_000099.2 Cystatin C (CST3) 5,97 2,58
Ribonuclease, RNase A family, 3 (eosinophil
RNASE3 NM_002935.2 5,57 2,48
cationic protein) (RNASE3)
CPA3 NM_001870.1 Carboxypeptidase A3 (mast cell) (CPA3) 5,56 2,48
TYRO protein tyrosine kinase binding
TYROBP NM_003332.2 5,34 2,42
protein (TYROBP), transcript variant 1
Par-6 partitioning defective 6 homolog
PARD6G NM_032510.3 5,21 2,38
gamma (C. elegans) (PARD6G)
IL1B NM_000576.2 Interleukin 1, beta (IL1B) 5,10 2,35
CHAPTER 5

Matrix metallopeptidase 9 (gelatinase

MMP9 NM_004994.2 B, 92kDa gelatinase, 92kDa type IV 4,91 2,30
collagenase) (MMP9)
Signal-regulatory protein alpha (SIRPA),
SIRPA NM_001040023.1 4,89 2,29
transcript variant 2
CXCL2 NM_002089.3 Chemokine (C-X-C motif) ligand 2 (CXCL2) 4,76 2,25
CCL2 NM_002982.3 Chemokine (C-C motif) ligand 2 (CCL2) 4,75 2,25
KIF1A NM_004321.4 Kinesin family member 1A (KIF1A) 4,73 2,24
Solute carrier family 15, member 3
SLC15A3 NM_016582.1 4,73 2,24
(SLC15A3)
Kynureninase (L-kynurenine hydrolase)
KYNU NM_003937.2 4,55 2,18
(KYNU), transcript variant 1

104
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

Table I – Continued

Gene symbol Genbank ID Gene FC* log FC**

Fascin homolog 1, actin-bundling protein
FSCN1 NM_003088.2 4,51 2,17
(Strongylocentrotus purpuratus) (FSCN1)
Major histocompatibility complex, class II,
HLA-DMB NM_002118.3 4,48 2,16
DM beta (HLA-DMB)
Chemokine (C-X-C motif) ligand 1
CXCL1 NM_001511.1 (melanoma growth stimulating activity, 4,40 2,14
alpha) (CXCL1)
CCNA1 NM_003914.2 Cyclin A1 (CCNA1) 4,37 2,13
Gardner-Rasheed feline sarcoma viral
FGR NM_001042729.1 (v-fgr) oncogene homolog (FGR), transcript 4,19 2,07
variant 3
ANKRD57 NM_023016.3 Ankyrin repeat domain 57 (ANKRD57) 4,17 2,06
TSC22 domain family, member 1 (TSC22D1),
TSC22D1 NM_006022.2 4,13 2,04
transcript variant 2
IL6 NM_000600.1 Interleukin 6 (interferon, beta 2) (IL6) 4,12 2,04
CCL7 NM_006273.2 Chemokine (C-C motif) ligand 7 (CCL7) 4,05 2,02
Ectonucleotide pyrophosphatase/
ENPP2 NM_001040092.1 phosphodiesterase 2 (ENPP2), transcript 4,05 2,02
variant 2
MAD1 mitotic arrest deficient-like 1 (yeast)
MAD1L1 NM_001013836.1 4,02 2,01
(MAD1L1), transcript variant 2

* Numbers represent the mean fold change (FC) in gene expression in IL-10+ T cells versus
IFNγ+/ IL-10- T cells of 3 independent donors
** Numbers represent the mean log2 fold change in gene expression

Cytokine secretion assay Intracellular staining

CHAPTER 5
IL-10

IFNγ

Figure 2. IL-10/ IFNγ cytokine secretion assay and intracellular FACS staining of DC-primed
T cells. Naïve CD4+ T cells were co-cultured with IL-10 tDC in the presence of anti-TNFα
F(ab)2 for 2 weeks and subsequently restimulated with PMA and ionomycin. An IL-10/ IFNγ
combined cytokine secretion assay was performed, or an intracellular FACS staining for IL-10
and IFNγ. Representative dotplots of 6 independent experiments are shown.

105
 CHAPTER 5

Table II – List of genes downregulated 4-fold or more in IL-10+ T cells compared

to IFNγ+/ IL-10- T cells
Gene symbol Genbank ID Gene FC* log FC**
Eomesodermin homolog (Xenopus laevis)
EOMES NM_005442.2 -6,26 -2,65
(EOMES)
IQCG NM_032263.2 IQ motif containing G (IQCG) -6,01 -2,59
Killer cell lectin-like receptor subfamily D,
KLRD1 NM_002262.2 member 1 (KLRD1; CD94), transcript -5,12 -2,36
variant 1
CCL1 NM_002981.1 Chemokine (C-C motif) ligand 1 (CCL1) -4,73 -2,24
XCL1 NM_002995.1 Chemokine (C motif) ligand 1 (XCL1) -4,43 -2,15
Granzyme H (cathepsin G-like 2, protein
GZMH NM_033423.3 -4,39 -2,13
h-CCPX) (GZMH)
CDC45 cell division cycle 45-like
CDC45L NM_003504.3 -4,27 -2,10
(S. cerevisiae) (CDC45L)
Ubiquitin-like with PHD and ring finger
UHRF1 NM_001048201.1 -4,09 -2,03
domains 1 (UHRF1), transcript variant 1

* Numbers represent the mean fold change (FC) in gene expression in IL-10+ T cells versus
IFNγ+/ IL-10- T cells of 3 independent donors
** Numbers represent the mean log2 fold change in gene expression

IL-10- T cells (Table II). The IL-10 gene transcript was highly upregulated in the IL-10
sorted cells (Table I and Figure 3A), confirming proper sorting of IL-10+ T cells.
Since the anti-inflammatory cytokine IL-10 is associated with immune
suppression 1,2 and produced by regulatory T cells 9,10,13,14, we further examined
the gene expression of Treg signature genes in IL-10+ T cells (Table III and Figure
3A).31-36 The Treg-associated gene transcripts of IL-10, galectin-10 (CLC), IDO1
and legumain were highly upregulated (fold change ≥ 6) in IL-10+ T cells compared
to IFNγ+/ IL-10- T cells. The Treg-associated gene transcripts of GARP (LRRC32),
CHAPTER 5

LAG-3, galectin-3 (LGALS3), TGFβ, granzyme A and CTLA-4 were slightly

upregulated (fold change ≥ 1.3). Other well-known Treg signature genes like CD25
(IL-2Rα), PD-1 (PDCD1), GITR (TNFRSF18) and Foxp3 were not induced in IL-10+
T cells compared to IFNγ+/ IL-10- T cells (Table III).
In addition, we examined typical Th1-associated gene transcripts (Figure 3B
and Table IV).37,38 IFNγ gene expression was not changed between the two
populations, which is in line with the large proportion of IL-10+ T cells that co-
expressed IFNγ. Strikingly, the Th1 transcription factors eomesodermin and
T-bet (TBX21), the Th1 specific co-stimulatory molecule KLRD1, and cytokines
lymphotoxin α and TNFα were downregulated in IL-10+ T cells. Thus, despite
IFNγ co-expression, the IL-10+ T cells display a strong regulatory gene profile and
a reduced Th1 gene profile.

106
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

A
IL10 CLC IDO1
100000 10000 1000

probe intensity
probe intensity
probe intensity

10000 1000 100

1000 100 10

100 10 1
IL-10 total IFNγγ +IL-10- IL-10 total IFNγγ + IL-10- IL-10 total IFNγγ + IL-10-

LGMN LRRC32 LAG3

1000 1000 100000

probe intensity
probe intensity

probe intensity
100 100 10000

10 10 1000
IL-10 total IFNγγ +IL-10- IL-10 total IFNγγ + IL-10- IL-10 total IFNγγ +IL-10-

LGALS3 TGFB1 CTLA4

1000 100 10000
probe intensity

probe intensity
probe intensity

10 1000

100 1 100
IL-10 total IFNγγ + IL-10- IL-10 total
+
IFNγγ IL-10
-
IL-10 total IFNγγ +IL-10-

B
EOMES KLRD1 TBX21
10 100
10000
probe intensity

probe intensity
probe intensity

1000

100

10 1 10
+ -
IL-10 total IFNγγ +IL-10- IL-10 total IFNγγ IL-10 IL-10 total IFNγγ +IL-10-

TNF
LTA IFNG
10000
10000 100000
CHAPTER 5
probe intensity
probe intensity

probe intensity

1000

100 1000 10000

IL-10 total IFNγγ + IL-10- IL-10 total IFNγγ + IL-10- IL-10 total IFNγγ + IL-10-

Figure 3. IL-10+ T cells display a regulatory gene profile with whole genome gene
expression analysis. Naïve CD4+ T cells were co-cultured with IL-10 tDC in the presence
of anti-TNFα F(ab)2 for 2 weeks and subsequently restimulated with PMA and ionomycin.
The total IL-10+ T cell population and the IFNγ+/ IL-10- T cell population were isolated with
an IL-10/ IFNγ combined cytokine secretion assay. Micro arrays were performed on samples
of three independent donors. With LIMMA analysis a mean log2 fold change of the log2
probe intensities was calculated for each gene. Genes were subsequently filtered on fold
change. Micro array probe intensities of three independent donors are plotted for selected
Treg-associated gene transcripts (A) and Th1-associated gene transcripts (B).

107
 CHAPTER 5

Table III – List of Treg signature genes in IL-10+ T cells compared to IFNγ+/
IL-10- T cells
Gene symbol Genbank ID Gene FC* log FC**
IL10 NM_000572.2 Interleukin 10 (IL10) 22,15 4,47
Charcot-Leyden crystal protein (CLC;
CLC NM_001828.4 11,91 3,57
galectin-10)
IDO1 NM_002164.4 Indoleamine 2,3-dioxygenase 1 (IDO1) 11,87 3,57
LGMN NM_001008530.1 Legumain (LGMN), transcript variant 2 6,11 2,61
Leucine rich repeat containing 32
LRRC32 NM_005512.1 2,62 1,39
(LRRC32; GARP)
LAG3 NM_002286.4 Lymphocyte-activation gene 3 (LAG3) 1,92 0,94
Interleukin 12A (natural killer cell
stimulatory factor 1, cytotoxic
IL12A NM_000882.2 1,53 0,61
lymphocyte maturation factor 1, p35)
(IL12A)
Lectin, galactoside-binding, soluble,
LGALS3 NM_002306.1 1,50 0,59
3 (LGALS3; galectin-3)
TGFB1 NM_000660.3 Transforming growth factor, beta 1 (TGFB1) 1,46 0,55
Granzyme A (granzyme 1, cytotoxic
GZMA NM_006144.2 T-lymphocyte-associated serine esterase 3) 1,40 0,48
(GZMA)
Cytotoxic T-lymphocyte-associated protein
CTLA4 NM_005214.3 1,35 0,43
4 (CTLA4), transcript variant 1
TNF receptor superfamily,
FAS NM_152877.1 member 6 (TNFRSF6; FAS), 1,13 0,17
transcript variant 7
IL2RA NM_000417.1 Interleukin 2 receptor, alpha (IL2RA; CD25) 1,12 0,17
Histocompatibility antigen, class I,
HLA-G NM_002127.3 1,08 0,12
G (HLA-G)
IL7R NM_002185.2 Interleukin 7 receptor (IL7R; CD127) -1,01 -0,02
PDCD1 NM_005018.1 Programmed cell death 1 (PDCD1; PD-1) -1,03 -0,04
Lectin, galactoside-binding, soluble,
LGALS1 NM_002305.2 -1,04 -0,05
1 (LGALS1; galectin-1)
CHAPTER 5

Tumor necrosis factor receptor superfamily,

TNFRSF18 NM_148901.1 member 18 (TNFRSF18; GITR), -1,05 -0,07
transcript variant 2
FOXP3 NM_014009.2 Forkhead box P3 (FOXP3) -1,12 -0,17
EBI3 NM_005755.2 Epstein-Barr virus induced 3 (EBI3) -1,39 -0,48

* Numbers represent the mean fold change (FC) in gene expression in IL-10+ T cells versus
IFNγ+/ IL-10- T cells of 3 independent donors
** Numbers represent the mean log2 fold change in gene expression

108
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

Table IV – List of Th1 signature genes in IL-10+ T cells compared to IFNγ+/ IL-10-
T cells
Gene symbol Genbank ID Gene FC * log FC**
Eomesodermin homolog (Xenopus laevis)
EOMES NM_005442.2 -6,26 -2,65
(EOMES)
Killer cell lectin-like receptor subfamily D,
KLRD1 NM_002262.2 -5,12 -2,36
member 1 (KLRD1; CD94), transcript variant 1
TBX21 NM_013351.1 T-box 21 (TBX21; T-bet) -1,84 -0,88
Lymphotoxin alpha (TNF superfamily,
LTA NM_000595.2 -1,52 -0,60
member 1) (LTA), transcript variant 2
Tumor necrosis factor (TNF superfamily,
TNF NM_000594.2 -1,39 -0,48
member 2) (TNF)
IFNG NM_000619.2 Interferon, gamma (IFNG) 1,00 -0,00
IL12RB2 NM_001559.2 Interleukin 12 receptor, beta 2 (IL12RB2) 1,03 0,04
Signal transducer and activator of
STAT4 NM_003151.2 1,09 0,12
transcription 4 (STAT4)
Signal transducer and activator of transcription
STAT1 NM_139266.1 1,13 0,18
1, 91kDa (STAT1), transcript variant beta
Signal transducer and activator of transcription
STAT1 NM_007315.2 1,49 0,58
1, 91kDa (STAT1), transcript variant alpha

* Numbers represent the mean fold change (FC) in gene expression in IL-10+ T cells versus
IFNγ+/ IL-10- T cells of 3 independent donors
** Numbers represent the mean log2 fold change in gene expression

IL-10+/ IFNγ+ T cells show a strong regulatory phenotype

We analyzed a selection of regulatory markers found with micro array analysis
on the IL-10/ IFNγ co-expressing population and IL-10 and IFNγ single positive
T cell populations to specifically determine the phenotype of the IL-10/ IFNγ co-
expressing T cells in relation to the single positive T cells. CTLA-4 and galectin-3
were highly expressed on both IL-10+ and IL-10+/ IFNγ+ T cells (Figure 4), suggesting
an association of these regulatory molecules with IL-10. LAG-3 showed the same CHAPTER 5
trend. In contrast, galectin-10 was higher expressed on IL-10+/ IFNγ+ and IFNγ+
T cells than on IL-10+ T cells, suggesting an association with IFNγ. The IL-10+/ IFNγ+
double positive cells showed enhanced expression of GARP and PD-1, and possibly
legumain, compared to both IL-10 and IFNγ single positive T cells. Overall, the
IL-10+/ IFNγ+ T cell population displays a profound regulatory phenotype, with
GARP and PD-1 as potential markers.

IL-10+/ IFNγ+ T cells show a strong regulatory function

Considering the regulatory phenotype of IL-10+/ IFNγ+ T cells, we investigated
the suppressive capacity of this population compared to IFNγ single positive
T cells in functional suppression assays. IL-10+/ IFNγ+ T cells, IFNγ single positive
T cells and IL-10-/ IFNγ- T cells were isolated based on cytokine expression.

