at the maternal-fetal interface throughout human pregnancy

The handle http://hdl.handle.net/1887/84689 holds various files of this Leiden University dissertation.

Author: Zwan, A. van der

Title: The immune compartment at the maternal-fetal interface throughout human pregnancy

Issue Date: 2020-02-06

at the maternal-fetal interface throughout human pregnancy

Anita van der Zwan

All rights reserved. No part of this thesis may be reproduced or transmitted in any form, by any means, electronic or mechanical without prior permission of the author, or where appropriate, of the publisher of the articles.

The research presented in this thesis was performed at:

The Department of Stem Cell and Regenerative Biology at Harvard University and the Department of Immunohematology and Blood Transfusion at Leiden University Medical Center.

Financial support for the research conducted at Harvard University was kindly provided by: VSBfonds and Studiefonds Ketel1.

Financial support for the publication of this thesis was kindly provided by:

Stichting Oranjekliniek, National Reference Center for Histocompatibility Testing, GenDx, Fluidigm, CleanAir by Baker, U-CyTech Biosciences, ChipSoft, ABN AMRO, and Pfizer.

Cover design and layout by: Nicolene van der Zwan Printed by: GVO drukkers & vormgevers B.V.

ISBN: 978-94-6332-603-2

at the maternal-fetal interface throughout human pregnancy

Anita van der Zwan Proefschrift

ter verkrijging van de graad van Doctor aan de Universiteit Leiden,

op gezag van Rector Magnificus prof. mr. C.J.J.M. Stolker, volgens besluit van het College voor Promoties

te verdedigen op donderdag 6 februari 2020, klokke 11.15 uur

door

geboren te Pretoria, Zuid Afrika

in 1988

Co-promotoren: Dr. S. Heidt

Dr. T. Tilburgs (Cincinnati Children’s Hospital, USA)

Leden promotiecommissie: Prof. Dr. J. Borst Prof. Dr. C. van Kooten

Prof. Dr. A. Moffett (University of Cambridge, UK)

Prof. Dr. S. Saito (University of Toyama, Japan)

The answers we have found have only served to raise a whole set of new questions.

In some ways we feel that we are as confused as ever, but we think we are confused on a higher level, and about more important things.”

Earl C. Kelley

The Workshop Way of Learning (1951)

Mass cytometry quality control: a crucial step not to be neglected

General introduction 9

61

91

117

133

145

183

Three types of functional regulatory T cells control 29

T cell responses at the human maternal-fetal interface

Mixed signature of activation and dysfunction allows human decidual CD8

T cells to provide both tolerance and immunity

Cross-reactivity of virus-specific CD8

T cells against allogeneic HLA-C: possible implications for pregnancy outcome

Cytotoxic potential of decidual NK cells and CD8

T cells awakened by infections

Visualizing dynamic changes at the maternal-fetal interface throughout human pregnancy by mass cytometry

Summarising discussion

01

03 04 05 06 07 08 02

09 English summary

Nederlandse samenvatting List of publications About the author

203

General introduction

Chapter 01

01

General introduction

1. The human immune system

The human immune system is shaped by cell populations, known as white blood cells or leukocytes, that are dedicated to protecting their host against tumor cells and micro- organisms, such as viruses, bacteria and parasites. Cells from the immune system originate from hematopoietic stem cells in the bone marrow, which after development circulate in the blood or undergo further education in the thymus, followed by migration to other (lymphoid) tissues. Traditionally, the immune system is divided into the innate and adaptive immune compartment, though several cell types (i.e. dendritic cells, TCRγγ γδ T cells and NKT cells) form a link between these two arms of the immune system (1, 2). To function properly, immune cells must distinguish between ‘self’ (one’s own healthy cells) and ‘non-self’ antigens such as tumor cells and pathogens. Immune tolerance towards self-antigens is established early in life where most cells reactive against self-antigens (autoreactive cells) are deleted to prevent autoimmunity. Upon breaching physical barriers such as the skin and mucosal surfaces, the first line of defense a pathogen encounters is the innate immune response that relies on several recognition mechanisms. This response involves soluble proteins (cytokines, chemokines), the complement system, and cells of both the myeloid lineage, such as macrophages, dendritic cells (DC), neutrophils and eosinophils, and the lymphoid lineage, including natural killer (NK) cells and innate lymphoid cells (ILCs). The innate immune system mounts a rapid and non-specific immune response against pathogens.

Infections that manage to surpass or outnumber innate immunity, encounter cells of the adaptive immune system that are more specific and consist of B and T lymphocytes, the latter including CD8+ and CD4+ T cells. B and T lymphocytes express somatically rearranged receptors that have specificity for the antigens they bind. The adaptive immune arm provides long-term immunological memory that allows stronger and faster immune responses upon re-encounter with the same pathogen. This immunological memory forms the basis for effective vaccination.

Crosstalk between the innate and adaptive immune compartment mediates an effective immune response (3) (Fig. 1). Processed peptides of an antigen are presented by antigen-presenting cells (APC), such as macrophages, dendritic cells and B cells, in their major histocompatibility complex (MHC) molecules to T cells. B cells recognize and internalize intact pathogens with their B cell receptors (BCR), after which they produce immuno-globulins (antibodies) and present antigen to CD4+ T cells to activate them.

Subsequently, T cells, that can be subdivided into effector helper T cells, cytotoxic T cells

and regulatory T cells, recognize these processed peptides bound to MHC molecules

with their T cell receptor (TCR). CD4+ T cells recognize peptides in the context of MHC

class II, hereby providing help to cytotoxic T cells and B cells, while CD8+ T cells can

recognize peptide-MHC class I resulting in an effector response that includes the generation

of long-live memory cells (4) (Fig. 2A).

2. HLA and alloreactive-immune responses

Whether a response is elicited by the immune compartment depends on the recognition of ‘self’ and ‘non-self’. MHC genes are essential for this recognition and encode for MHC class I and II molecules. MHC class I molecules are expressed by all nucleated cells and platelets, and present peptides derived from intracellular pathogens. MHC class II molecules are expressed by APC, activated T cells, and activated endothelial cells, and present self and non-self peptides that appear in the extracellular milieu. In humans, MHC molecules are known as human leukocyte antigens (HLA) that are located on the short arm of chromosome 6. HLA class I consists of the polymorphic HLA-A, -B, and -C and the non-polymorphic HLA-E, -F, and -G molecules. HLA class II is divided into the three major HLA-DR, -DQ, and -DP antigens (5). Every individual inherits one allele from their mother and one allele from their father, making a fetus semi-allogeneic to the immune system of the mother. There are numerous genetic variants, polymorphisms, of HLA molecules in human populations. These polymorphisms ensure that a population does not succumb to new pathogens or mutated pathogens, because several individuals in the population will be able to mount an immune response to these pathogens. Essentially, the function of HLA molecules is to present peptide fragments derived from pathogens to the TCR expressed on T cells. Here, T cells are activated by non-self antigens presented in self

Figure 1. A simplistic overview of the main players in an immune response.

During an immune response, pathogens (viruses, bacteria, parasites) can be taken up by antigen-presenting cells (e.g. dendritic cells and macrophages) and presented to specific T cells. CD8+ T cells become activated cytotoxic T cells (CTL) that secrete pro-inflammatory cytokines and cytotoxic molecules, such as perforin and granzymes.

CD4+ T cells differentiate into several T helper (Th) cell subsets (e.g. Th1, Th2, Th17, and Treg cells) depending on the presence of certain cytokines. B cells can activate CD4+ T cells that subsequently provide help to B cells.

Antigen-presenting cells also play an important role in controlling ILC homeostasis and function.

Macrophage Dendritic cell

B cell CD4+ T cell CD8+ T cell

Innate lymphoid cell

Treg Virus Bacteria Parasite

CD8+ T cell (CTL)

Granzymes Perforin Cytokines

Th1 cell Th2 cell Th17 cell

01

HLA. Conversely, T cells can also get activated by the presence and recognition of non- self (allogeneic) HLA molecules as is the case in transplantation, blood transfusions and pregnancy. When a donor and recipient have different HLA molecules, the recipient’s immune system mounts a response against the HLA molecules of the donor as the graft expressing these donor molecules is perceived as ‘foreign’. This response against the donor is referred to as an allo-immune response. Alloreactivity of T cells comes in two flavors: direct and indirect allorecognition. In direct allorecognition, recipient T cells recognize allogeneic (donor) peptide/MHC complexes (pMHC) presented on the surface of donor cells. Indirect allorecognition involves the response of recipient T cells to processed donor MHC allo-peptides presented in the context of self MHC molecules expressed on recipient APC (6). It was this phenomenon of alloreactivity in transfusion and pregnancy that led to the initial discovery of MHC (7, 8). Immune cell allo-reactivity can also occur in an antigen-independent manner, as is the case with NK cells. NK cell inhibitory killer cell immunoglobulin-like receptors (KIR) and C-type lectins can recognize specific MHC class I molecules. However, cells that fail to express self MHC-I molecules on their cell surface are eliminated by NK cells via activating NK cell receptors, known as the missing-self hypothesis (9). Major ligands for inhibitory KIRs are HLA-C molecules, that are classified as HLA-C1 and HLA-C2 molecules based on their sequence. In a transplantation setting, when the recipient’s NK cells fail to recognize self-KIR ligands (HLA-C molecules) on the donor cells, they kill the donor cells as a result of the lack of inhibitory signals and the engagement of activating receptors (10).

3. T cell immunity

3.1 T cell development and activation

T cells are generated in the bone marrow, after which they migrate to the thymus to mature and undergo gene rearrangements followed by negative and positive selection. In the thymus, T cell precursors express a TCR that is generated by the recombination of gene segments (majority of γγ αβ T cells; minority of γγγδ T cells) and differentiate into CD4+ and CD8+ T cells when they have affinity for, and bind peptides presented in MHC class I or II.

This process is referred to as positive selection and only involves γγ αβ T cells as γγγδ T cells

seem indifferent to peptides presented by MHC molecules. T cells that bind peptide-MHC

complexes and are chosen by positive selection are subsequently subjected to negative

selection where T cells with receptors that bind too strong to self-MHC undergo cell death

(autoreactive T cells). These processes ensure that the T cells that leave the thymus

(1-2%) are tolerant to self-antigens and capable of initiating an immune response against

foreign antigens presented by self-MHC molecules. This process is imperfect, and a small

fraction of autoreactive T cells escapes negative selection and needs to be regulated in

the periphery. For this purpose, a subset of autoreactive CD4+ T cells that are not

deleted differentiate into regulatory T cells (Treg) characterized by the expression of

CD25 and FOXP3, and are equipped to maintain peripheral immune tolerance (11, 12).

These Treg are indicated as thymus-derived naturally occurring Treg (nTreg).

Mature T cells that leave the thymus are considered naïve (N) cells and recirculate between the blood and secondary lymphoid organs. Encounter with a foreign antigen triggers proliferation, clonal expansion and differentiation of these naïve T cells into effector (EFF) T cells that secrete pro-inflammatory cytokines, such as IL-2 and IFN-γ γ, and cytotoxic molecules, such as perforin and granzymes (primary immune reaction). This T cell activation requires three signals: 1) strong and specific binding of the TCR to its pMHC accompanied by the binding of the co-receptors CD4 or CD8 for additional stabilization, 2) binding of the co-stimulatory molecule CD28 on the T cell surface to CD80/CD86 on the APC surface, and 3) an additional signal provided by inflammatory cytokines to determine which type of responder the cell will become (13). CD8+ T cells become activated cytotoxic T cells, whereas CD4+ T cells differentiate into several T helper (Th) subsets, namely Th1, Th2, Th9, Th17, Tfh (follicular helper) and Treg cells depending on the presence of certain cytokines (14) (Fig. 1). Next, these effector T cells exit the lymphoid tissue to scout out the site of infection. Once the infection is cleared, the majority (90- 95%) of effector cells undergo apoptosis (cell death) with a small percentage differentiating into antigen-specific central and effector memory (CM, EM) T cells. These memory T cells have the capacity to immediately respond upon re-encounter with the same pathogen leading to a considerably shorter response time of the secondary immune reaction (15).

CM T cells express homing receptors that allow them to migrate to secondary lymphoid organs and exhibit higher proliferative capacity but slower effector function than EM T cells. EM T cells express a different set of chemokine receptors that allows them access to inflamed peripheral tissues. Apart from differentiating into fully functional T cells, CD8+

T cells can also become dysfunctional during an immune response, as is described below.

3.2 Regulatory T cells (Treg)

In an ideal immunological response, Treg play their part in making sure the effector response is regulated and does not cause any damage to the host thereby preventing autoimmune diseases and limiting chronic inflammatory diseases. In addition to the previously mentioned nTreg, Treg can also extrathymically be ‘induced’ or ‘converted’

from naïve conventional CD4+ T cells into regulatory T cells (iTreg) during inflammatory conditions in peripheral tissues (16). Induced Treg have generally been described to lack expression of FOXP3 (a marker that specifically identifies nTreg) and HELIOS, although controversy remains about whether lack of HELIOS is a defining marker for iTreg (17).

In contrast to nTreg, iTreg frequently have a restricted TCR specificity for distinct cell

types or foreign antigens. Recent studies have revealed heterogeneity within the Treg

compartment, where many different Treg subsets exist. These subsets may rely on multiple

regulatory mechanisms such as: suppression by inhibitory cytokines (IL-10, TGF-γ β),

suppression by cytolysis (granzymes), suppression by metabolic disruption and

suppression by modulating DC maturation and thereby their function to activate effector

01

T cells. A division of labor between nTreg and iTreg may be conceivable to regulate immune responses (16, 18).