109
 CHAPTER 5

CTLA-4 Galectin-3 LAG-3 Galectin-10

120 *** 120 120 120 *
*** ** **
relative MFI

relative MFI

relative MFI

relative MFI
80 80 80 80

40 40 40 40

0 0 0 0
+

+
-

-
0

0
γ

γ
γ

γ
N

N
-1

-1

-1

-1
N

N
IF

IF

IF

IF
IF

IF

IF

IF
IL

IL

IL

IL
+

+
+

+
+

+
0

0
0

0
γ

γ
-1

-1

-1

-1
N

N
-1

-1

-1

-1
IL

IF

IL

IF

IL

IF

IL

IF
IL

IL

IL

IL
GARP PD-1 Legumain
120 120 120
* *** *** ** *
relative MFI

relative MFI

relative MFI
80 80 80

40 40 40

0 0 0
+

+
-

-
0

0
γ

γ
γ

γ
N

N
-1

-1

-1
N

N
IF

IF

IF
IF

IF

IF
IL

IL

IL
+

+
+

+
+

+
0

0
0

0
γ

γ
-1

-1

-1
N

N
-1

-1

-1
IL

IF

IL

IF

IL

IF
IL

IL

IL

Figure 4. IL-10+/ IFNγ+ T cells display a regulatory phenotype. Naïve CD4+ T cells were
co-cultured with IL-10 tDC in the presence of anti-TNFα F(ab)2 for 2 weeks and subsequently
restimulated with PMA and ionomycin. Primed T cells were stained intracellularly for IL-10 and
IFNγ in combination with specific monoclonal antibodies directed against Treg-associated
molecules. Mean fluorescence intensity (MFI; relative to IL-10+/ IFNγ+ cells) is depicted for
IL-10+/ IFNγ-, IL-10+/ IFNγ+ and IL-10-/ IFNγ+ T cells. Mean + SEM of 6 or 5 (GARP and LAG-3)
independent experiments is shown. *p≤0.05, **p≤0.01, ***p≤0.001

Suppression
(relative to start population)

120
% Tresp proliferation

80
CHAPTER 5

40

0
+

-
)

γ
C

N
N

N
D

/IF
/IF

IF
-
0

-
+
-1

0
0

-1

-1
IL

-1

IL

IL
r(

IL
Tp

Figure 5. IL-10+/ IFNγ+ T cells display potent suppressive capacity. Naïve CD4+ T cells were
co-cultured with IL-10 tDC in the presence of anti-TNFα F(ab)2 for 2 weeks and subsequently
restimulated with PMA and ionomycin. IL-10+/ IFNγ+, IL-10-/ IFNγ+ and IL-10-IFNγ- T cell
populations were isolated with an IL-10/ IFNγ combined cytokine secretion assay. The total
population of IL-10 tDC-primed T cells (Tpr (IL-10 DC); start population) and the different
isolated populations were tested for their suppressive capacity on responder T cell proliferation.
Primed T cells were labelled with PKH26 and added to CFSE-labelled responder memory CD4+
T cells (Tresp) together with mDC as stimulation. Experiments were performed in triplicate.
Percentage of responder T cell proliferation (% of CFSElow Tresp) relative to the start population
is shown. A representative experiment out of 3 independent experiments is depicted.

110
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

The suppressive capacity on mDC-stimulated CD4+ responder T cells of these

populations was compared to the total T cell (start) population primed by IL-10
tDC. Due to low numbers of IL-10 single positive T cells, we were not able to isolate
this population in a sufficient amount. IL-10/ IFNγ co-expressing T cells strongly
suppressed proliferation of stimulated responder T cells when compared to IFNγ
single positive T cells and the total T cell population primed by IL-10 tDC (Figure 5).
Taken together, our data demonstrate a strong regulatory phenotype and potent
suppressive capacity for IL-10+/ IFNγ+ T cells, and identified GARP and PD-1 as
potential markers for this T cell population.

Discussion
Co-expression of seemingly contradictory cytokines IL-10 and IFNγ has been
described for T-bet+ Th1 cells 22-26 and for Tr1 cells.11,15,16 To learn more about this
T cell population, we investigated the phenotypic and functional properties of IL-10/
IFNγ co-expressing CD4+ T cells primed by tDC, and compared this population
to IL-10 and IFNγ single positive T cells. We demonstrated that the IL-10/ IFNγ
co-expressing T cells exhibit a potent suppressive function, in line with the strong
regulatory gene profile and phenotype of this population.
Regulatory T cells are one of the key IL-10 producing cells.1,21 In our study we
compared gene expression of IL-10+ T cells, which for a large part co-expressed
IFNγ, with IFNγ+/ IL-10- T cells and observed an enhanced regulatory gene profile,
indicating that IL-10 expression is linked to the expression of regulatory molecules.
Galectin-10, IDO1, legumain, GARP, LAG-3, galectin-3, TGFβ, granzyme A and
CTLA-4 gene transcripts were highly upregulated in IL-10+ T cells, and these
molecules are all associated with suppressive T cell function.31-36
Based on the micro array analysis in which the IL-10+ T cells showed a regulatory
signature, we analyzed on protein level the expression of a selection of molecules
on IL-10/ IFNγ co-expressing T cells and IL-10 and IFNγ single positive T cell
populations. GARP and legumain were highest expressed on IL-10+/ IFNγ+ T cells,
suggesting that they may represent specific markers for this T cell population.
Galectin-3 was highly expressed on IL-10+/ IFNγ+ and IL-10 single positive T cells. CHAPTER 5
Previously, GARP, legumain and galectin-3 were identified as key molecules for human
CD4+CD25+ natural Treg suppressive function.32 GARP (LRRC32) is a transmembrane
glycoprotein containing leucine rich repeats.39 GARP overexpression in T cells lead
to upregulation of Foxp3, galectin-3 and legumain, and a suppressive function 32,40,
which was cell contact dependent.32 Downregulation of GARP impaired the
suppressive function of Treg cells.32 The extracellular domain of GARP, but not the
cytoplasmic region, was important for suppression 40, suggesting that the interaction
of GARP with other signalling receptors is important. Additionally, overexpression
of galectin-3 and legumain enhanced Foxp3 and GARP expression.32 The exact
mechanism of legumain, an asparaginyl endopeptidase localized in endosomal/
lysosomal compartments 41, and galectin-3, an intracellular lectin 42, in inducing
Foxp3 and GARP is unclear. Although we did not observe enhanced gene expression
of Foxp3, the other molecules of this regulatory network were highly expressed on

111
 CHAPTER 5

gene and protein level by IL-10+/ IFNγ+ T cells and/ or IL-10+ T cells. This suggests
an important role in suppression for GARP, legumain and galectin-3 by Treg subsets
other than natural Treg.
PD-1 was highest expressed on IL-10+/ IFNγ+ T cells, suggesting that PD-1 may
also represent a marker for this T cell population. PD-1 and its ligands, PD-L1 and
PD-L2, have been linked to T cell tolerance. The ligands PD-L1 and PD-L2 are
expressed by T cells and APC, whereas PD-1 is expressed on T cells only after
activation.43 Both PD-1 and its ligands deliver inhibitory signals, thereby inhibiting
T cell responses.34 Expression of PD-1 by IL-10+/ IFNγ+ T cells may represent one of
its inhibitory mechanisms hereby suppressing T cell responses. On the other hand,
it may also characterize this population as strongly activated effector T cells.
Protein expression of galectin-10 was similar for IFNγ single positive T cells, as for
IL-10+/ IFNγ+ T cells, while lower expressed on the IL-10 single positive population,
indicating an association with IFNγ. It was shown before by proteome analysis that
human CD25+Foxp3+ natural Treg specifically express galectin-10 when compared
to CD4+CD25- T cells, which was important for their suppressive function.31 The exact
mechanism of galectin-10 in suppression is not clear yet. Galectin-10, also known as
Charcot-Leyden crystal (CLC) protein, is expressed intracellularly, therefore it seems
unlikely to be involved in cell contact dependent suppression by Treg.
LAG-3 and CTLA-4 were highly expressed on both IL-10 single positive as well
as IL-10+/ IFNγ+ T cells compared to IFNγ single positive T cells, suggesting an
association of these regulatory molecules with IL-10. Our data is in agreement with
a study showing LAG-3 expression by induced Treg and activated natural Treg.44
LAG-3 is a CD4 homolog that binds MHCII with high affinity.44 It was shown that
LAG-3 engagement to MHCII induces suppression via inhibition of DC maturation
and immunostimulatory capacity.45 Overexpression of LAG-3 confers regulatory
activity in T cells 44,45 in a manner dependent on cell contact and MHCII engagements,
but independent of LAG-3 signalling in T cells.45 Interestingly, LAG-3+ T cells have
been associated before with production of large amounts of IL-10 and moderate
amounts of IFNγ.46
CTLA-4, a homolog of CD28, is an inhibitory molecule that inhibits effector T cells
by competing with CD28 for CD80/ CD86 ligation and/ or by downregulation of IL-2
CHAPTER 5

production by the effector T cells.47 In addition, CTLA-4 is constitutively expressed

by Foxp3+CD25+ natural Treg and important for suppressive function.35,36,47 CTLA-4+
Treg can suppress DC by inhibiting maturation 48,49 and by inducing IDO, an enzyme
that induces tryptophan catabolism into pro-apoptotic metabolites.50,51 CTLA-4
expression was also demonstrated on IL-10 tDC-induced Treg, which was necessary
for their suppressive function on DC-stimulated responder T cells.52 Thus, CTLA-4
expression on IL-10/ IFNγ co-expressing T cells may be important for DC-mediated
suppression of responder T cells.
So far, most of the above described regulatory molecules are described for
CD25+Foxp3+ (natural) Treg, but might represent molecules with a more general
mechanism of suppression, as is suggested by the upregulated expression of these
molecules on IL-10+ and/ or IL-10+/ IFNγ+ T cells. Interestingly, although IFNγ levels

112
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

were comparable between IL-10+ T cells, which mainly consist of IFNγ co-expressing
T cells, and IFNγ single positive T cells, other Th1 associated genes were lower
expressed in IL-10+/ IFNγ+ T cells compared to IFNγ+/ IL-10- T cells, indicating no
association of IL-10/ IFNγ co-expressing T cells with Th1 functions.
It will be of interest to determine whether the IL-10+/ IFNγ+ T cell population
differentiates directly from naïve T cells or are derived from either Th1 or Treg cells. It
is possible that these double positive cells have originally differentiated along a Th1
cell pathway. Various reports have shown that through repeated high-level antigenic
stimulation 53,54, IL-27 exposure 55-57 or IL-12 exposure 58, IL-10 production is induced in
Th1 cells. A study by Cardone and co-workers showed that IL-10/ IFNγ co-expressing
cells as well as IL-10 single positive cells can originate from sorted IFNγ+ T cells.59 In
favour of the other possibility, it was shown that Treg cells can lose Foxp3 expression
and/ or acquire the capacity to produce pro-inflammatory cytokines under strong
Th1 polarizing conditions, such as Toxoplasma gondii infection in mice 60, in patients
with relapsing remitting multiple sclerosis 61 or by IL-12 exposure.61
Corresponding with the regulatory phenotype, the IL-10+/ IFNγ+ T cells displayed
a suppressive function as they potently suppressed the proliferation of responder
T cells. This is in agreement with the fact that despite differential characterization
as either Th1 cells or Treg cells, IL-10/ IFNγ co-expressing cells are associated
with immune suppression, as was shown in in vivo infection/ inflammation animal
models 24,25,62 as well as in in vitro suppression assays with human IL-10/ IFNγ co-
expressing T cells.11,20,59,63,64 Suppression by these cells is reported to be mediated
via IL-10 production 11,63,64 or via cell contact dependent mechanisms.62 It needs
to be further determined how suppression by tDC-induced IL-10+/ IFNγ+ T cells is
mediated, and whether IL-10 and/ or the regulatory molecules galectin-10, legumain,
GARP, LAG-3, galectin-3, CTLA-4 and PD-1 play a role in this. The expression of
transmembrane molecules GARP, LAG-3, CTLA-4 and PD-1 suggests a role in cell
contact-dependent suppression.
In this study we demonstrated an extended regulatory phenotype and function
for IL-10+/ IFNγ+ T cells induced by tolerogenic DC and a potent suppressive
capacity. GARP and PD-1 were found to be potential markers for these IL-10/ IFNγ
co-expressing T cells. Regardless of the origin of IL-10/ IFNγ co-expressing CD4+ CHAPTER 5
T cells, these cells are functionally suppressive and contribute to host survival
by the production of pro-inflammatory cytokines and at the same time limiting
immunopathology by IL-10 production.

Acknowledgements
This work was supported by a Sanquin PPOC grant (PPOC06-026). We thank
Sanquin Blood Bank North West Region for leukaphaeresis and elutriation and Gijs
van Schijndel and Suzanne Lissenberg-Thunnissen for monocyte isolations. We
thank Erik Mul and Floris van Alpghen for excellent FACS sorting. We thank Jacques
Neefjes, Petra Paul and Arno Velds from The Netherlands Cancer Institute for help
with analysis of the micro array data.

113
 CHAPTER 5

References
1. Moore, K.W., de Waal, M.R., Coffman, by T helper type 1 (Th1)- and Th2-
R.L., O’Garra, A. Interleukin-10 and inducing cytokines. J. Exp. Med. 2002;
the interleukin-10 receptor. Annu. Rev. 195:603-616.
Immunol. 2001; 19:683-765. 11. Roncarolo, M.G., Gregori, S., Battaglia,
2. Couper, K.N., Blount, D.G., Riley, E.M. M., Bacchetta, R., Fleischhauer, K.,
IL-10: The Master Regulator of Immunity Levings, M.K. Interleukin-10-secreting
to Infection. J Immunol 2008; 180:5771- type 1 regulatory T cells in rodents and
5777. humans. Immunol. Rev. 2006; 212:28-50.
3. Kuhn, R., Lohler, J., Rennick, D., Rajewsky, 12. Fiorentino, D.F., Bond, M.W., Mosmann,
K., Muller, W. Interleukin-10-deficient T.R. Two types of mouse T helper cell. IV.
mice develop chronic enterocolitis. Cell Th2 clones secrete a factor that inhibits
1993; 75:263-274. cytokine production by Th1 clones. J.
4. Sellon, R.K., Tonkonogy, S., Schultz, M., Exp. Med. 1989; 170:2081-2095.
Dieleman, L.A., Grenther, W., Balish, 13. Vieira, P.L., Christensen, J.R., Minaee, S.,
E., Rennick, D.M., Sartor, R.B. Resident O’Neill, E.J., Barrat, F.J., Boonstra, A.,
enteric bacteria are necessary for Barthlott, T., Stockinger, B., Wraith, D.C.,
development of spontaneous colitis and O’Garra, A. IL-10-secreting regulatory
immune system activation in interleukin- T cells do not express Foxp3 but
10-deficient mice. Infect. Immun. 1998; have comparable regulatory function
66:5224-5231. to naturally occurring CD4+CD25+
5. Groux, H., Cottrez, F., Rouleau, M., regulatory T cells. J. Immunol. 2004;
Mauze, S., Antonenko, S., Hurst, S., 172:5986-5993.
McNeil, T., Bigler, M., Roncarolo, M.G., 14. Jonuleit, H., Schmitt, E., Schuler, G.,
Coffman, R.L. A transgenic model to Knop, J., Enk, A.H. Induction of Interleukin
analyze the immunoregulatory role of IL- 10-producing, Nonproliferating CD4+
10 secreted by antigen-presenting cells. T Cells with Regulatory Properties by
J. Immunol. 1999; 162:1723-1729. Repetitive Stimulation with Allogeneic
6. Fiorentino, D.F., Zlotnik, A., Vieira, P., Immature Human Dendritic Cells. The
Mosmann, T.R., Howard, M., Moore, Journal of Experimental Medicine 2000;
K.W., O’Garra, A. IL-10 acts on the 192:1213-1222.
antigen-presenting cell to inhibit 15. Levings, M.K., Sangregorio, R., Galbiati,
cytokine production by Th1 cells. J. F., Squadrone, S., Waal Malefyt, R.,
Immunol. 1991; 146:3444-3451. Roncarolo, M.G. IFN-{{alpha}} and IL-
7. Joss, A., Akdis, M., Faith, A., Blaser, K., 10 Induce the Differentiation of Human
Akdis, C.A. IL-10 directly acts on T cells Type 1 T Regulatory Cells. J Immunol
by specifically altering the CD28 co- 2001; 166:5530-5539.
stimulation pathway. Eur. J. Immunol. 16. Levings, M.K., Gregori, S., Tresoldi, E.,
2000; 30:1683-1690. Cazzaniga, S., Bonini, C., Roncarolo,
8. Schandene, L., Alonso-Vega, C., M.G. Differentiation of Tr1 cells by
Willems, F., Gerard, C., Delvaux, A., immature dendritic cells requires IL-
CHAPTER 5