3.3 Heterologous immunity

To ensure a comprehensive immune response to foreign antigens, cross-reactivity has been described as an intrinsic and essential feature of antigen recognition by T cells, where one TCR can react with approximately 10

different MHC-associated processed peptides and thereby provide sufficient protection against different pathogens (19). Thus, CD8+

T cells confer cross-reactivity against all influenza virus strains and cross-protection has been observed for virus-specific CD8+ T cells that recognize both the hepatitis C virus and the Influenza A virus, or both the Epstein-Barr virus (EBV) and Influenza A virus, and recently protection by Dengue virus-specific CD8+ T cells against Zika virus during pregnancy in a murine infection model has emerged (20, 21, 22, 23). Aside from cross- reactivity against other viruses, a significant proportion of virus-specific CD8+ T cells in healthy (non-HLA sensitized) individuals display cross-reactivity against allogeneic HLA molecules (allo-HLA) (24, 25). This phenomenon that enables the same TCR to recognize the pMHC that was involved in its initial priming and cross-react with a different pMHC is referred to as heterologous immunity (Fig. 2B). Both naïve and memory T cells show alloreactive potential, though memory T cells pose a superior threat given that their activation threshold is significantly lower as they have less need for co-stimulation (26).

Heterologous immunity may have implications in transplantation where allo-reactive T cells are a cause of graft-versus-host disease and graft rejection (27). Furthermore, potential cross-reactivity of maternal virus-specific T cells against fetal allo-HLA may have detrimental consequences for the success of pregnancy, a subject that is discussed in more detail in Chapter 4 of this thesis.

Figure 2. MHC-TCR interactions and heterologous immunity.

(A) T cells recognize processed peptides bound to major histocompatibility complex (MHC) molecules with their T cell receptor (TCR). CD4+ T cells recognize peptides in the context of MHC class II, while CD8+ T cells recognize peptides in the context of MHC class I. (B) Heterologous immunity is the phenomenon where virus-specific CD8+

T cells can recognize the peptide-MHC complex that was involved in their initial priming (viral peptide) as well as cross-react with an allo-peptide-MHC complex presenting self-peptides.

CD4+ T cell

IFNy IFNy

CD8+ T cell

Antigen-presenting cell

TCR TCR

MHC-II MHC-I

Cross-reactive virus-specfic CD8+ T cell Virus-infected

autologous cell Allogeneic cell

viral peptide self peptide

A B

3.4 Immune checkpoint receptors to preserve tolerance

The extent of an allo-immune response by T cells is not only dependent on the TCR and pMHC interaction, but also on the expression of immune checkpoint receptors, specifically co-stimulatory and co-inhibitory pathways. These receptors play a key role in regulating immune responses in infections, especially for latent pathogens, and pose opportunities for manipulation of immune responses in cancer. The expression of co-inhibitory receptors such as programmed cell death-1 (PD-1), T cell Ig mucin-3 (TIM-3), cytotoxic

T-lymphocyte–associated protein 4 (CTLA-4) and T-cell immunoreceptor with Ig and ITIM domains (TIGIT) by T cells is essential to maintain self-tolerance, but in the setting of cancer (e.g. colorectal cancer and melanoma) and chronic infections has been associated with T cell dysfunction (28, 29, 30). Dysfunctional T cells include: 1) exhausted T cells that lose their effector functions during chronic antigen exposure, 2) anergic T cells that are primed without co-stimulation and thus fail to gain effector functions, and 3) T cells that are temporarily suppressed by suppressor cells such as Treg (31, 32). When T cells lose their effector functions, they downregulate production of cytokines such as IL-2, IFNγ- γ and TNF- αγ and reveal reduced cytotoxicity and proliferation. Blocking antibodies targeting PD-1 and CTLA-4 have led to improved T cell effector function and hereby regression in cancer patients (33). In contrast, in the transplantation setting the use of CTLA-4 antibodies that bind their targets CD80/CD86 blocked positive co-stimulatory signals through CD28, initially resulting in prolonged allograft survival in both murine and non-human primate models. However, in clinical trials acute rejection episodes were observed with one of the underlying mechanisms most likely being the simultaneous blocking of CTLA-4 on Treg (34). Important to note is that these co-inhibitory receptors are not exclusively expressed by dysfunctional T cells but are also observed in activated T cells. Inhibitory receptors counterbalance the activation signals provided by stimulatory receptors such as CD28, CD69 and Inducible T-cell Costimulator (ICOS). Excessive co-stimulation and/or inadequate co-inhibition can result in diminished self-tolerance and subsequently auto- immunity (35). Therefore, a sufficient balance between co-stimulation and co-inhibition is necessary to guide an effective immune response.

4. Immunological paradox of pregnancy

The work of Billingham, Brent and Medawar on immunological tolerance and graft

rejection has proven fundamental in the field of transplantation (36, 37). Moreover,

their work played a major role in the field of reproductive immunology as Sir Peter

Medawar first described the ‘immunological paradox of pregnancy’ in 1953: ‘The

immunological problem of pregnancy may be formulated thus: how does the pregnant

mother contrive to nourish within itself, for many weeks or months, a fetus that is an

antigenically foreign body?’. Pregnancy resembles organ transplantation in that the fetus

expresses inherited paternal antigens and is hereby considered a semi-allogeneic graft that

is immunologically foreign to the mother. After solid organ transplantation, given the HLA

01

mismatches between donor and recipient, recipients rely on life-long immunosuppressive drugs to prevent graft rejection. The success of pregnancy, without immunosuppressive treatment, could be explained by three hypotheses according to Medawar: 1) anatomical separation of mother and child, 2) antigenic immaturity of the fetus and 3) a non-active maternal immune system (38). Although the exact mechanisms of immune tolerance that are at play in the placenta have not fully been unraveled, it is clear that all three of Medawar’s hypotheses are essentially incorrect. Evidence contradicting his ideas are discussed below.