Velu, T., Devos, R., de, B.M., Goldman, 10 but not CD25+CD4+ Tr cells. Blood
M. B7/CD28-dependent IL-5 production 2005; 105:1162-1169.
by human resting T cells is inhibited by 17. Roncarolo, M.G., Levings, M.K.,
IL-10. J. Immunol. 1994; 152:4368-4374. Traversari, C. Differentiation of T
9. Groux, H., O’Garra, A., Bigler, M., Regulatory Cells by Immature Dendritic
Rouleau, M., Antonenko, S., de Vries, Cells. The Journal of Experimental
J.E., Roncarolo, M.G. A CD4+ T-cell Medicine 2001; 193:5F-10.
subset inhibits antigen-specific T-cell 18. Gregori, S., Tomasoni, D., Pacciani, V.,
responses and prevents colitis. Nature Scirpoli, M., Battaglia, M., Magnani,
1997; 389:737-742. C.F., Hauben, E., Roncarolo, M.G.
10. Barrat, F.J., Cua, D.J., Boonstra, A., Differentiation of type 1 T regulatory (Tr1)
Richards, D.F., Crain, C., Savelkoul, cells by tolerogenic DC-10 requires the
H.F., de Waal-Malefyt, R., Coffman, R.L., IL-10-dependent ILT4/HLA-G pathway.
Hawrylowicz, C.M., O’Garra, A. In vitro Blood 2010; 116:935-944.
generation of interleukin 10-producing 19. Wakkach, A., Fournier, N., Brun, V.,
regulatory CD4(+) T cells is induced by Breittmayer, J.P., Cottrez, F., Groux, H.
immunosuppressive drugs and inhibited Characterization of Dendritic Cells that

114
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

Induce Tolerance and T Regulatory 1 tolerogenic dendritic cells are optimal

Cell Differentiation In Vivo. Immunity for functional regulatory T cell induction
2003; 18:605-617. - A comparative study of human clinical-
20. Fujio, K., Okamura, T., Yamamoto, K. The applicable DC. Clin. Immunol. 2012;
Family of IL-10-secreting CD4+ T cells. 142:332-342.
Adv. Immunol. 2010; 105:99-130. 30. Boks, M.A., Kager-Groenland, J.R., van
21. Saraiva, M., O’Garra, A. The regulation Ham, S.M., ten Brinke, A. Inhibition
of IL-10 production by immune cells. Nat of TNFRII signalling by anti-TNFalpha
Rev Immunol 2010; 10:170-181. polarizes naive CD4+ T cell priming
22. Gerosa, F., Nisii, C., Righetti, S., towards IL-10+ cells with strong
Micciolo, R., Marchesini, M., Cazzadori, regulatory phenotype and function.
A., Trinchieri, G. CD4(+) T cell clones Submitted 2012.
producing both interferon-gamma 31. Kubach, J., Lutter, P., Bopp, T., Stoll,
and interleukin-10 predominate in S., Becker, C., Huter, E., Richter, C.,
bronchoalveolar lavages of active Weingarten, P., Warger, T., Knop,
pulmonary tuberculosis patients. Clin. J., Mullner, S., Wijdenes, J., Schild,
Immunol. 1999; 92:224-234. H., Schmitt, E., Jonuleit, H. Human
23. Pohl-Koppe, A., Balashov, K.E., Steere, CD4+CD25+ regulatory T cells:
A.C., Logigian, E.L., Hafler, D.A. proteome analysis identifies galectin-10
Identification of a T cell subset capable as a novel marker essential for their
of both IFN-gamma and IL-10 secretion anergy and suppressive function. Blood
in patients with chronic Borrelia 2007; 110:1550-1558.
burgdorferi infection. J. Immunol. 1998; 32. Probst-Kepper, M., Geffers, R., Kroger,
160:1804-1810. A., Viegas, N., Erck, C., Hecht, H.J.,
24. Jankovic, D., Kullberg, M.C., Feng, C.G., Lunsdorf, H., Roubin, R., Moharregh-
Goldszmid, R.S., Collazo, C.M., Wilson, Khiabani, D., Wagner, K., Ocklenburg,
M., Wynn, T.A., Kamanaka, M., Flavell, F., Jeron, A., Garritsen, H., Arstila, T.P.,
R.A., Sher, A. Conventional T-bet(+) Kekalainen, E., Balling, R., Hauser, H.,
Foxp3(-) Th1 cells are the major source of Buer, J., Weiss, S. GARP: a key receptor
host-protective regulatory IL-10 during controlling FOXP3 in human regulatory
intracellular protozoan infection. J. Exp. T cells. J. Cell Mol. Med. 2009; 13:3343-
Med. 2007; 204:273-283. 3357.
25. Anderson, C.F., Oukka, M., Kuchroo, 33. Corthay, A. How do regulatory T cells
V.J., Sacks, D. CD4(+)CD25(-)Foxp3(-) work? Scand. J. Immunol. 2009; 70:326-
Th1 cells are the source of IL-10- 336.
mediated immune suppression in 34. Keir, M.E., Butte, M.J., Freeman, G.J.,
chronic cutaneous leishmaniasis. J. Exp. Sharpe, A.H. PD-1 and its ligands in
Med. 2007; 204:285-297. tolerance and immunity. Annu. Rev.
26. Trinchieri, G. Regulatory role of T cells Immunol. 2008; 26:677-704.
producing both interferon gamma and 35. Vignali, D.A.A., Collison, L.W., Workman,
interleukin 10 in persistent infection. J. C.J. How regulatory T cells work. Nat.
Exp. Med. 2001; 194:F53-F57. Rev. Immunol. 2008; 8:523-532.
36. Shevach, E.M. Mechanisms of Foxp3+ CHAPTER 5
27. Zielinski, C.E., Mele, F., Aschenbrenner,
D., Jarrossay, D., Ronchi, F., Gattorno, T regulatory cell-mediated suppression.
M., Monticelli, S., Lanzavecchia, A., Immunity 2009; 30:636-645.
Sallusto, F. Pathogen-induced human 37. Zhu, J., Yamane, H., Paul, W.E.
TH17 cells produce IFN-gamma or IL-10 Differentiation of effector CD4 T cell
and are regulated by IL-1beta. Nature populations. Annu. Rev. Immunol. 2010;
2012; 484:514-518. 28:445-489.
28. Boks, M.A., Zwaginga, J.J., van Ham, 38. Meyers, J.H., Ryu, A., Monney, L.,
S.M., ten Brinke, A. An optimized Nguyen, K., Greenfield, E.A., Freeman,
CFSE-based T-cell suppression assay G.J., Kuchroo, V.K. Cutting edge: CD94/
to evaluate the suppressive capacity NKG2 is expressed on Th1 but not Th2
of regulatory T-cells induced by human cells and costimulates Th1 effector
tolerogenic dendritic cells. Scand. J. functions. J. Immunol. 2002; 169:5382-
Immunol. 2010; 72:158-168. 5386.
29. Boks, M.A., Kager-Groenland, J.R., 39. Ollendorff, V., Noguchi, T., deLapeyriere,
Haasjes, M.S., Zwaginga, J.J., van Ham, O., Birnbaum, D. The GARP gene
S.M., ten Brinke, A. IL-10-generated encodes a new member of the family of

115
 CHAPTER 5

leucine-rich repeat-containing proteins. 50. Fallarino, F., Grohmann, U., Hwang,

Cell Growth Differ 1994; 5:213-219. K.W., Orabona, C., Vacca, C., Bianchi,
40. Wang, R., Wan, Q., Kozhaya, L., Fujii, R., Belladonna, M.L., Fioretti, M.C.,
H., Unutmaz, D. Identification of a Alegre, M.L., Puccetti, P. Modulation
regulatory T cell specific cell surface of tryptophan catabolism by regulatory
molecule that mediates suppressive T cells. Nat. Immunol. 2003; 4:1206-
signals and induces Foxp3 expression. 1212.
PLoS. One. 2008; 3:e2705. 51. Grohmann, U., Orabona, C., Fallarino,
41. Watts, C., Matthews, S.P., Mazzeo, D., F., Vacca, C., Calcinaro, F., Falorni, A.,
Manoury, B.+., Moss, C.X. Asparaginyl Candeloro, P., Belladonna, M.L., Bianchi,
endopeptidase: case history of a R., Fioretti, M.C., Puccetti, P. CTLA-4-Ig
class II MHC compartment protease. regulates tryptophan catabolism in vivo.
Immunological Reviews 2005; 207:218- Nat. Immunol. 2002; 3:1097-1101.
228. 52. Kubsch, S., Graulich, E., Knop, J.,
42. Hsu, D.K., Chen, H.Y., Liu, F.T. Galectin-3 Steinbrink, K. Suppressor activity of
regulates T-cell functions. Immunol. Rev. anergic T cells induced by IL-10-treated
2009; 230:114-127. human dendritic cells: association with
IL-2- and CTLA-4-dependent G1 arrest
43. Greenwald, R.J., Freeman, G.J., Sharpe, of the cell cycle regulated by p27Kip1.
A.H. The B7 Family Revisited. Annu. Rev. Eur. J Immunol. 2003; 33:1988-1997.
Immunol. 2004; 23:515-548.
53. Gabrysova, L., Nicolson, K.S., Streeter,
44. Huang, C.T., Workman, C.J., Flies, D., H.B., Verhagen, J., Sabatos-Peyton, C.A.,
Pan, X., Marson, A.L., Zhou, G., Hipkiss, Morgan, D.J., Wraith, D.C. Negative
E.L., Ravi, S., Kowalski, J., Levitsky, H.I., feedback control of the autoimmune
Powell, J.D., Pardoll, D.M., Drake, C.G., response through antigen-induced
Vignali, D.A. Role of LAG-3 in regulatory differentiation of IL-10-secreting Th1
T cells. Immunity. 2004; 21:503-513. cells. J. Exp. Med. 2009; 206:1755-1767.
45. Liang, B., Workman, C., Lee, J., Chew, 54. Saraiva, M., Christensen, J.R., Veldhoen,
C., Dale, B.M., Colonna, L., Flores, M., M., Murphy, T.L., Murphy, K.M., O’Garra,
Li, N., Schweighoffer, E., Greenberg, A. Interleukin-10 production by Th1
S., Tybulewicz, V., Vignali, D., Clynes, cells requires interleukin-12-induced
R. Regulatory T cells inhibit dendritic STAT4 transcription factor and ERK MAP
cells by lymphocyte activation gene-3 kinase activation by high antigen dose.
engagement of MHC class II. J. Immunol. Immunity. 2009; 31:209-219.
2008; 180:5916-5926.
55. Awasthi, A., Carrier, Y., Peron, J.P.,
46. Okamura, T., Fujio, K., Shibuya, M., Bettelli, E., Kamanaka, M., Flavell, R.A.,
Sumitomo, S., Shoda, H., Sakaguchi, Kuchroo, V.K., Oukka, M., Weiner, H.L.
S., Yamamoto, K. CD4+CD25-LAG3+ A dominant function for interleukin 27
regulatory T cells controlled by the in generating interleukin 10-producing
transcription factor Egr-2. Proc. Natl. anti-inflammatory T cells. Nat. Immunol.
Acad. Sci. U. S. A 2009; 106:13974- 2007; 8:1380-1389.
13979. 56. Fitzgerald, D.C., Zhang, G.X., El-
CHAPTER 5

47. Wing, K., Yamaguchi, T., Sakaguchi, S. Behi, M., Fonseca-Kelly, Z., Li, H.,
Cell-autonomous and -non-autonomous Yu, S., Saris, C.J., Gran, B., Ciric, B.,
roles of CTLA-4 in immune regulation. Rostami, A. Suppression of autoimmune
Trends in Immunology 2011; 32:428- inflammation of the central nervous
433. system by interleukin 10 secreted by
48. Serra, P., Amrani, A., Yamanouchi, J., Han, interleukin 27-stimulated T cells. Nat.
B., Thiessen, S., Utsugi, T., Verdaguer, J., Immunol. 2007; 8:1372-1379.
Santamaria, P. CD40 ligation releases 57. Stumhofer, J.S., Silver, J.S., Laurence,
immature dendritic cells from the control A., Porrett, P.M., Harris, T.H., Turka,
of regulatory CD4+CD25+ T cells. L.A., Ernst, M., Saris, C.J., O’Shea, J.J.,
Immunity. 2003; 19:877-889. Hunter, C.A. Interleukins 27 and 6 induce
49. Misra, N., Bayry, J., Lacroix-Desmazes, STAT3-mediated T cell production of
S., Kazatchkine, M.D., Kaveri, S.V. interleukin 10. Nat. Immunol. 2007;
Cutting edge: human CD4+CD25+ 8:1363-1371.
T cells restrain the maturation and 58. Meyaard, L., Hovenkamp, E., Otto,
antigen-presenting function of dendritic S.A., Miedema, F. IL-12-induced IL-
cells. J. Immunol. 2004; 172:4676-4680. 10 production by human T cells as a

116
 IL-10+/ IFNγ+ CD4+ T CELLS DISPLAY REGULATORY PHENOTYPE AND FUNCTION

negative feedback for IL-12-induced helper type 1-like, Foxp3+ regulatory

immune responses. J. Immunol. 1996; T cells in human autoimmune disease.
156:2776-2782. Nat. Med. 2011; 17:673-675.
59. Cardone, J., Le, F.G., Vantourout, 62. Hsu, C.Y., Leu, S.J., Chiang, B.L., Liu,
P., Roberts, A., Fuchs, A., Jackson, H.E., Su, H.C., Lee, Y.L. Cytokine gene-
I., Suddason, T., Lord, G., Atkinson, modulated dendritic cells protect
J.P., Cope, A., Hayday, A., Kemper, C. against allergic airway inflammation by
Complement regulator CD46 temporally inducing IL-10+IFN-[gamma]+CD4+
regulates cytokine production by T cells. Gene Ther 2010; 17:1011-1021.
conventional and unconventional T cells. 63. Haringer, B., Lozza, L., Steckel,
Nat. Immunol. 2010; 11:862-871. B., Geginat, J. Identification and
60. Oldenhove, G., Bouladoux, N., characterization of IL-10/IFN-{gamma}-
Wohlfert, E.A., Hall, J.A., Chou, D., Dos, producing effector-like T cells with
S.L., O’Brien, S., Blank, R., Lamb, E., regulatory function in human blood. J.
Natarajan, S., Kastenmayer, R., Hunter, Exp. Med. 2009; 206:1009-1017.
C., Grigg, M.E., Belkaid, Y. Decrease of 64. Kemper, C., Chan, A.C., Green, J.M.,
Foxp3+ Treg cell number and acquisition Brett, K.A., Murphy, K.M., Atkinson, J.P.
of effector cell phenotype during lethal Activation of human CD4+ cells with
infection. Immunity. 2009; 31:772-786. CD3 and CD46 induces a T-regulatory
61. Dominguez-Villar, M., Baecher-Allan, cell 1 phenotype. Nature 2003; 421:388-
C.M., Hafler, D.A. Identification of T 392.

CHAPTER 5

117
 6
CHAPTER
Summarizing discussion
 SUMMARIZING DISCUSSION

Our immune system is efficiently orchestrated to protect us from disease. Innate

and adaptive immunity cooperate to eradicate pathogens and tumour cells, while
maintaining homeostasis in absence of danger. To terminate immune responses
after pathogen clearance and to prevent excessive immunopathology, active
tolerance induction and/ or maintenance is essential. When tolerance fails, immune-
mediated inflammatory diseases can occur for which immune suppressive therapies
might be necessary. Suppressive dendritic cells (DC) and T cells play a crucial role
in maintaining immune homeostasis and preventing immunopathology, and may
thereby be ideal for specific tolerance inducing therapies. Tolerance inducing
therapy can be applied to autoimmune diseases and allergies, where the immune
system over-reacts against self and innocuous foreign antigens. But also patients
that need transplantation might benefit from immune suppressive therapy to
prevent rejection caused by immune recognition of non-self MHC molecules and/
or antigens.

Potential of immunosuppressive therapy with tolerogenic IL-10 DC

Because DC are key in regulating immune responses, they possess immense
therapeutic potential for treatment of a variety of immune disorders. DC can be
generated in vitro and delivered to patients to correct dysfunctional immune
responses, as has already been performed for over a decade in clinical trials with
cancer patients.1 In chapter 3 we compared different types of tolerogenic (t)DC
generated with clinically-applicable protocols with the aim to identify the best-suited
tDC for cellular immunosuppressive therapy. In order to effectively induce immune
suppression, tolerogenic DC must possess some essential functional characteristics,
i.e. induction of regulatory T cells (Treg), migration to the site where they can induce
Treg, and a stable tolerogenic phenotype and function. We demonstrated that
CCR7-directed migration towards CCL21 as well as a stable phenotype and function
was achieved by additional maturation stimuli in combination with a tolerogenic
stimulus. IL-10-generated DC were superior tolerogenic DC compared to TGFβ-,
vitamin D3-, dexamethasone- or rapamycin-generated DC, as they produced high
amounts of IL-10, inhibited memory T cell proliferation and potently induced Treg
from naïve T cells. IL-10 tDC therefore show great promise as cellular therapy to
induce immune suppression.
We analysed Treg-inducing potential as most important immunosuppressive
function. Treg induction by tDC is crucial for peripheral tolerance as Treg can
subsequently induce more widespread tolerance by suppressing T cell responses
via secretion of anti-inflammatory cytokines, cytotoxicity, induction of infectious
tolerance, or mediating metabolic distress.2-5 In addition, the memory population
of Treg will provide long lasting immune suppression.6 Furthermore, Treg not only
CHAPTER 6

suppress T cell responses, but also suppress APC by downregulating co-stimulatory

molecules, inhibiting maturation and upregulating inhibitory molecules.7-12 Thus,
tDC vaccination may provide a self-maintaining regulatory loop in which tDC induce
Treg and Treg subsequently are able to program the generation of tDC.

121
 CHAPTER 6

Various animal models have provided proof of principle of the therapeutic

benefits of IL-10-modulated tDC. IL-10 tDC protected against autoimmune
diseases 13,14, graft-versus-host disease 15 and asthma.16 Tolerance showed to
be T cell-mediated, via anergy induction 14 or Treg induction 13,16, indicating the
importance of induction of T cell tolerance for tDC therapy. Furthermore, ex vivo
generated IL-10 tDC from monocytes of asthmatic donors were tolerogenic as they
suppressed T helper (Th)2 responses and induced functional Treg.17 These studies
show the great potential of IL-10 tDC in tolerance induction protocols in various
disease models.
Despite the promising results in animal models, difficulties on the way towards
clinical practice must be overcome.18-20 Amongst others these include; defining
which DC subset to be used for a particular application, standardizing protocols
for tDC generation, identifying the optimal administration route, number of DC,
frequency of administration, and which antigens to be used for antigen-specific
tolerance induction. Further translational studies and clinical trials are needed to
examine how tDC therapy can be implemented. To date, the first phase I safety
studies with tDC have started. Two ongoing studies aim at treatment of rheumatoid
arthritis with vitamin D3/ dexamethasone generated DC (Newcastle University) and
with NF-κB inhibited DC (University of Queensland). Another study used genetically
engineered immature DC, with low expression of CD40, CD80 and CD86, for type
I diabetes treatment.21 In this study it was shown that tDC were safely tolerated
without adverse effects. These pioneering clinical trials, together with DC vaccination
studies for cancer therapy, will provide more insight about optimal DC generation
and administration protocols.

Measuring regulatory T cell induction with an in vitro suppression

assay
Studying human regulatory T cells is technically difficult, because markers for
human induced Treg are either lacking or often also expressed by recently activated
T cells. Since a consensus phenotype that uniquely defines Treg does not exist, the
best way to identify human Treg is by functional activity. For this we developed a
suppression assay in chapter 2. With this assay we are able to measure suppression
of proliferation of activated responder T cells, which we use as readout for the
suppressive capacity of Treg. Suppression assays can be performed in several ways
and some parameters are critical for reliable readout of the assay.
Whereas suppression of proliferation of CD4+ responder T cells is most used
to examine Treg function, it is important to acknowledge that Treg cells possess
other mechanisms of suppression and may target other cell types. For example,
suppression of APC, CD8+ T cells and B cells is another major mechanism of Treg
CHAPTER 6

suppressive function.8,22,23 Suppression assays can be performed with antigen

presenting cells (APC) as stimulation for responder T cell proliferation, or with
polyclonal stimulation of CD3 and CD28. The type of stimulation could have a major
effect on the outcome of the assay. We chose to use mature DC as stimulation, since
Treg cells can suppress APC function, for example via engagement of inhibitory

122
 SUMMARIZING DISCUSSION

molecules LAG-3 24 or CTLA-4 25 to MHC class II or CD80/ CD86, respectively,

and thereby suppress proliferation of responder T cells. If APC are not included
in the suppression assay, this possible mechanism would be missed. However, to
determine whether Treg exert their suppressive function via DC or directly suppress
responder T cells, the use of both DC stimulation and polyclonal stimulation can
provide insight into this.
For reliable readout of the suppression assay, we established that it is essential
to use a negative control condition to rule out non-specific suppression. We used a
negative control condition of naïve CD4+ T cells primed by mature DC. This negative
control is necessary to exclude cell death of responder T cells, due to overgrowth,
IL-2 depletion or nutrient consumption. In addition, non-specific suppression can be
induced by cell death of primed T cells, as the presence of dying cells may lead to
reduced proliferation and can be erroneously interpreted as suppression.26
We found that the amount of stimulation, i.e. the number of mature DC to
stimulate responder T cells, is essential to optimally determine the suppressive
capacity of primed T cells. Depending on the strength of stimulating responder
T cells, Treg cells may appear more or less suppressive.27 This may be explained
by altered sensitivity to suppression of responder T cells, as strongly activated
responder T cells are more resistant to Treg-mediated suppression.27,28
Thus, for optimal and reliable readout of a suppression assay, one should consider
the type and strength of stimulation to stimulate responder T cell proliferation,
and include a negative control condition to rule out non-specific suppression. A
functional suppression assay will be of great value for evaluating Treg suppressive
capacity as a measure for the therapeutic efficacy of ex vivo generated tDC or ex
vivo induced Treg. For cellular therapy, it is preferable to induce only relevant Treg
with certain antigen specificity involved in the targeted disease, to prevent general
immune suppression and reduce the risk of unwanted side effects. The in vitro
suppression assay can indeed be used to determine antigen specificity of induced
Treg (see below).

Antigen-specific immune suppression

In chapters 2 and 3 we investigated Treg induction by tDC in an allogeneic in vitro
suppression assay, using DC from one donor and T cells, both primed T cells and
responder T cells, from another donor. This is a useful approach for determining
the effect of donor tDC on recipient Treg induction that will suppress recipient
T cell responses. Allorecognition after solid organ transplantation occurs in three
different ways; I) donor DC that migrate from the graft and present donor MHC
to recipient T cells (direct pathway), leading to acute graft rejection, II) recipient
DC that present allogeneic donor antigens to recipient T cells (indirect pathway),
CHAPTER 6

leading to chronic graft rejection, and III) recipient DC that present donor MHC,
which can be transferred from donor DC to recipient DC, to recipient T cells
(semi-direct pathway).29 To target the pathways of allorecognition, either donor-
or recipient-derived tDC can be used, which both have been shown to induce
indefinite murine allograft survival.30 Use of recipient tDC loaded with donor

123
 CHAPTER 6

alloantigen may be more promising, since donor tDC can only be used in case of
live donor renal or liver transplantation. In addition, donor DC can be recognized
as foreign and thereby be taken up, processed and presented to recipient T cells,
risking sensitization of the indirect pathway.18 If antigen-specific tolerance can
be established, there should be no major impact on host immune responses
against tumour cells or pathogen infections. As the antigens involved in some
of the autoimmune diseases and allergies are known, it is possible to generate
antigen-specific tDC by ‘loading’ the DC with peptides or whole protein. We have
explored (allo)antigen-specificity of IL-10 tDC-induced Treg by using 3rd party
donors or model antigens and autologous T cells. Our preliminary results look
promising, but more work is needed to conclude antigen specificity of IL-10 tDC-
induced Treg. For type I diabetes and multiple sclerosis (MS) the autoantigens are
delineated; islet β cell components (including insulin) and myelin, respectively.31,32
A recent report demonstrated that loading of in vitro generated tDC from MS
patients with myelin peptide resulted in antigen-specific hyporesponsiveness
of autologous T cells in an in vitro assay.33 However, in many autoimmune
diseases, for example in rheumatoid arthritis (RA), the target antigens have not
been delineated.20 Several potential autoantigens may be used to load DC, but
responses to these antigens could only be detected in subgroups of patients. For
the clinical trial at the University of Queensland, tDC are pulsed with a mixture of
four citrullinated peptide antigens and used to treat patients who have anti-cyclic
citrullinated peptide antibodies.34

Regulatory T cell induction by tolerogenic IL-10 DC

In chapter 2 we showed that IL-10 tDC-induced Treg suppressed the proliferation
of responder T cells in a cell contact-dependent manner, suggesting that interaction
between Treg and the responder T cells or the stimulatory DC is necessary for
suppression. Since IL-10 tDC produced relatively high amounts of TNFα, we
studied the effect of TNFα on Treg induction with IL-10 tDC while simultaneously
neutralizing TNFα (chapter 4). We identified a strong regulatory gene profile
in these induced Treg. Inhibitory molecules that were highly upregulated are;
galectin-3, legumain, GARP and LAG-3, which may mediate suppressive action on
responder T cells. Galectin-3, legumain and GARP are part of a regulatory network
and are key molecules for the suppressive function of natural Treg.35 Whether
galectin-3, an intracellular lectin 36, and legumain, an endopeptidase localized in the
endosomal/ lysosomal compartments 37, would have a (direct) effect in cell contact-
mediated suppression is unclear. GARP is a transmembrane glycoprotein of which
the extracellular domain is important for suppression 35, suggesting that interaction
with other signalling molecules is necessary for (cell contact-mediated) suppression.
CHAPTER 6

The CD4 homolog LAG-3 has been shown to inhibit DC maturation via engagement
to MHC class II 24 and may thereby induce suppression via DC.
Thus, IL-10 tDC induce Treg that suppress in a cell contact-dependent manner.
Neutralizing TNFα during Treg induction by IL-10 tDC enhances gene expression of
inhibitory molecules that may be involved in cell contact-dependent suppression,

124
 SUMMARIZING DISCUSSION

such as GARP and LAG-3. It will be interesting to study the role of these inhibitory
molecules in Treg-mediated suppression. Furthermore, to test whether Treg exert
their suppressive function via DC or directly via responder T cells, a suppression
assay in which T cells are stimulated with polyclonal CD3/ CD28 stimulation instead
of DC may be performed. If suppression is mediated via inhibition of DC then CD3/
CD28-stimulated responder T cells will not be suppressed.

Anti-TNFα therapy and regulation of T cell responses

In chapter 4 we studied the effect of TNFα neutralization on priming of naïve
CD4+ T cells. Anti-TNFα induced IL-10+ T cells, inhibited IFNγ polarization,
inhibited T cell activation and survival, and enhanced a regulatory phenotype
and function. This shows that neutralizing TNFα results in a more regulated T cell
response. CD4+ T cells, especially Th1 and Th17 cells, play a dominant role in the
immunopathogenesis of autoimmune diseases.38-40 Impaired regulatory mechanisms
might allow the breakdown of peripheral tolerance, after which a detrimental Th1-/
Th17-driven immune response evolves and proceeds to chronic inflammation. That
Treg play an important role in preventing autoimmune diseases was proven by
mouse studies where removal of CD4+CD25+ T cells lead to the development of
various autoimmune diseases 41, and adoptive transfer of this population effectively
prevented the development of experimentally induced autoimmune diseases.42
Inhibiting the activation of effector Th cells and simultaneously upregulating Treg
responses as we show in chapter 4 may be one of the beneficial effects seen with
anti-TNFα therapy.
In RA patients, it was shown that peripheral blood Treg function is severely
impaired and anti-TNFα therapy showed clinical improvement and induction/
expansion of Treg with a restored function.43-45 However, controversy exists with
regard to the numbers and function of human Treg cells in peripheral blood of
RA patients.46,47 In the inflamed synovial tissue it appears that elevated numbers
of Treg cells are present, which were fully functional and suppressed proliferation
and cytokine production of autologous CD4+CD25- T cells in vitro.46,48,49 The
general approach of developing strategies to replace/ induce Treg cells neglects to
acknowledge that Treg are present at the site of inflammation.50 The main question
is why autoimmunity develops if fully functional Treg are present in abundance. What
might contribute to this apparent paradox is the fact that inflammatory mediators
may inhibit Treg function. It has been shown that IL-7, IL-15 and TNFα abrogated
the suppressive abilities of Treg 45,51-53, which indicates that inflammatory cytokines
present in inflamed tissue can interfere with Treg activity or with the susceptibility
to suppression of responder T cells. A study in an EAE mouse model indicated that
the local inflammatory cytokine milieu of the target tissue needs to be controlled
CHAPTER 6

for Treg to exert their suppressive function.54 Thus, neutralizing pro-inflammatory

cytokines like TNFα will contribute to tolerance induction, as anti-TNFα enhances
the induction of Treg as well as the suppressive function of Treg. This also indicates
that Treg inducing therapy, e.g. with tDC vaccination, will be most efficacious when
combined with anti-TNFα treatment.