5. The maternal-fetal interface

After fertilization, the human blastocyst adheres to the wall of the uterus and fetal cells (trophoblasts) of the blastocyst attach to the receptive endometrial epithelium. Successful implantation is vital for the establishment and completion of a healthy pregnancy (39). The inner cell mass of the blastocyst develops into the embryo and the outer cell mass will form the placenta. The placenta is a key player in pregnancy as it connects the developing fetus via the umbilical cord to the wall of the uterus to allow for the exchange of nutrients, oxygen, waste products and thermo-regulation via the mother’s blood flow. Subsequently, the trophoblasts deeply invade the uterine decidualized epithelium (hereafter referred to as

‘decidua’) and spiral arteries where they come into direct contact with maternal immune cells, the so-called maternal-fetal interface (Fig. 3). Three maternal-fetal interfaces where maternal and fetal cells intermingle can be identified: 1) the decidua basalis that is located at the implantation site, 2) the decidua parietalis that is attached to the fetal membranes (chorion and amnion) that line the uterine cavity and surround the fetus, and 3) the branched villi network that baths in maternal peripheral blood mononuclear cells (mPBMC) at the intervillous space. The existence of these maternal-fetal interfaces, together with evidence of prolonged persistence of fetal progenitor cells and cell-free DNA in maternal blood and persistence of maternal cells in offspring well into adult life (microchimerism), refutes Medawar’s hypothesis of an anatomical separation between mother and child (40, 41).

6. Immunity in pregnancy

Broadly, trophoblasts can be divided into villous syncytiotrophoblasts and extravillous trophoblasts (EVT) that differentiate out from trophoblast stem cells. Remarkably, the villous syncytiotrophoblasts are devoid of any MHC expression and do not express NK activating receptors, thereby preventing their recognition by the maternal immune system. EVT on the other hand express polymorphic HLA-C and non-polymorphic HLA-E, -F and -G, but lack expression of the classical HLA-A and -B antigens and HLA class II molecules (42).

During pregnancy, EVT migrate into the maternal decidua and contribute to spiral artery remodelling and vascularization to increase blood flow to the fetus. By not expressing the classical HLA-A and -B antigens, EVT avoid direct allo-antigen recognition by decidual T cells.

Nevertheless, maternal NK cells express KIR that recognize fetal HLA-C expressed on

EVT and expression of allogeneic HLA-C was correlated with the activation of maternal

T cells and induction of Treg (43, 44). Furthermore, HLA-G+ EVT can interact with NK cells, APC and T cells to induce immune tolerance (45, 46) and they express and secrete immunomodulatory molecules such as galectin-1, Fas-ligand, IDO and TRAIL (47, 48, 49, 50). Thus, although fetal cells reveal altered MHC expression, they are not antigenically immature and experience crosstalk with maternal immune cells.

Since Medawar postulated the ‘immunological paradox of pregnancy’, a lot of research has focused on resolving the question of whether decidual immune cells at the maternal-fetal interface reside in an inert state (51). While tolerance towards the semi-allogeneic fetus needs to be preserved, maternal immune cells must simultaneously provide protection against environmental pathogens. Circulating immune cells undergo modifications during pregnancy (52) and likewise within the decidua local maternal immune adaptations need to be strictly regulated. ILCs and primarily NK cells are the most prevalent cell type (70%) in early pregnancy and decrease in time, while T cells make up 5-15% of immune cells in early pregnancy and their proportions increase to 40-70% over the course of gestation. APC remain relatively constant throughout pregnancy and B cells have been described as a sparse population (53). (Fig. 4)

Figure 3. The maternal-fetal interface.

Left panel; the uterus, placenta and fetus. The placenta consists of the chorionic plate, villi and the decidua basalis

which is adjacent to the maternal myometrium. The fetal membranes consist of the amnion, chorion and decidua

parietalis. The fetus is connected to the placenta via the umbilical cord. Right panel a; at the maternal-fetal interface

there is direct contact between the fetal villous tissue and maternal decidual tissue. Villous cytotrophoblasts grow out

to form cell columns (CT) that attach to the decidua. From these columns, extravillous trophoblasts (EVT) enter the

decidua as interstitial trophoblasts (T) and then endovascular trophoblast cells (E) move down the uterine spiral arteries

(A) to replace the endothelial cells. Right panel b; Interstitial trophoblasts (T) are in direct contact with maternal stromal

cells (S), natural killer cells (K), macrophages (M) and T cells (L). Adapted from A. Moffett. Nat. Rev. Immunol. 2002.

01

Decidual NK cells (dNK) differ from peripheral blood NK cells in that they express high levels of CD56, lack the marker CD16, contain equally high levels of the cytolytic molecules perforin and granzyme B, and increased levels of granulysin, but are not able to kill target cells as efficiently as peripheral blood NK cells (54). Rejection of the semi-allogeneic fetus by dNK is prevented by the expression of HLA-C on EVT (missing self-hypothesis). dNK cooperate with EVT, by KIR and HLA-C and/or HLA-G interactions, to remodel spiral arteries, assist in decidualization and may play a role in fighting of placental infections (55). dNK revealed cytotoxic function against decidual stromal cells (DSC) infected with human cytomegalovirus (HCMV), but were unable to degranulate or secrete cytokines against HCMV-infected EVT (56). This emphasizes that EVT maintain protective mechanisms against an attack from maternal immune cells, even during an infection, resulting in an immunological challenge to clear placental viral infections during pregnancy.

Macrophages are the most abundant APC at the maternal-fetal interface. Tissue-resident macrophage populations are replenished by bone marrow-derived blood monocytes.

They have a dual function in protecting the host against infections and maintaining tissue homeostasis thereby also promoting tissue repair (57). Decidual macrophages exhibit regulatory and suppressive properties by the production of immunomodulatory molecules such as IDO and IL-10. Moreover, they maintain homeostasis by clearance of cell debris and apoptotic cells, undertake crosstalk with T cells and play a role in spiral artery remodeling and angiogenesis by interacting with dNK and EVT (58, 59, 60, 61).

Dendritic cells (DC) form a crucial link between the innate sensing of invading pathogens, the presentation of antigens and thereby effectively activating the adaptive immune system. The numbers of DC in the placental bed are significantly lower when compared to the number of macrophages. They appear in a resting state, most likely to maintain immune tolerance (62). Decidual DC are unable to mature in response to bacterial ligands (63), and may be involved in promoting a Th2 T cell phenotype and the induction of Treg (64, 65). Furthermore, close contact between decidual DC and dNK has been observed in early human pregnancy (66). Additionally, granulocytic myeloid-derived suppressor cells (MDSC) have recently been described in first trimester decidua where they induce FOXP3 expression in CD4+CD25- T cells (67).

Of decidual CD4+ T cells, the CD4+ Treg have most widely been studied given their

immune-regulatory function. Higher levels of CD4+CD25+FOXP3+ Treg have been

observed in decidua compared to mPBMC, where preferential recruitment of fetus-

specific Treg from mPBMC to the decidua has been proposed (68, 69). Treg can be

induced by the crosstalk between dNK and CD14+ APC (59), decidual stromal cells

(DSC) (70), and interaction of naïve CD4+ T cells with EVT (46). The importance of Treg in

maintaining a healthy pregnancy was illustrated by the observation of diminished circulating and local proportions of Treg in complicated pregnancies, such as preterm birth, pre-eclampsia and recurrent spontaneous miscarriages (71, 72, 73, 74). In murine models, an increased resorption rate was observed in allogeneic matings, but not syngeneic matings, upon depletion of Treg (75). The exact mechanisms and factors that play a role in diminished Treg numbers and function during pregnancy complications have thus far not been identified. In addition to CD4+ Treg, decidual Th1, Th2 and Th17 CD4+

T cell subsets, with a preference of Th1 over Th2, exist that possess transcriptional profiles of antigen-presentation, proliferation and activation (76, 77). The effector and memory CD4+

T cell compartment is understudied and thus far little is known about the exact antigen- specificity of decidual CD4+ T cell subsets, including CD4+ Treg.