125
 CHAPTER 6

Anti-TNFα and regulation of immune cell trafficking

Neutralizing TNFα regulates the T cell response towards immune suppression
(chapter 4). However, the question remains whether anti-TNFα therapy affects the
T cells that are present at the sites of inflammation, for example the rheumatic
joint. In other words; will the peripheral blood Treg that are induced/ expanded
by anti-TNFα 43-45 reach the sites of inflammation? Guiding the expression of
chemokine receptors by Treg might lead to faster or more specific migration of
these cells to inflamed tissues. In addition, chemokine secretion by Treg itself
might regulate the chemoattraction of specific cell subsets. We observed a change
in chemokine and chemokine receptor gene profiles upon TNFα neutralization
(chapter 4). CC-chemokine receptor (CCR)2 was highly upregulated on anti-TNFα-
treated T cells (Table I). CCR2 has been linked to autoimmune diseases because
it leads to chemotaxis of monocytes, NK cells and T cells to CCL2 that is present
at the sites of inflammation.55 Interference of this chemoattraction is now being
studied to possibly inhibit progression of disease. Conversely, upregulation of CCR2
expression on Treg cells might provide a means of trafficking to and suppressing
inflammation at inflamed sites where CCL2 is present. In a mouse model of
spontaneous development of generalized autoimmune disease, CCR2-transfected
Treg showed more pronounced CCL2-directed migration to the sites of inflammation
and ameliorated disease progression.56
In addition, anti-TNFα treated T cells expressed higher levels of various
chemokines, e.g. IL-8, CCL2 and CCL18 (Table I). Natural Treg were shown to produce
IL-8 and CCL2 57, suggesting a role in the orchestration of immune cell migration.

Table I – Chemokine (receptor) gene profile in anti-TNFα-treated T cells compared

to non-treated T cells
Gene symbol Genbank ID Gene FC* log FC**
CCL13 NM_005408.2 Chemokine (C-C motif) ligand 13 (CCL13) 23,73 4,57
CCL2 NM_002982.3 Chemokine (C-C motif) ligand 2 (CCL2) 14,88 3,90
CCL8 NM_005623.2 Chemokine (C-C motif) ligand 8 (CCL8) 7,64 2,93
Chemokine (C-C motif) ligand 18 (pulmonary
CCL18 NM_002988.2 5,98 2,58
and activation-regulated) (CCL18)
CXCL2 NM_002089.3 Chemokine (C-X-C motif) ligand 2 (CXCL2) 4,74 2,25
Chemokine (C-C motif) receptor 2 (CCR2),
CCR2 NM_000647.3 4,54 2,18
transcript variant A
IL8 NM_000584.2 Interleukin 8 (IL8) 3,00 1,59
Chemokine (C-X-C motif) ligand 13
CXCL13 NM_006419.1 -4,01 -2,00
CHAPTER 6

(B-cell chemoattractant) (CXCL13)

CCL1 NM_002981.1 Chemokine (C-C motif) ligand 1 (CCL1) -7,13 -2,83

* Numbers represent the mean fold change (FC) in gene expression in anti-TNFα-treated T cells
versus non-treated T cells of 3 independent donors
** Numbers represent the mean log2 fold change in gene expression

126
 SUMMARIZING DISCUSSION

CCL2 is a ‘Th2 chemokine’ that stimulates Th2 cell polarization 58, which might
mediate immune deviation away from Th1 and Th17. CCL18 displays tolerance-
inducing properties as it differentiates monocyte-derived DC into tolerogenic DC 59
and generates Treg from memory T cells.60
Thus, altering the trafficking of Treg and chemoattraction by Treg may modulate
immune responses by determining which accessory cells the Treg will encounter.
Induction or reactivation of Treg in patients with autoimmune diseases may restore
the balance of the immune response and halt the inflammatory process, thereby
preventing further disease. Before Treg induction/ targeting is to be used in the
clinic, future studies will have to focus on how Treg cells traffic to the inflamed tissues
and how the micro-environment within inflamed tissues influences their function.

Anti-TNFα and CD4+ T cell survival

Neutralizing TNFα not only affects T cell polarization and Treg induction, as
described above, but also inhibits T cell activation and survival (chapter 4). The
effect of anti-TNFα coincided with the kinetics of TNFRII expression after naïve CD4+
T cell activation, as TNFα must be neutralized before significant TNFRII expression
for IL-10 induction, anti-TNFα reduced expression of late T cell activation markers,
but not early activation marker CD69, and anti-TNFα reduced T cell survival in the
second phase of DC–T cell co-culture. These results indicate that TNFα has major
effects on T cell polarization, activation and survival in later phases of T cell priming,
when TNFRII is expressed. T cell responses start with a proliferation phase in which
naïve T cells are activated and differentiate into effector T cells. A contraction (or
transition) phase occurs in a later stage, in which a large population of effector
T cells are cleared and a selection of effector T cells transit to memory T cells.61
Depending on the signalling pathway engaged by ligation of TNFα, different
biological responses, such as apoptosis or survival, may ensue. Binding of TNFα
to TNFRI may result in apoptosis via caspase cascades or result in pro-survival
signals via NF-κB, wich is cell- and context-dependent.62 TNFRII signalling induces
cell survival and cell proliferation via NF-κB and MAP kinases.63 Absence of
TNFRII signalling during T cell priming may therefore lead to reduced survival of
the T cells. We observed that neutralizing TNFα during T cell priming resulted in
reduced expression levels of IL-2 gene transcript (data not shown). IL-2 signalling
during priming of CD4+ T cells has been shown to significantly enhance growth,
differentiation and survival of effector T cells.64,65 With IL-2 deficient CD4+ T cells
it was shown that autocrine IL-2 production has an important role in regulating
contraction during influenza infection, as fewer IL-2 deficient T cells were detected
after the contraction phase compared to wild type cells.61 This indicates that TNFα-
induced IL-2 regulates survival of the T cells during the contraction phase. A study
CHAPTER 6

with TNFRII deficient CD8+ T cells indeed demonstrated that TNFRII signalling is
important for IL-2 production and cell survival.66 In addition, TNFα showed to be
important for transition of effector CD8+ T cells into memory T cells.67 Our data
indicate that TNFα has the same role in IL-2 induction and regulating the contraction
phase of CD4+ T cells.

127
 CHAPTER 6

IL-10/ IFNγ co-expressing suppressive T cells

IL-10/ IFNγ co-expressing CD4+ T cells are found in chronic infections.68-73 These cells
are characterized as T-bet+ Th1 cells associated with production of protective IL-10 68,69
and display suppressive function in vitro.72,74-76 In chapter 5 we demonstrated an
extensive regulatory network in IL-10+/ IFNγ+ T cells, with upregulated expression of
CTLA-4, galectin-3, LAG-3, galectin-10, GARP, PD-1 and legumain. In addition, the
expression of various Th1 gene transcripts was downregulated in this population.
GARP, PD-1 and legumain were specifically upregulated in IL-10+/ IFNγ+ T cells when
compared to IL-10 and IFNγ single positive T cells. GARP, legumain and galectin-3
are part of a regulatory network described to induce suppressive function in natural
Treg.35 This demonstrates the uniqueness of this population, with downregulated
Th1/ IFNγ associated genes and specific expression of inhibitory molecules which
are lower expressed by IL-10 or IFNγ single positive T cells.
It is not clear whether IL-10/ IFNγ co-expressing cells originated from Th1 or Treg
cells, or directly differentiated from naïve T cells. The IL-10+/ IFNγ+ T cell population
has been described as either Th1 cells because of T-bet expression 68,69,73,75, or as
regulatory cells because of suppressive function.73-77 However, Th cells and Treg
cells are highly plastic under polarizing conditions and can co-express signature
cytokines and transcription factors.78,79 Many reports have shown IL-10 induction
in Th1 cells under various polarizing conditions.80-85 However, IFNγ induction in
Treg cells is also possible.86-89 In addition, Treg can lose Foxp3 expression under
inflammatory conditions and become activated memory T cells that produce pro-
inflammatory cytokines like IFNγ and IL-17.86,87,90 Switching of Foxp3+ Treg to Th
cells occurred especially in inflamed tissues as was demonstrated in a colitis model
and in NOD mice.90,91 Further studies are needed to determine the origin of this
intriguing T cell population.
Although the origin of IL-10/ IFNγ co-expressing T cells is not clear yet, or may
be variable, the reason for their induction seems more obvious. IL-10 is an anti-
inflammatory cytokine with its most important function to reduce immunopathology
and maintain homeostasis by limiting inflammatory responses.92 Host survival often
depends not only on effector pro-inflammatory cytokines, but also on the expression
of IL-10.92,93 Although IFNγ secretion is essential for control of most intracellular
pathogens, IL-10 is important for limiting immunopathology. Regulatory T cells
are one of the key IL-10 producing cells, although it is now clear that other Th
subsets have been described to produce IL-10, including Th1 cells.93 In this way, the
immune response has evolved to protect the host from a wide range of potentially
pathogenic microorganisms, but parallel mechanisms to control overexuberant
immune responses and prevent reactivity to self are required to limit host damage.
CD4+ T cell plasticity may be advantageous in terms of host defence, i.e. switching
CHAPTER 6

to pro-inflammatory effector Th cells in case of infection/ inflammation, or to IL-10-

producing regulatory cells to control immunopathology. A better understanding
of the signals that control stability and plasticity will have important therapeutic
applications to combat infections and to control autoimmunity.

128
 SUMMARIZING DISCUSSION

Manipulation of regulatory T cell induction

Aberrant Th cell responses can contribute to immunopathology in the context of
infection, autoimmunity, allergy and other inflammatory conditions. Therefore, Th
cell responses must be carefully regulated to ensure that they are initiated only
when appropriate and are efficiently resolved upon pathogen eradication. Specific
immune suppression targeting inflammation only when and where needed may help
in limiting aberrant Th cell responses. Fine-tuning of tDC generation and thereby
fine-tuning of Treg induction may optimize suppressive therapy. In chapter 3 we
showed that generation of tDC with different immunosuppressive compounds
results in differential phenotypic and functional properties by tDC, i.e. maturation
status, cytokine production, T cell stimulatory capacity and Treg induction. The
type of immunosuppressive compound, but also the type of maturation stimulus,
leaves room for fine-tuning of tDC generation and the resulting T cell response.
As maturation stimulus, we compared a cytokine cocktail, consisting of TNFα, IL-1β
and PGE2, with MPLA, which is a synthetic form of LPS (chapter 3). This showed that
maturation of tDC with the cytokine cocktail induced CCR7 expression and migration
towards CCL21, whereas MPLA maturation did not show migration of tDC. Because
de novo Treg induction in secondary lymphoid organs is an important mechanism
of tolerance induction by tDC, the cytokine cocktail is more optimal for achieving
Treg induction. In a study comparing vitamin D3 tDC and dexamethasone tDC, it
was demonstrated that both tDC induced IL-10+ Treg, however only vitamin D3
tDC-induced Treg displayed antigen-specific suppression.94 Another study showed
that TNFα tDC, dexamethasone tDC and IL-10 tDC all prevented collagen-induced
arthritis when injected before disease induction, but their mechanisms of tolerance
induction differed.95 All tDC induced IL-10+ T cells, however TNFα tDC also specifically
induced IL-5+ T cells, whereas IL-10 tDC reduced IFNγ+ T cells, leading to altered
Th cell balances. In addition, TNFα- and IL-10-modulated DC showed a decrease
in IgG2a antibody titers in an antigen-specific manner, whereas the effects induced
by dexamethasone-stimulated DC were not antigen specific. This indicates that
different tDC display differential tolerogenic features that may be useful in different
applications.In addition to tDC generation, the inflammatory environment can also
be manipulated in order to improve immune suppressive therapy and fine-tune Treg
induction. As described above, neutralizing TNFα regulates the immune response
by inducing IL-10+ T cells, reducing IFNγ polarization and enhancing suppressive
function of tDC-induced Treg (chapter 4). Anti-TNFα induced a regulatory gene
profile with upregulated expression of inhibitory molecules galectin-3, legumain,
GARP and LAG-3, which may contribute to more specific suppressive function
by Treg cells.24,35 Furthermore, anti-TNFα-treated T cells displayed an altered
chemokine and chemokine receptor gene expression profile (Table I), possibly
CHAPTER 6

regulating chemoattraction of accessory cells. This indicates that fine-tuning of

T cell responses is possible by regulating the inflammatory environment which may
lead to specific expression of cytokines, chemokines or inhibitory molecules, and
thereby may lead to better clinical outcome.

129
 CHAPTER 6

Polarizing factors can drive Treg into functionally different subsets 79 or drive
CD4+ Th cells from one subset to another.78 This T cell plasticity seen in vivo and in
vitro can be used to influence T cell responses by regulating the (cytokine) micro-
environment with tDC or anti-TNFα. The plasticity of CD4+ T cells leaves room
for fine-tuning T cell responses, but the highly plastic nature of CD4+ T cells also
makes it necessary to take into account the stability of induced Treg before going
into clinical practice.

Concluding remarks
This thesis describes the role of tolerogenic DC and anti-TNFα agents in
tolerance induction. IL-10-generated tDC potently induce Treg, while inhibiting
CD4+ T cell proliferation and cytokine production by Th1 and Th2 cell subsets.
Anti-TNFα shares this dual function; inducing IL-10 production and a regulatory
phenotype and function in naïve CD4+ T cells, and at the same time counteracting
CD4+ effector T cell priming by inhibiting activation status, survival and IFNγ
production. Anti-TNFα upregulates inhibitory molecules galectin-3, legumain,
GARP and LAG-3 which may be involved in (cell contact-dependent) suppressive
function, as is described for natural Treg. In addition, chemokine and chemokine
receptor expression are altered after TNFα neutralization, suggesting a change in
chemoattraction by Treg and chemotaxis of Treg, thereby regulating the encounter
with other immune cells. The picture emerges that tDC and anti-TNFα agents
synergistically enhance immune suppression (Figure 1). Combination therapy of
tDC and anti-TNFα agents may therefore improve tolerance induction therapy in
patients with TNFα-mediated inflammatory diseases such as rheumatoid arthritis,
ankylosing spondylitis, psoriatic arthritis, multiple sclerosis and inflammatory
bowel diseases.
CHAPTER 6

130
 SUMMARIZING DISCUSSION

Activation
Survival CD4+ Th

tDC vaccination IFNγ

& anti-TNFα
Proliferation

Maturation 96-98

IL-10

IL-10 GARP

Chemoattraction

CCR2 Suppression 35

IL-10 tDC
Treg induction Galectin-3 Legumain
IL-10

MHCII CD4+ iTreg

LAG-3

APC suppression 24
ILT3/ ILT4

Figure 1. Tolerance-inducing potential of IL-10 tDC and anti-TNFα. Generation of DC with

the immunosuppressive cytokine IL-10 induces potent tolerogenic features. IL-10 tDC secrete
high amounts of IL-10, express inhibitory molecules ILT3 and ILT4, suppress proliferation and
IFNγ production of CD4+ T cells, while enhancing IL-10 production. In addition, priming of
naïve CD4+ T cells with IL-10 tDC induces Treg that potently suppress the proliferation of
memory CD4+ T cells. Neutralizing TNFα with anti-TNFα therapeutic antibodies during naïve
CD4+ T cell priming enhances IL-10 expression by T cells, while reducing IFNγ expression,
T cell activation and T cell survival. Anti-TNFα may alter chemoattractive protential of T cells
by alterting chemokine and chemokine receptor profile. In addition, neutralizing TNFα
induces Treg that express inhibitory molecules LAG-3, GARP, galectin-3 and legumain, by
which they may mediate their suppressive function, either directly on T cells or via DC. Th;
T helper cell, iTreg; induced regulatory T cell, tDC; tolerogenic DC, ILT3/4; immunoglobulin-
like transcript 3/4.
CHAPTER 6

131
 CHAPTER 6

References
1. Figdor, C.G., de Vries, I.J., Lesterhuis, Cutting edge: human CD4+CD25+
W.J., Melief, C.J.M. Dendritic cell T cells restrain the maturation and
immunotherapy: mapping the way. Nat antigen-presenting function of dendritic
Med 2004; 10:475-480. cells. J. Immunol. 2004; 172:4676-4680.
2. Shevach, E.M. Mechanisms of Foxp3+ 12. Cederbom, L., Hall, H., Ivars, F.
T regulatory cell-mediated suppression. CD4+CD25+ regulatory T cells down-
Immunity 2009; 30:636-645. regulate co-stimulatory molecules
3. Vignali, D.A.A., Collison, L.W., Workman, on antigen-presenting cells. Eur. J.
C.J. How regulatory T cells work. Nat. Immunol. 2000; 30:1538-1543.
Rev. Immunol. 2008; 8:523-532. 13. Tai, N., Yasuda, H., Xiang, Y., Zhang,
4. Andersson, J., Tran, D.Q., Pesu, M., L., Rodriguez-Pinto, D., Yokono, K.,
Davidson, T.S., Ramsey, H., O’Shea, J.J., Sherwin, R., Wong, F.S., Nagata, M.,
Shevach, E.M. CD4+ FoxP3+ regulatory Wen, L. IL-10-conditioned dendritic cells
T cells confer infectious tolerance in a prevent autoimmune diabetes in NOD
TGF-beta-dependent manner. J Exp. and humanized HLA-DQ8/RIP-B7.1
Med. 2008; 205:1975-1981. mice. Clin. Immunol. 2011; 139:336-349.
5. Collison, L.W., Workman, C.J., Kuo, T.T., 14. Perona-Wright, G., Anderton, S.M.,
Boyd, K., Wang, Y., Vignali, K.M., Cross, Howie, S.E.M., Gray, D. IL-10 permits
R., Sehy, D., Blumberg, R.S., Vignali, transient activation of dendritic cells to
D.A.A. The inhibitory cytokine IL-35 tolerize T cells and protect from central
contributes to regulatory T-cell function. nervous system autoimmune disease.
Nature 2007; 450:566-569. Int. Immunol. 2007; dxm084.
6. Tosello, V., Odunsi, K., Souleimanian, 15. Sato, K., Yamashita, N., Baba, M.,
N.E., Lele, S., Shrikant, P., Old, L.J., Matsuyama, T. Modified myeloid
Valmori, D., Ayyoub, M. Differential dendritic cells act as regulatory dendritic
expression of CCR7 defines two distinct cells to induce anergic and regulatory
subsets of human memory CD4+CD25+ T cells. Blood 2003; 101:3581-3589.
Tregs. Clinical Immunology 2008; 16. Lu, M., Dawicki, W., Zhang, X., Huang,
126:291-302. H., Nayyar, A., Gordon, J.R. Therapeutic
7. Mahnke, K., Bedke, T., Enk, A.H. induction of tolerance by IL-10-
Regulatory conversation between differentiated dendritic cells in a mouse
antigen presenting cells and regulatory model of house dust mite-asthma.
T cells enhance immune suppression. Allergy 2011; 66:612-620.
Cell Immunol 2007; 250:1-13. 17. Li, X., Yang, A., Huang, H., Zhang, X.,
8. Onishi, Y., Fehervari, Z., Yamaguchi, T., Town, J., Davis, B., Cockcroft, D.W.,
Sakaguchi, S. Foxp3+ natural regulatory Gordon, J.R. Induction of type 2 T
T cells preferentially form aggregates on helper cell allergen tolerance by IL-
dendritic cells in vitro and actively inhibit 10-differentiated regulatory dendritic
their maturation. Proc. Natl. Acad. Sci. U. cells. Am. J. Respir. Cell Mol. Biol. 2010;
S. A 2008; 105:10113-10118. 42:190-199.
9. Kleijwegt, F.S., Laban, S., Duinkerken, 18. van Kooten, C., Lombardi, G.,
G., Joosten, A.M., Koeleman, B.P.C., Gelderman, K.A., Sagoo, P., Buckland,
Nikolic, T., Roep, B.O. Transfer of M., Lechler, R., Cuturi, M.C. Dendritic
Regulatory Properties from Tolerogenic cells as a tool to induce transplantation
to Proinflammatory Dendritic Cells via tolerance: obstacles and opportunities.
Induced Autoreactive Regulatory T Cells. Transplantation 2011; 91:2-7.
J Immunol 2011; 187:6357-6364. 19. Hu, J., Wan, Y. Tolerogenic dendritic
10. Min, W.P., Zhou, D., Ichim, T.E., Strejan, cells and their potential applications.
G.H., Xia, X., Yang, J., Huang, X., Garcia, Immunology 2011; 132:307-314.
CHAPTER 6

B., White, D., Dutartre, P., Jevnikar, 20. Stoop, J.N., Harry, R.A., von, D.A.,
A.M., Zhong, R. Inhibitory feedback loop Isaacs, J.D., Robinson, J.H., Hilkens,
between tolerogenic dendritic cells and C.M. Therapeutic effect of tolerogenic
regulatory T cells in transplant tolerance. dendritic cells in established collagen-
J. Immunol. 2003; 170:1304-1312. induced arthritis is associated with a
11. Misra, N., Bayry, J., Lacroix-Desmazes, reduction in Th17 responses. Arthritis
S., Kazatchkine, M.D., Kaveri, S.V. Rheum. 2010; 62:3656-3665.

132
 SUMMARIZING DISCUSSION

21. Giannoukakis, N., Phillips, B., Finegold, autoimmune diabetes. Curr. Opin.
D., Harnaha, J., Trucco, M. Phase I Immunol. 2011; 23:739-745.
(safety) study of autologous tolerogenic 32. Bielekova, B., Sung, M.H., Kadom, N.,
dendritic cells in type 1 diabetic patients. Simon, R., McFarland, H., Martin, R.
Diabetes Care 2011; Epub ahead of Expansion and functional relevance of
print. high-avidity myelin-specific CD4+ T cells
22. Mempel, T.R., Pittet, M.J., Khazaie, in multiple sclerosis. J. Immunol. 2004;
K., Weninger, W., Weissleder, R., 172:3893-3904.
von Boehmer, H., von Andrian, U.H. 33. Raich-Regue, D., Grau-Lopez, L.,
Regulatory T Cells Reversibly Suppress Naranjo-Gomez, M., Ramo-Tello, C.,
Cytotoxic T Cell Function Independent Pujol-Borrell, R., Martinez-Caceres, E.,
of Effector Differentiation. Immunity
Borras, F.E. Stable antigen-specific T-
2006; 25:129-141.
cell hyporesponsiveness induced by
23. Eddahri, F., Oldenhove, G., Denanglaire, tolerogenic dendritic cells from multiple
S.+., Urbain, J., Leo, O., Andris, F. sclerosis patients. Eur. J. Immunol. 2012;
CD4+CD25+ regulatory TGÇäcells 42:771-782.
control the magnitude ofT-dependent
humoral immune responses to 34. Khan, S., Greenberg, J.D., Bhardwaj,
exogenous antigens. Eur. J. Immunol. N. Dendritic cells as targets for therapy
2006; 36:855-863. in rheumatoid arthritis. Nat. Rev.
Rheumatol. 2009; 5:566-571.
24. Liang, B., Workman, C., Lee, J., Chew,
C., Dale, B.M., Colonna, L., Flores, M., 35. Probst-Kepper, M., Geffers, R., Kroger,
Li, N., Schweighoffer, E., Greenberg, A., Viegas, N., Erck, C., Hecht, H.J.,
S., Tybulewicz, V., Vignali, D., Clynes, Lunsdorf, H., Roubin, R., Moharregh-
R. Regulatory T cells inhibit dendritic Khiabani, D., Wagner, K., Ocklenburg,
cells by lymphocyte activation gene-3 F., Jeron, A., Garritsen, H., Arstila, T.P.,
engagement of MHC class II. J. Immunol. Kekalainen, E., Balling, R., Hauser, H.,
2008; 180:5916-5926. Buer, J., Weiss, S. GARP: a key receptor
25. Grohmann, U., Orabona, C., Fallarino, controlling FOXP3 in human regulatory
F., Vacca, C., Calcinaro, F., Falorni, A., T cells. J. Cell Mol. Med. 2009; 13:3343-
Candeloro, P., Belladonna, M.L., Bianchi, 3357.
R., Fioretti, M.C., Puccetti, P. CTLA-4-Ig 36. Hsu, D.K., Chen, H.Y., Liu, F.T. Galectin-3
regulates tryptophan catabolism in vivo. regulates T-cell functions. Immunol. Rev.
Nat. Immunol. 2002; 3:1097-1101. 2009; 230:114-127.
26. McMurchy, A.N., Levings, M.K. 37. Watts, C., Matthews, S.P., Mazzeo, D.,
Suppression assays with human T Manoury, B.+., Moss, C.X. Asparaginyl
regulatory cells: a technical guide. Eur. J. endopeptidase: case history of a
Immunol. 2012; 42:27-34. class II MHC compartment protease.
27. Baecher-Allan, C., Viglietta, V., Immunological Reviews 2005; 207:218-
Hafler, D.A. Inhibition of Human 228.
CD4+CD25+high Regulatory T Cell 38. Skapenko, A., Leipe, J., Lipsky, P.E.,
Function. J Immunol 2002; 169:6210- Schulze-Koops, H. The role of the T cell
6217. in autoimmune inflammation. Arthr. Res.
28. George, T.C., Bilsborough, J., Viney, J.L., Ther. 2005; 7 Suppl 2:S4-14.
Norment, A.M. High antigen dose and 39. Chabaud, M., Lubberts, E., Joosten, L.,
activated dendritic cells enable Th cells van Den, B.W., Miossec, P. IL-17 derived
to escape regulatory T cell-mediated from juxta-articular bone and synovium
suppression in vitro. Eur. J. Immunol. contributes to joint degradation in
2003; 33:502-511. rheumatoid arthritis. Arthritis Res. 2001;
29. Ezzelarab, M., Thomson, A.W. 3:168-177.
Tolerogenic dendritic cells and their 40. Canete, J.D., Martinez, S.E., Farres,
role in transplantation. Semin. Immunol. J., Sanmarti, R., Blay, M., Gomez,
CHAPTER 6

2011; 23:252-263. A., Salvador, G., Munoz-Gomez, J.

30. Morelli, A.E., Thomson, A.W. Tolerogenic Differential Th1/Th2 cytokine patterns
dendritic cells and the quest for in chronic arthritis: interferon gamma
transplant tolerance. Nat. Rev. Immunol. is highly expressed in synovium of
2007; 7:610-621. rheumatoid arthritis compared with
31. Haskins, K., Cooke, A. CD4 T cells and seronegative spondyloarthropathies.
their antigens in the pathogenesis of Ann. Rheum. Dis. 2000; 59:263-268.

133
 CHAPTER 6

41. Sakaguchi, S., Sakaguchi, N., Asano, rheumatoid arthritis. Eur. J. Immunol.
M., Itoh, M., Toda, M. Immunologic 2003; 33:215-223.
self-tolerance maintained by activated 50. Mellanby, R.J., Thomas, D.C., Lamb, J.
T cells expressing IL-2 receptor alpha- Role of regulatory T-cells in autoimmunity.
chains (CD25). Breakdown of a single Clinical Science 2009; 116:639-649.
mechanism of self-tolerance causes 51. van Amelsfort, J.M.R., van Roon, J.A.G.,
various autoimmune diseases. J Immunol Noordegraaf, M., Jacobs, K.M.G.,
1995; 155:1151-1164. Bijlsma, J.W.J., Lafeber, F.P.J.G.,
42. McHugh, R.S., Shevach, E.M. Cutting Taams, L.S. Proinflammatory mediator-
Edge: Depletion of CD4+CD25+ induced reversal of CD4+, CD25+
regulatory T cells is necessary, but not regulatory T cell-mediated suppression
sufficient, for induction of organ-specific in rheumatoid arthritis. Arthritis &
autoimmune disease. J Immunol 2002; Rheumatism 2007; 56:732-742.
168:5979-5983. 52. Ruprecht, C.R., Gattorno, M., Ferlito, F.,
43. Nadkarni, S., Mauri, C., Ehrenstein, Gregorio, A., Martini, A., Lanzavecchia,
M.R. Anti-TNF-alpha therapy induces A., Sallusto, F. Coexpression of CD25
a distinct regulatory T cell population and CD27 identifies FoxP3+ regulatory
in patients with rheumatoid arthritis via T cells in inflamed synovia. J. Exp. Med.
TGF-beta. J Exp. Med. 2007; 204:33-39. 2005; 201:1793-1803.
44. Ehrenstein, M.R., Evans, J.G., Singh, A., 53. Nagar, M., Jacob-Hirsch, J., Vernitsky,
Moore, S., Warnes, G., Isenberg, D.A., H., Berkun, Y., Ben-Horin, S., Amariglio,
Mauri, C. Compromised function of N., Bank, I., Kloog, Y., Rechavi, G.,
regulatory T cells in rheumatoid arthritis Goldstein, I. TNF activates a NF-
and reversal by anti-TNFalpha therapy. kappa B-regulated cellular program in
J. Exp. Med. 2004; 200:277-285. human CD45RA- regulatory T cells that
45. Valencia, X., Stephens, G., Goldbach- modulates their suppressive function. J.
Mansky, R., Wilson, M., Shevach, E.M., Immunol. 2010; 184:3570-3581.
Lipsky, P.E. TNF downmodulates the 54. Korn, T., Reddy, J., Gao, W., Bettelli, E.,
function of human CD4+CD25hi T- Awasthi, A., Petersen, T.R., Backstrom,
regulatory cells. Blood 2006; 108:253- B.T., Sobel, R.A., Wucherpfennig,
261. K.W., Strom, T.B., Oukka, M., Kuchroo,
46. van Amelsfort, J.M., Jacobs, K.M., V.K. Myelin-specific regulatory T cells
Bijlsma, J.W., Lafeber, F.P., Taams, L.S. accumulate in the CNS but fail to control
CD4(+)CD25(+) regulatory T cells in autoimmune inflammation. Nat Med
rheumatoid arthritis: differences in the 2007; 13:423-431.
presence, phenotype, and function 55. Rafei, M., Galipeau, J. A CCL2-based
between peripheral blood and synovial fusokine as a novel biopharmaceutical
fluid. Arthritis Rheum. 2004; 50:2775- for the treatment of CCR2-driven
2785. autoimmune diseases. Crit Rev.
47. Lawson, C.A., Brown, A.K., Bejarano, Immunol. 2010; 30:449-461.
V., Douglas, S.H., Burgoyne, C.H., 56. Hasegawa, H., Inoue, A., Muraoka, M.,
Greenstein, A.S., Boylston, A.W., Yamanouchi, J., Miyazaki, T., Yasukawa,
Emery, P., Ponchel, F., Isaacs, J.D. Early M. Therapy for pneumonitis and
rheumatoid arthritis is associated with a sialadenitis by accumulation of CCR2-
deficit in the CD4+CD25high regulatory expressing CD4+CD25+ regulatory
T cell population in peripheral blood. T cells in MRL/lpr mice. Arthritis Res.
Rheumatology 2006; 45:1210-1217. Ther. 2007; 9:R15.
48. Liu, M.F., Wang, C.R., Fung, L.L., Lin, 57. Himmel, M.E., Crome, S.Q., Ivison, S.,
L.H., Tsai, C.N. The presence of cytokine- Piccirillo, C., Steiner, T.S., Levings, M.K.
suppressive CD4+CD25+ T cells in Human CD4+ FOXP3+ regulatory T cells
the peripheral blood and synovial fluid produce CXCL8 and recruit neutrophils.
CHAPTER 6

of patients with rheumatoid arthritis. Eur. J. Immunol. 2011; 41:306-312.