Decidual CD8+ T cells form the other arm of the T cell compartment and present a predominantly effector-memory phenotype. These CD8+ T cells have decreased protein expression of perforin and granzyme B, but higher mRNA content of these cytolytic molecules when compared to peripheral blood CD8+ T cells (78). CD8+ T cells specific for the HY-antigen (male tissue-specific antigen, in case of a male fetus) and HLA-A and -B-restricted virus-specific CD8+ T cells populate the decidua, indicating that decidual CD8+ T cells can recognize viral, fetal and placental antigens during pregnancy (79, 80, 81). To prevent fetal rejection by a cytolytic response of these effector-memory CD8+ T cells, interaction with Treg in concert with a balance between the expression of co-stimulatory and co-inhibitory receptors is essential and will be further discussed in this thesis.

Remaining immune cell populations present at lower frequencies in the decidua are the B cells, and non-conventional T cells such as mucosal associated invariant T (MAIT) cells, natural killer T cells (NKT) and TCRγγ γδ T cells. Reactivation of B cells is suggested in recurrent miscarriages where a higher incidence of anti-HLA-C antibodies was observed in women suffering from recurrent miscarriage (82). MAIT cells are T cells restricted by MHC- related molecule 1(MR1) and abundant in tissues where they can be rapidly activated.

NKT cells have a restricted TCR repertoire, express NK cell markers and do not recognize peptide antigens presented by MHC molecules, but glycolipid antigens presented by the non-polymorphic CD1d molecule. TCRγγ γδ T cells retain a specialized TCR on their cell surface that is believed to predominantly recognize lipid antigens not bound to MHC molecules (51). It is clear that the maternal immune system is aware of paternal antigens expressed by the fetus and plays an active role in bringing pregnancy to a success.

Furthermore, the maternal immune system holds the challenging task of providing

tolerance against the semi-allogeneic fetus, yet simultaneously providing placental

immunity against invading pathogens, a task with a conflicting nature.

01

7. Immunity in complicated pregnancy

Not in all cases is a pregnancy successful. Aberrant functioning of the maternal immune system has been suggested to play a role in pregnancy complications such as pre-eclampsia, recurrent miscarriage, preterm birth and fetal growth restrictions.

Pre-eclampsia (PE) is a pregnancy complication characterized by high blood pressure and proteinuria and the incidence rate is circa 3-5% of pregnancies (83). An imbalance between Treg subsets, lower numbers of decidual CD8+ T cells and NK cells, and aberrantly activated macrophages that secrete pro-inflammatory cytokines may be associated with PE (72, 84, 85). Recurrent miscarriage (RM) is defined as three or more consecutive miscarriages prior to the 20

week of gestation, which is the case in 1 to 2% of all couples trying to conceive (86). Reduced CD4+CD25+FOXP3+ Treg numbers were observed in the decidua and mPBMC of women suffering from miscarriage. Furthermore, increased levels of Th17 cells (generally pro-inflammatory cells) and an inverse relationship between Th17 cells and Treg were found in women with RM. In addition, increased Th1 immunity (pro-inflammatory) has been suggested and increased numbers of NK cells were observed in the mPBMC and decidua of women with RM compared to normal pregnancies (87, 73, 88). Premature birth is defined as babies born alive before 37 weeks of gestation. A role for cells from both the innate and adaptive immune arm have been indicated in preterm birth. Increased numbers of neutrophils, altered proportions of macrophages and the presence of effector and activated T cells at the maternal-fetal interface have been observed during preterm birth (89, 90). In a murine model, an indirect defect in dNK

Figure 4. Decidual leukocytes at the maternal-fetal interface throughout human pregnancy. Adapted from Gomez-Lopez et al. JLB 2010.

Relative Leukocyte Density

Gestational age

Early Late Labor

Granulocytes T cells Mast cells Monocytes/

Macrophages

B cells

NK cells

differentiation and migration was implicated in fetal growth restriction (91). To obtain a better understanding of the immunological component in pregnancy complications it is essential to fully comprehend the immune network and its regulatory mechanisms at the maternal-fetal interface in healthy pregnancy.

8. Outline of this thesis

More than 60 years after the paradox of pregnancy was first described, a gap remains in our understanding of the role and functionality of maternal immune cells and their interactions with fetal cells in accepting the semi-allogeneic fetus during a nine-month pregnancy. Incomplete knowledge of this dynamic network at the maternal-fetal interface has hindered our understanding of aberrant pregnancies such as recurrent miscarriages and pre-eclampsia. The aim of the studies described in this thesis was to further explore the underlying mechanisms of immune regulation and recognition by decidual CD4+ and CD8+ T cells and their interaction with EVT throughout healthy human pregnancy. Moreover, the need for both placental tolerance and immunity and how these two components may be in conflict is addressed. A main driver of immune changes at the maternal-fetal interface is the regulatory CD4+ T cell. Here, we identified three decidual Treg subtypes that suppress effector T cell responses and are induced by EVT and decidual macrophages (Chapter 2).

The presence of Treg appears to not be the only mechanism necessary to maintain tolerance. Decidual CD8+ T cells exhibit a mixed transcriptional profile of activation and dysfunction enabling them to provide both tolerance and immunity (Chapter 3). A normal consequence of human pregnancy is the development of fetus-specific CD8+ T cells, accompanied by the presence of virus-specific CD8+ T cells to protect the fetus against invading pathogens. Virus-specific CD8+ T cells are known to be able to cross-react with allo-HLA-A and -B molecules. Potential cross-reactivity of maternal decidual virus- specific T cells against fetal allo-HLA-C (the only polymorphic HLA molecule expressed by EVT) may have detrimental consequences for pregnancy outcome and is investigated in Chapter 4. Regulation of the cytotoxicity of decidual CD8+ T cells and NK cells and the ability of HLA-C to elicit their effector responses is further described in Chapter 5. Up until this point, functional analyses were performed of specific decidual T cell subsets. Next, we applied mass cytometry to obtain a system-wide and data-driven overview of the dynamics of both the innate and adaptive maternal immune compartment at different time points throughout pregnancy (Chapter 6 and Chapter 7). This resulted in an immune atlas of healthy pregnancy that serves as a foundation for understanding pregnancy complications.