Scand. J. Immunol. 2005; 62:312-317. 58. Daly, C., Rollins, B.J. Monocyte
49. Cao, D., Malmstrom, V., Baecher-Allan, chemoattractant protein-1 (CCL2) in
C., Hafler, D., Klareskog, L., Trollmo, C. inflammatory disease and adaptive
Isolation and functional characterization immunity: therapeutic opportunities and
of regulatory CD25brightCD4+ T cells controversies. Microcirculation. 2003;
from the target organ of patients with 10:247-257.

134
 SUMMARIZING DISCUSSION

59. Azzaoui, I., Yahia, S.A., Chang, Y., Th1 cells are the source of IL-10-
Vorng, H., Morales, O., Fan, Y., Delhem, mediated immune suppression in
N., Ple, C., Tonnel, A.B., Wallaert, B., chronic cutaneous leishmaniasis. J. Exp.
Tsicopoulos, A. CCL18 differentiates Med. 2007; 204:285-297.
dendritic cells in tolerogenic cells able 70. Pohl-Koppe, A., Balashov, K.E., Steere,
to prime regulatory T cells in healthy A.C., Logigian, E.L., Hafler, D.A.
subjects. Blood 2011; 118:3549-3558. Identification of a T cell subset capable
60. Chang, Y., de, N.P., Azzaoui, I., Morales, of both IFN-gamma and IL-10 secretion
O., Delhem, N., Vorng, H., Tomavo, in patients with chronic Borrelia
S., Ait, Y.S., Zhang, G., Wallaert, B., burgdorferi infection. J. Immunol. 1998;
Chenivesse, C., Tsicopoulos, A. The 160:1804-1810.
chemokine CCL18 generates adaptive 71. Gerosa, F., Nisii, C., Righetti, S.,
regulatory T cells from memory CD4+ Micciolo, R., Marchesini, M., Cazzadori,
T cells of healthy but not allergic A., Trinchieri, G. CD4(+) T cell clones
subjects. FASEB J. 2010; 24:5063-5072. producing both interferon-gamma
61. McKinstry, K.K., Strutt, T.M., Swain, S.L. and interleukin-10 predominate in
Regulation of CD4+ T-cell contraction bronchoalveolar lavages of active
during pathogen challenge. Immunol. pulmonary tuberculosis patients. Clin.
Rev. 2010; 236:110-124. Immunol. 1999; 92:224-234.
62. Karin, M., Gallagher, E. TNFR signaling: 72. Hsu, C.Y., Leu, S.J., Chiang, B.L., Liu,
ubiquitin-conjugated TRAFfic signals H.E., Su, H.C., Lee, Y.L. Cytokine gene-
control stop-and-go for MAPK signaling modulated dendritic cells protect
complexes. Immunol Rev 2009; 228:225- against allergic airway inflammation by
240. inducing IL-10+IFN-[gamma]+CD4+
63. Faustman, D., Davis, M. TNF receptor T cells. Gene Ther 2010; 17:1011-1021.
2 pathway: drug target for autoimmune 73. Fang, R., Ismail, N., Shelite, T., Walker,
diseases. Nat Rev Drug Discov 2010; D.H. CD4+ CD25+ Foxp3GêÆ T-
9:482-493. Regulatory Cells Produce both Gamma
64. Dooms, H., Kahn, E., Knoechel, B., Interferon and Interleukin-10 during
Abbas, A.K. IL-2 induces a competitive Acute Severe Murine Spotted Fever
survival advantage in T lymphocytes. J Rickettsiosis. Infection and Immunity
Immunol 2004; 172:5973-5979. 2009; 77:3838-3849.
65. Malek, T.R., Yu, A., Scibelli, P., 74. Haringer, B., Lozza, L., Steckel,
Lichtenheld, M.G., Codias, E.K. Broad B., Geginat, J. Identification and
programming by IL-2 receptor signaling characterization of IL-10/IFN-{gamma}-
for extended growth to multiple producing effector-like T cells with
cytokines and functional maturation of regulatory function in human blood. J.
antigen-activated T cells. J Immunol Exp. Med. 2009; 206:1009-1017.
2001; 166:1675-1683. 75. Kemper, C., Chan, A.C., Green, J.M.,
66. Kim, E.Y., Teh, H.S. Critical Role of TNF Brett, K.A., Murphy, K.M., Atkinson, J.P.
Receptor Type-2 (p75) as a Costimulator Activation of human CD4+ cells with
for IL-2 Induction and T Cell Survival: CD3 and CD46 induces a T-regulatory
A Functional Link to CD28. J Immunol cell 1 phenotype. Nature 2003; 421:388-
2004; 173:4500-4509. 392.
67. Shi, M., Ye, Z., Umeshappa, K.S., Moyana, 76. Fujio, K., Okamura, T., Yamamoto, K. The
T., Xiang, J. Alpha tumor necrosis factor Family of IL-10-secreting CD4+ T cells.
contributes to CD8(+) T cell survival in Adv. Immunol. 2010; 105:99-130.
the transition phase. Biochem. Biophys. 77. Ochi, H., Abraham, M., Ishikawa, H.,
Res. Commun. 2007; 360:702-707. Frenkel, D., Yang, K., Basso, A.S., Wu, H.,
68. Jankovic, D., Kullberg, M.C., Feng, C.G., Chen, M.L., Gandhi, R., Miller, A., Maron,
Goldszmid, R.S., Collazo, C.M., Wilson, R., Weiner, H.L. Oral CD3-specific
M., Wynn, T.A., Kamanaka, M., Flavell, antibody suppresses autoimmune
CHAPTER 6

R.A., Sher, A. Conventional T-bet(+) encephalomyelitis by inducing CD4+

Foxp3(-) Th1 cells are the major source of CD25- LAP+ T cells. Nat. Med. 2006;
host-protective regulatory IL-10 during 12:627-635.
intracellular protozoan infection. J. Exp. 78. Nakayamada, S., Takahashi, H., Kanno,
Med. 2007; 204:273-283. Y., O’Shea, J.J. Helper T cell diversity
69. Anderson, C.F., Oukka, M., Kuchroo, and plasticity. Curr. Opin. Immunol.
V.J., Sacks, D. CD4(+)CD25(-)Foxp3(-) 2012.

135
 CHAPTER 6

79. Beyer, M., Schultze, J.L. Plasticity of T cells in human autoimmune disease.
T(reg) cells: is reprogramming of T(reg) Nat. Med. 2011; 17:673-675.
cells possible in the presence of FOXP3? 88. Duhen, T., Duhen, R., Lanzavecchia, A.,
Int. Immunopharmacol. 2011; 11:555- Sallusto, F., Campbell, D.J. Functionally
560. distinct subsets of human FOXP3+ Treg
80. Gabrysova, L., Nicolson, K.S., Streeter, cells that phenotypically mirror effector
H.B., Verhagen, J., Sabatos-Peyton, C.A., TH cells. Blood 2012; 119:4430-4440.
Morgan, D.J., Wraith, D.C. Negative 89. Stroopinsky, D., Avivi, I., Rowe, J.M.,
feedback control of the autoimmune Avigan, D., Katz, T. Allogeneic induced
response through antigen-induced human FOXP3(+)IFN-gamma(+) T cells
differentiation of IL-10-secreting Th1 exhibit selective suppressive capacity.
cells. J. Exp. Med. 2009; 206:1755-1767. Eur. J Immunol 2009; 39:2703-2715.
81. Saraiva, M., Christensen, J.R., Veldhoen, 90. Zhou, X., Bailey-Bucktrout, S.L.,
M., Murphy, T.L., Murphy, K.M., O’Garra, Jeker, L.T., Penaranda, C., Martinez-
A. Interleukin-10 production by Th1 Llordella, M., Ashby, M., Nakayama, M.,
cells requires interleukin-12-induced Rosenthal, W., Bluestone, J.A. Instability
STAT4 transcription factor and ERK MAP of the transcription factor Foxp3 leads to
kinase activation by high antigen dose. the generation of pathogenic memory
Immunity. 2009; 31:209-219. T cells in vivo. Nat. Immunol. 2009;
82. Awasthi, A., Carrier, Y., Peron, J.P., 10:1000-1007.
Bettelli, E., Kamanaka, M., Flavell, R.A., 91. Murai, M., Turovskaya, O., Kim, G.,
Kuchroo, V.K., Oukka, M., Weiner, H.L. Madan, R., Karp, C.L., Cheroutre, H.,
A dominant function for interleukin 27 Kronenberg, M. Interleukin 10 acts on
in generating interleukin 10-producing regulatory T cells to maintain expression
anti-inflammatory T cells. Nat. Immunol. of the transcription factor Foxp3 and
2007; 8:1380-1389. suppressive function in mice with colitis.
83. Fitzgerald, D.C., Zhang, G.X., El- Nat. Immunol. 2009; 10:1178-1184.
Behi, M., Fonseca-Kelly, Z., Li, H., 92. Moore, K.W., de Waal, M.R., Coffman,
Yu, S., Saris, C.J., Gran, B., Ciric, B., R.L., O’Garra, A. Interleukin-10 and
Rostami, A. Suppression of autoimmune the interleukin-10 receptor. Annu. Rev.
inflammation of the central nervous Immunol. 2001; 19:683-765.
system by interleukin 10 secreted by 93. Saraiva, M., O’Garra, A. The regulation
interleukin 27-stimulated T cells. Nat. of IL-10 production by immune cells. Nat
Immunol. 2007; 8:1372-1379. Rev Immunol 2010; 10:170-181.
84. Stumhofer, J.S., Silver, J.S., Laurence, 94. Unger, W.W., Laban, S., Kleijwegt,
A., Porrett, P.M., Harris, T.H., Turka, F.S., van der Slik, A.R., Roep, B.O.
L.A., Ernst, M., Saris, C.J., O’Shea, J.J., Induction of Treg by monocyte-derived
Hunter, C.A. Interleukins 27 and 6 induce DC modulated by vitamin D(3) or
STAT3-mediated T cell production of dexamethasone: Differential role for PD-
interleukin 10. Nat. Immunol. 2007; L1. Eur. J. Immunol. 2009; 39:3147-3159.
8:1363-1371. 95. van Duivenvoorde, L.M., Han, W.G.,
85. Meyaard, L., Hovenkamp, E., Otto, Bakker, A.M., Louis-Plence, P.,
S.A., Miedema, F. IL-12-induced IL- Charbonnier, L.M., Apparailly, F., van der
10 production by human T cells as a Voort, E.I., Jorgensen, C., Huizinga, T.W.,
negative feedback for IL-12-induced Toes, R.E. Immunomodulatory dendritic
immune responses. J. Immunol. 1996; cells inhibit Th1 responses and arthritis
156:2776-2782. via different mechanisms. J. Immunol.
86. Oldenhove, G., Bouladoux, N., 2007; 179:1506-1515.
Wohlfert, E.A., Hall, J.A., Chou, D., Dos, 96. Baldwin, H.M., Ito-Ihara, T., Isaacs, J.D.,
S.L., O’Brien, S., Blank, R., Lamb, E., Hilkens, C.M.U. TNF alpha blockade
Natarajan, S., Kastenmayer, R., Hunter, impairs dendritic cell survival and
CHAPTER 6

C., Grigg, M.E., Belkaid, Y. Decrease of function in rheumatoid arthritis. Ann.

Foxp3+ Treg cell number and acquisition Rheum. Dis. 2010; 69:1200-1207.
of effector cell phenotype during lethal 97. Balanescu, A., Radu, E., Nat, R., Regalia,
infection. Immunity. 2009; 31:772-786. T., Bojinca, V., Ionescu, R., Balanescu,
87. Dominguez-Villar, M., Baecher-Allan, S., Savu, C., Predeteanu, D. Early and
C.M., Hafler, D.A. Identification of T late effect of infliximab on circulating
helper type 1-like, Foxp3+ regulatory dendritic cells phenotype in rheumatoid

136
 SUMMARIZING DISCUSSION

arthritis patients. Int. J. Clin. Pharmacol. TNF alpha during maturation of dendritic
Res. 2005; 25:9-18. cells results in the development of semi-
mature cells: a potential mechanism
98. van Lieshout, A.W., Barrera, P., Smeets, for the beneficial effects of TNF alpha
R.L., Pesman, G.J., van Riel, P.L., van den blockade in rheumatoid arthritis. Ann.
Berg, W.B., Radstake, T.R. Inhibition of Rheum. Dis. 2005; 64:408-414.

CHAPTER 6

137
 &
ADDENDUM
Nederlandse samenvatting
Curriculum Vitae
Dankwoord
 NEDERLANDSE SAMENVATTING

Nederlandse samenvatting
Ons afweersysteem (immuunsysteem) beschermt ons tegen ziekteverwekkers
(pathogenen) zoals bacteriën, virussen, parasieten en schimmels. Dendritische
cellen (DC) zijn antigeen presenterende cellen die het immuunsysteem reguleren
doordat ze pathogenen kunnen herkennen. Na herkenning van een pathogeen,
wordt deze opgenomen door de DC en in kleine stukjes (antigeen) geknipt om
vervolgens aan T cellen gepresenteerd te worden. T cellen worden gezien als
uitvoerende cellen van het immuunsysteem. Er zijn twee soorten T cellen; de CD8+
cytotoxische T cellen kunnen geïnfecteerde cellen opruimen. CD4+ T cellen worden
ook wel T helper cellen genoemd, omdat ze immuunreacties kunnen ondersteunen
of activeren. De samenwerking en communicatie tussen deze cellen is van belang
voor een goede en efficiënte opruiming van pathogenen en geïnfecteerde cellen.
Communicatie tussen immuuncellen gebeurt onder andere door het uitscheiden
van signaalstoffen zoals cytokinen. Cytokinen kunnen een immuun-activerende
(pro-inflammatoire) of inhiberende (anti-inflammatoire) werking hebben op cellen
die receptoren voor deze cytokinen tot expressie brengen.
Behalve activatie van het immuunsysteem in reactie op pathogenen, is er
soms ook onderdrukking van het immuunsysteem nodig. Dit is bijvoorbeeld het
geval wanneer er geen pathogenen zijn en homeostase bewaard moet worden,
maar ook om een immuunreactie te beëindigen als een infectie effectief is
bestreden. Als de onderdrukking van het immuunsysteem niet goed verloopt, kan
er weefselschade ontstaan door excessieve immuunreacties of kunnen er auto-
immuunziekten ontstaan. DC spelen een belangrijke rol bij de onderdrukking van
het immuunsysteem. DC die immuunreacties onderdrukken worden tolerogene
dendritische cellen genoemd en geven immuunsuppressie door reactieve T cellen
te onderdrukken en door suppressieve (regulatoire) T cellen te induceren. Door
suppressieve omgevingsfactoren en de afwezigheid van pathogenen zullen DC
uitgroeien tot tolerogene DC.
Bij auto-immuunziekten en allergieën is er sprake van ongewenste en
schadelijke immuunreacties gericht tegen lichaamseigen weefsel of onschadelijke
omgevingsfactoren. Ook patiënten die een transplantatie ondergaan lopen risico
op ongewenste afstotingsreacties. Deze patiënten hebben immuunsuppressive
medicijnen nodig, die veel bijwerkingen hebben. In de toekomst kunnen
de immuunsuppressieve capaciteiten van tolerogene DC en van regulatoire
T cellen mogelijk uitkomst bieden om de ongewenste immuunreacties specifiek te
onderdrukken.

Inductie van tolerogene dendritische cellen en regulatoire

T cellen
Een van de belangrijkste immuunsuppressieve mechanismen van tolerogene DC is
de inductie van regulatoire T cellen. Omdat specifieke markers voor geïnduceerde
regulatoire T cellen niet bestaan, is een functionele assay de meest betrouwbare
manier om geïnduceerde regulatoire T cellen aan te tonen. In hoofdstuk 2 hebben

141
ADDENDUM

we een suppressie assay opgezet om de suppressieve functie van regulatoire

T cellen te kunnen meten. In deze assay wordt de remming op de celdeling van
gestimuleerde T cellen uitgelezen. We laten zien dat de hoeveelheid DC, die in
de assay gebruikt wordt voor T cel stimulatie, belangrijk is voor een optimale
uitlees van de assay. Daarnaast is een juiste negatieve controle, T cellen gekweekt
met immuun-activerende DC, van essentieel belang om specifieke suppressie te
onderscheiden van aspecifieke suppressie.
Vervolgens hebben we in hoofdstuk 3 van dit proefschrift verschillende typen
tolerogene DC bestudeerd met als doel een geschikte kandidaat te selecteren voor
een cellulaire immuunsuppressieve therapie. Tolerogene DC zijn geïnduceerd in
een klinisch-toepasbaar protocol met verschillende tolerogene stoffen; vitamine D3,
anti-inflammatoire cytokinen IL-10 en TGFβ en immuunsuppressieve medicijnen
dexamethason en rapamycine. Een additionele stimulatie in combinatie met deze
tolerogene stoffen is nodig voor een goede migratie van de DC naar de lymfeklieren,
waar de T cellen zich bevinden; en voor een stabiel fenotype en functie, een
eigenschap die nodig is om te voorkomen dat de tolerogene DC veranderen in
immuun-activerende DC. Tolerogene DC die geïnduceerd zijn met IL-10 beschikken
over de beste immuunsuppressieve kwaliteiten. IL-10 DC onderdrukken T cel
proliferatie, induceren regulatoire T cellen, en vertonen een hoge IL-10 productie.

De rol van TNFα in T cell differentiatie

In het tweede deel van dit proefschrift hebben we de rol van het pro-inflammatoire
cytokine TNFα in CD4+ T cel polarisatie en activatie bestudeerd. TNFα speelt een
rol in de initiatie en instandhouding van een immuunreactie. Daarnaast is TNFα ook
een belangrijke speler in veel auto-immuunziekten, onder andere in reumatoïde
artritis en multiple sclerosis. Patiënten met deze auto-immuunziekten kunnen
vaak succesvol worden behandeld met anti-TNFα therapieën. Het effect van
anti-TNFα op T cel activatie en polarisatie is echter nog onduidelijk. Hoofdstuk 4
laat zien dat het wegvangen van TNFα met anti-TNFα tijdens de activatie van
naïeve CD4+ T cellen door DC resulteert in een T cel populatie met meer IL-10+
T cellen. Daarnaast neemt het aantal IFNγ+ T cellen af door anti-TNFα en hebben
de T cellen een verhoogd regulatoir fenotype. Deze veranderingen tezamen
leiden tot een verhoogde suppressieve activiteit van DC-geïnduceerde regulatoire
T cellen. Daarnaast leidt neutralisatie van TNFα tot verminderde T cel activatie in
een latere fase van DC–T cel kweek. Deze effecten worden bewerkstelligd door
de afwezigheid van signalering via TNF receptor II. Deze resultaten wijzen erop
dat de immuunsuppressieve werking van anti-TNFα therapie gedeeltelijk via T cel
suppressie verloopt.

IL-10/ IFNγ dubbel positieve T cellen

De populatie IL-10 positieve T cellen geïnduceerd door anti-TNFα bestaat voor een
groot deel uit IL-10/ IFNγ dubbel positieve cellen. IFNγ is een pro-inflammatoir
cytokine voornamelijk betrokken bij de immuunreactie tegen virussen en
intracellulaire bacteriën. Terwijl IL-10 als anti-inflammatoir cytokine betrokken is bij

142
 NEDERLANDSE SAMENVATTING

de onderdrukking van immuunreacties en onder andere wordt geproduceerd door

tolerogene DC en regulatoire T cellen. Naast regulatoire T cellen is beschreven dat
ook pro-inflammatoire T cellen IL-10 kunnen produceren. In dit proefschrift hebben
we in hoofdstuk 5 het fenotype en de functie van deze IL-10+/ IFNγ+ T cellen verder
bestudeerd. IL-10/ IFNγ dubbel positieve T cellen hebben een regulatoir genotype
ten opzichte van IFNγ enkel positieve T cellen. Daarnaast hebben de IL-10+/ IFNγ+
T cellen ook een sterk regulatoir fenotype en suppressieve activiteit. Deze resultaten
laten zien dat de IL-10/ IFNγ dubbel positieve T cellen een suppressieve T cel
populatie is, die mogelijk betrokken is bij de bescherming tegen weefselschade
door immuunreacties te onderdrukken.

Samenvattend laat dit proefschrift zien dat zowel IL-10 tolerogene DC als
anti-TNFα T cel tolerantie induceren door enerzijds regulatoire T cellen te
induceren en anderzijds de deling en cytokinen productie van pro-inflammatoire
CD4+ T cellen te onderdrukken. De combinatie van tolerogene DC en anti-TNFα
heeft een synergistische tolerogene werking en zou mogelijk tot een verbeterde
immuunsuppressieve therapie kunnen leiden voor patiënten met TNFα-gemedieerde
(auto-)immuunziekten.

143
 CURRICULUM VITAE

Curriculum Vitae
De auteur van dit proefschrift, Martine Boks, werd geboren op 6 juli 1983 te Voorburg.
In 2001 behaalde zij haar VWO Gymnasiumdiploma aan het Ichthus College in
Kampen om vervolgens te beginnen aan de studie Biomedische Wetenschappen
aan de Vrije Universiteit te Amsterdam. Zij behaalde haar bachelor diploma in 2004
en heeft in deze tijd een korte stage gelopen bij de afdeling Ontwikkelingsbiologie
aan de VU, onder begeleiding van dr. A. Rebocho en Prof. dr. R. Koes. Hier deed
zij onderzoek naar het EVG gen in Petunia hybrida. Gedurende de masteropleiding
liep zij stage bij de afdeling Moleculaire Celbiologie en Immunologie van
het VU medisch centrum (VUmc), onder begeleiding van dr. S. van Vliet en
Prof. dr. Y. van Kooyk. Hier bestudeerde zij de internalisatie van de C-type lectine
receptor MGL op dendritische cellen. Tijdens een tweede stage op de afdeling
Moleculaire Biologie van het Nederlands Kanker Instituut, onder begeleiding van
dr. R. Wolthuis en Prof. dr. H. te Riele, onderzocht zij de rol van eiwitafbraak tijdens
mitose. In 2007 voltooide zij de masteropleiding Biomedische Wetenschappen,
waarna zij haar promotieonderzoek startte bij de afdeling Immunopathologie van
Sanquin Bloedvoorziening te Amsterdam, onder begeleiding van dr. J.A. ten Brinke,
Prof. dr. S.M. van Ham en dr. J.J. Zwaginga. De resultaten van dit onderzoek staan
beschreven in dit proefschrift. Vanaf juni 2012 is zij werkzaam als post-doc op de
afdeling Moleculaire Celbiologie en Immunologie van het VUmc in de groep van
Prof. dr. Y. van Kooyk.

145
 DANKWOORD

Dankwoord
En dan, na 5 jaar onderzoek doen, ligt er eindelijk een boekje. Dit proefschrift
was er nooit geweest zonder de steun van alle mensen om mij heen, zowel
wetenschappelijk als niet-wetenschappelijk. Bedankt daarvoor! Een aantal mensen
wil ik graag in het bijzonder bedanken:

Allereerst mijn promotor en co-promotores. Anja, je bent een fantastische mentor

voor me geweest. Jouw enthousiasme voor onderzoek werkt enorm aanstekelijk. Ik
kwam natuurlijk als ‘groentje’ binnen, maar ik heb in de afgelopen jaren ontzettend
veel van je geleerd. Dank je wel daarvoor.
Marieke, bedankt voor de kans die je me hebt gegeven om bij jou te mogen
promoveren. Ik heb een fantastische tijd gehad op het lab. Jouw enorme positivisme
heeft me er regelmatig doorheen gesleept. Een praatje met jou was altijd voldoende
om het weer positief in te zien en er weer voor te gaan, of om juist even wat beter
op mezelf te letten toen ik dat nodig had. Dank je wel hiervoor.
Jaap Jan, ook jij bedankt voor het begeleiden van mijn AIO traject. Jouw
klinische kijk op het onderzoek werkte vaak verfrissend. Bedankt voor je inzet, je
ideeën en de nuttige discussies die we tijdens de DC werkbesprekingen hadden.
En dan natuurlijk mijn paranimfen. Judith, zonder jou had ik het waarschijnlijk
nooit gered. Ik ben nog steeds ontzettend blij dat jij halverwege op mijn
project bent gekomen. Mijn dank voor jouw onuitputtelijke enthousiasme en
doorzettingsvermogen. Het DC project is geen makkelijk project en de proeven
waren lang niet altijd even leuk, maar dankzij jouw power en doortastendheid is
mijn project tot een succesvol einde gekomen, dank je wel!
Laura, mijn andere paranimf, ik ben zo blij dat jij bij ons op de afdeling bent
gekomen. Sindsdien heb ik er een goede vriendin bij! Het klikte meteen al goed
en al vrij snel waren we een goed team voor het initiëren van borrels en andere
gezelligheden. En dat gaan we nu buiten het lab zeker vasthouden!
De DC-groep: Gijs, Miranda, Sonja en Suzanne, en ook oud-DC-groep-leden:
Mirjam, Remco en Maria, bedankt voor al jullie hulp en de leuke samenwerking! Sonja,
ik vond het erg leuk om er een AIO-zusje bij te krijgen! Succes met je laatste jaar.
De studenten die op mijn project hebben gezeten: Michiel, Kiran en Sietske,
bedankt voor jullie hulp. Ik vond het ontzettend leuk om jullie te mogen begeleiden.
Michiel, bij jou is er zelfs een publicatie van gekomen!
Simone, Pauline en Laura, mede dankzij jullie heb ik een geweldige AIO tijd
gehad. Altijd gezelligheid met borrelen, terrasjes pakken, stappen, concerten,
festivals, etc. En sinds kort ook weekendjes weg. Kopenhagen was geweldig (en
koud…)! Het nieuwe weekendje staat ook alweer gepland, zin in! Ik zie een traditie
aankomen, die we ondanks buitenlandplannen gewoon doorzetten toch?!
Jelle, Femke en Theresa, samen met mijzelf de ‘oude garde’. Jelle, jij hebt het er
al op zitten, voor jou komt het grote buitenlandavontuur nu snel dichterbij! Femke,
jij moet nog even, maar ook voor jou komt er eerst een nieuw avontuur aan! En dat

147
ADDENDUM

promoveren komt daarna helemaal goed. Theresa, jij bent ook bijna zover, zet hem
op met de laatste loodjes!
De rest van de AIO kamer: Gerben, Iwan, Lotte, Mateusz, Mieke P en Mischa.
Ik denk dat wij de mooiste AIO kamer van Nederland hebben! Ik vertrouw er op
dat jullie de traditie van lelijke souvenirs doorzetten, de kerstversiering gewoon
lekker tot de volgende kerst laten hangen (dat is tenslotte alweer over een paar
maandjes) en dat er nog meer prullaria bij komt. Regelmatige updates worden op
prijs gesteld! Met z’n allen beneden was af en toe veel te gezellig. Maar er werd
ook nog wel eens gewerkt. Mijn boekje is in ieder geval af gekomen. Bedankt voor
de gezelligheid, de steun, de gevulde snoep- en koektrommels (hoeveel hadden
we er op een gegeven moment, een stuk of vijf?) en de fantastisch lelijke souvenirs!
En dan natuurlijk de rest van de afdeling Immunopathologie: Angela, Anneke,
Dorina, Dörte, Ellen, Els, Flavia, Gerard, Gert Jan, Gwen, Hanny, Ineke, Irma, Jillian,
Jolanda, Josine, Lucien, Margreet, Marja, Mieke B, Ninotska, Piet, Pleuni, Rishi,
Rob, Ruchira, Sacha, Shabnam, Steven, Susanne, Tamara, Theo, Tineke, Wouter en
Yuri. Bedankt voor alle hulp en gezelligheid op (en buiten) het lab. Dankzij jullie heb
ik een hele fijne AIO tijd gehad! Fatima en Kaoutar, bedankt voor alle goede zorgen
omtrent de congresreisjes en vooral ook tijdens het laatste gedeelte van mijn
promotie. Bouke, bedankt voor het op peil houden van alle voorraden. Door jou
had er bijna geen boekje gelegen en was ik een modellencarrière begonnen! Diana,
ik vond het superleuk om samen met jou het eerste (sinds lange tijd) labweekend
te organiseren! Die traditie is weer nieuw leven ingeblazen. Henk, Jelle, Ingrid en
Diana, we hebben een paar mooie borrels georganiseerd. Van oranje, tot cocktails,
en zelfs een keer vrijdag de 13e, altijd gezelligheid!
Erik en Floris, ik dacht heel lang dat ik er niet aan zou beginnen, maar uiteindelijk
zaten Judith en ik toch vrij vaak bij jullie voor het sorten. Bedankt voor jullie geduld
en de goede zorgen met sorten en FACSen.
Ook wil ik graag Sjaak, Petra en Arno van het NKI bedanken voor alle hulp bij
het uitvoeren van de micro-arrays. Petra, thank you for your time to run the samples
with us. En Arno, bedankt voor je hulp bij het analyseren van de data.

Ook buiten het werk om zijn er veel mensen belangrijk geweest voor het goede
einde van mijn promotie.
Lin, Suus en Renaat; lieverds, bedankt voor jullie vriendschap en bedankt voor
jullie steun in mindere tijden. Lin, op een gegeven moment op ‘iets’ meer afstand,
maar daarom niet minder! Het zit in de planning om volgend jaar weer bij jullie langs
te komen. Suus en Renaat, jullie gelukkig nog wel lekker dichtbij in Amsterdam. Nu
geen gezeur meer van mij over promoveren, en weer alle tijd voor zomibo’s en
BBQen!
Marijke, mijn ‘oudste’ vriendinnetje. ;-) Nooit gedacht dat dat brieven schrijven
tot een levenslange vriendschap zou leiden. Ook al zien we elkaar niet heel vaak,
de vriendschap is er nog steeds en daar ben ik heel erg blij om. Nu mijn boekje af
is, kom ik weer vaker langs! Friends forever.

148
 DANKWOORD

Bernadette, ook wij zijn al heel lang vriendinnen. Ik weet nog dat ik onze ‘scheiding’
door verschillende middelbare scholen verschrikkelijk vond. Maar dat hebben we
overleefd! Later allebei in (de buurt van) Amsterdam gaan wonen en studeren, dat
was wel zo makkelijk om elkaar regelmatig te zien. En dat houden we er in!
En verder alle anderen die van dichtbij of ver weg mijn promotiesores hebben
meegemaakt en mij hebben gesteund, opgevrolijkt, een (salsa)dansje hebben
gedaan of gewoon een wijntje hebben gegeven ;-), dank jullie wel!

Tot slot, lieve papa, mama en Jasper, bedankt voor jullie onvoorwaardelijke
steun en liefde. Mijn AIO tijd is niet altijd even makkelijk geweest, maar ik heb altijd
op jullie kunnen bouwen. Zonder jullie geloof en vertrouwen in mij had ik dit nooit
gekund. Ik hou van jullie.

149