Finally, the major findings in this thesis are described in Chapter 8 and implications of the

presented results for future research and clinical practice are discussed.

01

References

1. W. Holtmeier and D. Kabelitz. gammadelta T cells link innate and adaptive immune responses. Chem Immunol Allergy 86:151-183, 2005.

2. D. Godfrey, H. MacDonald, M. Kronenberg, M. Smyth and L. Van Kaer. NKT cells: what’s in a name?. Nat Rev Immunol 4:231-237, 2004.

3. A. Iwasaki and R. Medzhitov. Control of adaptive immunity by the innate immune system. Nature Immunology 16:343-353, 2015.

4. P. Parham. The immune system. Garland Science 4th edition, 2014.

5. J. Mccluskey and Peh Ca. The human leucocyte antigens and clinical medicine: an overview. Rev Immunogenet 1:3-20, 1999.

6. P. Hornick and R. Lechler. Direct and indirect pathways of alloantigen recognition: relevance to acute and chronic allograft rejection. Nephrol Dial Transplant 12:1806-1810, 1997.

7. J. Dausset. Iso-leuco-anticorps. Acta Haematol 20:156–66, 1958.

8. J. Van Rood, J. Eernisse and A. van Leeuwen. Leucocyte antibodies in sera from pregnant women. Nature 181:1735-1736, 1958.

9. H. Ljunggren and K. Kärre. In search of the ‘missing self’: MHC molecules and NK cell recognition. Immunology Today 11:237-244, 1990.

10. L. Castagna and D. Mavilio. Re-discovering NK cell allo-reactivity in the therapy of solid tumors. Journal for ImmunoTherapy of Cancer 4:54, 2016.

11. H. Takaba and H. Takayanagi. The mechanisms of T cell selection in the thymus. Trends in Immunology 38:805-816, 2017.

12. A. E. Moran and K. A. Hogquist. T-cell receptor affinity in thymic development. Immunology 135:261-267, 2012.

13. J. M. Curtsinger, C. S. Schmidt, A. Mondino, D. C. Lins, R. M. Kedl, M. K. Jenkins and M. F. Mescher. Inflammatory cytokines provide a third signal for activation of naive CD4+ and CD8+ T cells. Journal of Immunology 162:3256-62, 1999.

14. J. Geginat, M. Paroni, S. Maglie, J. S. Alfen, I. Kastirr, P. Gruarin, M. de Simone, M. Pagani and S. Abrignani.

Plasticity of human CD4 T cell subsets. Frontiers in Immunology 5:630, 2014.

15. N. D. Pennock, J. T. White, E. W. Cross, E. E. Cheney, B. A. Tamburini and R. M. Kedl. T cell responses: naïve to memory and everything in between. Advances in Physiology Education 37:273-283, 2013.

16. D. A. A. Vignali, L. W. Collison and C. J. Workman. How regulatory T cells work. Nature Reviews Immunology 8 523-532, 2008.

17. M. E. Himmel, K. G. MacDonald, R. Garcia, T. Steiner and M. K. Levings. Helios+ and Helios- cells coexist within the natural FOXP3+ T regulatory cell subset in humans. The Journal of Immunology 190:2001-2008, 2013.

18. D. O. Adeegbe and H. Nishikawa. Natural and induced T regulatory cells in cancer. Frontiers in Immunology 4:1-14, 2013.

19. D. Mason. A very high level of crossreactivity is an essential feature of the T-cell receptor. Immunology Today 19:395-404, 1998.

20. M. Koutsakos, P. T. Illing, T. H. Nguyen, N. A. Mifsud, J. C. Crawford, S. Rizzetto, A. A. Eltahla, E. B. Clemens, S. Sant, B. Y. Chua, C. Y. Wong, E. K. Allen, D. Teng, P. Dash, D. F. Boyd, L. Grzelak, W. Zeng, A. C. Hurt, I. Barr, S.

Rockman, D. C. Jackson, T. C. Kotsimbos, A. C. Cheng, M. Richards, G. P. Westall, T. Loudovaris, S. I. Mannering, M. Elliott, S. G. Tangye, L. M. Wakim, J. Rossjohn, D. Vijaykrishna, F. Luciani, P. G. Thomas, S. Gras, A. W. Purcell and K. Kedzierska. Human CD8 + T cell cross-reactivity across influenza A, B and C viruses. Nature Immunology 20:613-625, 2019.

21. H. Wedemeyer, E. Mizukoshi, A. R. Davis, J. R. Bennink and B. Rehermann. Cross-reactivity between Hepatitis C

Virus and Influenza A Virus determinant-specific cytotoxic T cells. Journal of Virology 75:11392-11400, 2001.

22. S. C. Clute, Y. N. Naumov, L. B. Watkin, N. Aslan, J. L. Sullivan, D. A. Thorley-Lawson, K. Luzuriaga, R. M. Welsh, R. Puzone, F. Celada and L. K. Selin. Broad cross-reactive TCR repertoires recognizing dissimilar Epstein-Barr and Influenza A Virus epitopes. The Journal of Immunology 185:6753-6764, 2010.

23. J. A. Regla-Nava, A. Elong Ngono, K. M. Viramontes, A. T. Huynh, Y. T. Wang, A. V. T. Nguyen, R. Salgado, A.

Mamidi, K. Kim, M. S. Diamond and S. Shresta. Cross-reactive Dengue virus-specific CD8+ T cells protect against Zika virus during pregnancy. Nature Communications 9:3042, 2018.

24. J. S. H. Gaston, A. B. Rickinson and M. A. Epstein. Cross-reactivity of self-HLA-restricted Epstein-Barr virus- specific cytotoxic T lymphocytes for allo-HLA determinants. J. Exp. Med 158:1804-1821, 1983.

25. A. L. Amir, L. J. A. D’Orsogna, D. L. Roelen, M. M. Van Loenen, R. S. Hagedoorn, R. De Boer, M. A. W. G. Van Der Hoorn, M. G. D. Kester, I. I. N. Doxiadis, J. H. F. Falkenburg, F. H. J. Claas and M. H. M. Heemskerk. Allo-HLA reactivity of virus-specific memory T cells is common. Blood 115:3146-3157, 2010.

26. L. J. A. D’Orsogna, D. L. Roelen, I. I. N. Doxiadis and F. H. J. Claas. Alloreactivity from human viral specific memory T-cells. Transplant Immunology 23:149-155, 2010.

27. A. B. Adams, M. A. Williams, T. R. Jones, N. Shirasugi, M. M. Durham, S. M. Kaech, E. J. Wherry, T. Onami, J. G.

Lanier, K. E. Kokko, T. C. Pearson, R. Ahmed and C. P. Larsen. Heterologous immunity provides a potent barrier to transplantation tolerance. Journal of Clinical Investigation 111:1887–1895, 2003.

28. J. Attanasio and E. J. Wherry. Costimulatory and coinhibitory receptor pathways in infectious disease. Immunity 44:1052-1068, 2016.

29. I. S. Okoye, M. Houghton, L. Tyrrell, K. Barakat and S. Elahi. Coinhibitory receptor expression and immune check- point blockade: Maintaining a balance in CD8 + T cell responses to chronic viral infections and cancer. Frontiers in Immunology 8:1215, 2017.

30. A. C. Anderson, N. Joller and V. K. Kuchroo. Lag-3, Tim-3, and TIGIT: Co-inhibitory receptors with specialized functions in immune regulation. Immunity 44:989-1004, 2016.

31. R. H. Schwartz. T cell anergy. Annual Review of Immunology 21:305-334, 2003.

32. E. J. Wherry and M. Kurachi. Molecular and cellular insights into T cell exhaustion. Nat Rev Immunol 15:486-99, 2015.

33. S. H. Baumeister, G. J. Freeman, G. Dranoff and A. H. Sharpe. Coinhibitory pathways in immunotherapy for cancer. Annual Review of Immunology 34:539-573, 2016.

34. M. L. Ford. T cell cosignaling molecules in transplantation. Immunity 44:1020-1033, 2016.

35. Q. Zhang and D. A. Vignali. Co-stimulatory and co-inhibitory pathways in autoimmunity. Immunity 44:1034-1051, 2016.

36. D. Anderson, R. E. Billingham, G. H. Lampkin and P. B. Medawar. The use of skin grafting to distinguish between monozygotic and dizygotic twins in cattle. Heredity 5:379-397, 1951.

37. R. Billingham, L. Brent and P. Medawar. The antigenic stimulus in transplantation immunity. Nature 178:514-519, 1956.

38. P. Medawar. Some immunological and endocrinological problems raised by the evolution of viviparity in vertebrates. Symp Soc Exp Biol 7:320–338, 1953.

39. R. W. Su and A. T. Fazleabas. Implantation and establishment of pregnancy in human and nonhuman primates.

Advances in Anatomy Embryology and Cell Biology 216:189-213, 2015.

40. D. W. Bianchi, G. K. Zickwolf, G. J. Weil, S. Sylvester and M. A. DeMaria. Male fetal progenitor cells persist in maternal blood for as long as 27 years postpartum. Proceedings of the National Academy of Sciences of the United States of America 93:705-708, 1996.

41. S. Maloney, A. Smith, D. Furst, D. Myerson, K. Rupert, P. Evans and J. Nelson. Microchimerism of maternal origin persists into adult life. J Clin Invest 104:41-47, 1999.

42. A. Moffett and C. Loke. Immunology of placentation in eutherian mammals. Nat Rev Immunol 6:584-594, 2006.

43. A. M. Sharkey, L. Gardner, S. Hiby, L. Farrell, R. Apps, L. Masters, J. Goodridge, L. Lathbury, C. A. Stewart, S.

Verma and A. Moffett. Killer Ig-Like Receptor expression in uterine NK cells is biased toward recognition of HLA-C and

alters with gestational age. The Journal of Immunology 181:39-46, 2014.

01

44. T. Tilburgs, S. A. Scherjon, B. J. van der Mast, G. W. Haasnoot, M. Versteeg-v.d.Voort-Maarschalk, D. L. Roelen, J. J. van Rood and F. H. Claas. Fetal-maternal HLA-C mismatch is associated with decidual T cell activation and induction of functional T regulatory cells. Journal of Reproductive Immunology 82:148-157, 2009.

45. L. Pazmany, O. Mandelboim, M. Vales-Gomez, D. Davis, H. Reyburn and J. Strominger. Protection from natural killer cell-mediated lysis by HLA-G expression on target cells. Science 274:792-795, 1996.

46. T. Tilburgs, Â. C. Crespo, A. van der Zwan, B. Rybalov, T. Raj, B. Stranger, L. Gardner, A. Moffett and J. L.

Strominger. Human HLA-G+ extravillous trophoblasts: Immune-activating cells that interact with decidual leukocytes.

Proceedings of the National Academy of Sciences 112:7219-7224, 2015.

47. I. Tirado-gonzález, N. Freitag, G. Barrientos, V. Shaikly, O. Nagaeva, M. Strand, L. Kjellberg, B. F. Klapp, L.

Mincheva-nilsson, M. Cohen and S. M. Blois. Galectin-I influences trophoblast immune evasion and emerges as a predictive factor for the outcome of pregnancy. Molecular Human Reproduction 19:43-53, 2013.

48. A. Hönig, L. Rieger, M. Kapp, M. Sütterlin, J. Dietl and U. Kämmerer. Indoleamine 2,3-dioxygenase (IDO) expression in invasive extravillous trophoblast supports role of the enzyme for materno-fetal tolerance. Journal of Reproductive Immunology 61:79-86, 2004.

49. V. M. Abrahams, S. L. Straszewski-Chavez, S. Guller and G. Mor. First trimester trophoblast cells secrete Fas ligand which induces immune cell apoptosis. Molecular Human Reproduction 10:55-63, 2004.

50. T. A. Phillips, J. Ni, G. Pan, S. M. Ruben, Y. F. Wei, J. L. Pace and J. S. Hunt. TRAIL (Apo-2L) and TRAIL receptors in human placentas: implications for immune privilege. Journal of immunology 162:6053-6059, 1999.

51. M. E. Solano. Decidual immune cells: Guardians of human pregnancies. Best Practice & Research Clinical Obstetrics & Gynaecology, 2019.

52. P. Luppi. How immune mechanisms are affected by pregnancy. Vaccine. 21:3352-3357, 2003.

53. N. Gomez-Lopez, L. J. Guilbert and D. M. Olson. Invasion of the leukocytes into the fetal-maternal interface during pregnancy. Journal of Leukocyte Biology 88:625-633, 2010.

54. L. A. Koopman, H. D. Kopcow, B. Rybalov, J. E. Boyson, J. S. Orange, F. Schatz, R. Masch, C. J. Lockwood, A.

D. Schachter, P. J. Park and J. L. Strominger. Human decidual natural killer cells are a unique NK cell subset with immunomodulatory potential. The Journal of Experimental Medicine 198:1201-1212, 2003.

55. A. Moffett and F. Colucci. Uterine NK cells: Active regulators at the maternal-fetal interface. The American Society for Clinical Investigation 124:1872-1879, 2014.

56. Â. C. Crespo, J. L. Strominger and T. Tilburgs. Expression of KIR2DS1 by decidual natural killer cells increases their ability to control placental HCMV infection. Proceedings of the National Academy of Sciences 113:15072-15077, 2016.

57. S. Gordon and A. Plüddemann. Tissue macrophages: Heterogeneity and functions. BMC Biology 15:53, 2017.

58. J. Svensson-Arvelund and J. Ernerudh. The role of macrophages in promoting and maintaining homeostasis at the fetal-maternal interface. The American Journal of Reproductive Immunology 74:100-109, 2015.

59. P. Vacca, C. Cantoni, M. Vitale, C. Prato, F. Canegallo, D. Fenoglio, N. Ragni, L. Moretta and M. C. Mingari. Cross- talk between decidual NK and CD14+ myelomonocytic cells results in induction of Tregs and immunosuppression.

Proceedings of the National Academy of Sciences 107:11918-11923, 2010.

60. S. Sayama, T. Nagamatsu, D. J. Schust, N. Itaoka, M. Ichikawa, K. Kawana, T. Yamashita, S. Kozuma and T. Fujii.

Human decidual macrophages suppress IFN- γ production by T cells through costimulatory B7-H1: PD-1 signaling in early pregnancy. Journal of Reproductive Immunology 100:109-117, 2013.

61. G. E. Lash, H. Pitman, H. L. Morgan, B. A. Innes, C. N. Agwu and J. N. Bulmer. Decidual macrophages: key regu- lators of vascular remodeling in human pregnancy. Journal of Leukocyte Biology 100:315-325, 2016.

62. M. M. Faas and P. De Vos. Innate immune cells in the placental bed in healthy pregnancy and preeclampsia.

Placenta 69:125-133, 2018.

63. L. Gorvel, A. Ben Amara, M. B. Ka, J. Textoris, J.-P. Gorvel and J.-L. Mege. Myeloid decidual dendritic cells and immunoregulation of pregnancy: defective responsiveness to Coxiella burnetii and Brucella abortus. Frontiers in Cellular and Infection Microbiology 4:179, 2014.

64. S. Miyazaki, H. Tsuda, M. Sakai, S. Hori, Y. Sasaki, T. Futatani, T. Miyawaki and S. Saito. Predominance of Th2-

promoting dendritic cells in early human pregnancy decidua. Journal of Leukocyte Biology 74:514-522, 2003.

65. P. Hsu, B. Santner-Nanan, J. E. Dahlstrom, M. Fadia, A. Chandra, M. Peek and R. Nanan. Altered decidual DC-SIGN+ antigen-presenting cells and impaired regulatory T-cell induction in preeclampsia. American Journal of Pathology 181:2149-2160, 2012.

66. U. Kämmerer, A. O. Eggert, M. Kapp, A. D. McLellan, T. B. Geijtenbeek, J. Dietl, Y. Van Kooyk and E. Kämpgen.

Unique appearance of proliferating antigen-presenting cells expressing DC-SIGN (CD209) in the decidua of early human pregnancy. American Journal of Pathology 162:887-896, 2003.

67. X. Kang, X. Zhang, Z. Liu, H. Xu, T. Wang, L. He and A. Zhao. Granulocytic myeloid-derived suppressor cells main- tain feto-maternal tolerance by inducing Foxp3 expression in CD4+CD25- T cells by activation of the TGF-γ/γ-catenin pathway. Molecular Human Reproduction 22:499-511, 2016.

68. Y. Sasaki, M. Sakai, S. Miyazaki, S. Higuma, A. Shiozaki and S. Saito. Decidual and peripheral blood CD4+

CD25+ regulatory T cells in early pregnancy subjects and spontaneous abortion cases. Molecular Human Reproduction 10:347-353, 2004.

69. T. Tilburgs, D. L. Roelen, B. J. van der Mast, G. M. de Groot-Swings, C. Kleijburg, S. A. Scherjon and F. H. Claas.

Evidence for a selective migration of fetus-specific CD4+CD25bright regulatory T cells from the peripheral blood to the decidua in human pregnancy. Journal of Immunology 180:5737-5745, 2008.

70. T. Erkers, S. Nava, J. Yosef, O. Ringdén and H. Kaipe. Decidual stromal cells promote regulatory T cells and suppress alloreactivity in a cell contact-dependent manner. stem cells and development 22:2596-2605, 2013.

71. L. Schober, D. Radnai, E. Schmitt, K. Mahnke, C. Sohn and A. Steinborn. Term and preterm labor: Decreased suppressive activity and changes in composition of the regulatory T-cell pool. Immunology and Cell Biology 90:935-944, 2012.

72. A. Steinborn, G. M. Haensch, K. Mahnke, E. Schmitt, A. Toermer, S. Meuer and C. Sohn. Distinct subsets of regulatory T cells during pregnancy: Is the imbalance of these subsets involved in the pathogenesis of preeclampsia?

Clinical Immunology 129:401-412, 2008.

73. K. Inada, T. Shima, A. Nakashima, K. Aoki, M. Ito and S. Saito. Characterization of regulatory T cells in decidua of miscarriage cases with abnormal or normal fetal chromosomal content. Journal of Reproductive Immunology 97:104-111, 2013.

74. K. Inada, T. Shima, M. Ito, A. Ushijima and S. Saito. Helios-positive functional regulatory T cells are decreased in decidua of miscarriage cases with normal fetal chromosomal content. Journal of reproductive immunology 107:10-19, 2015.

75. V. R. Aluvihare, M. Kallikourdis and A. G. Betz. Regulatory T cells mediate maternal tolerance to the fetus. Nature Immunology 5:266-271, 2004.

76. J. Mjösberg, G. Berg, M. C. Jenmalm and J. Ernerudh. FOXP3+ regulatory T cells and T helper 1, T helper 2, and T helper 17 cells in human early pregnancy decidua. Biology of Reproduction 82:698-705, 2010.

77. W. Zeng, Z. Liu, X. Liu, S. Zhang, A. Khanniche, Y. Zheng, X. Ma, T. Yu, F. Tian, X. R. Liu, J. Fan and Y. Lin. Distinct transcriptional and alternative splicing signatures of decidual CD4+T cells in early human pregnancy. Frontiers in Immunology 8:682, 2017.

78. T. Tilburgs, D. Schonkeren, M. Eikmans, N. M. Nagtzaam, G. Datema, G. M. Swings, F. Prins, J. M. van Lith, B.

J. van der Mast, D. L. Roelen, S. A. Scherjon and F. H. Claas. Human decidual tissue contains differentiated CD8+

effector-memory T cells with unique properties. The Journal of Immunology 185:4470-4477, 2010.

79. D. Lissauer, K. Piper, O. Goodyear, M. D. Kilby and P. A. H. Moss. Fetal-specific CD8+ cytotoxic T cell responses develop during normal human pregnancy and exhibit broad functional capacity. The Journal of Immunology 189:1072-1080, 2012.

80. T. Tilburgs and J. L. Strominger. CD8+ effector T cells at the fetal-maternal interface, balancing fetal tolerance and antiviral immunity. American Journal of Reproductive Immunology 69:395-407, 2013.

81. A. van Egmond, C. van der Keur, G. M. J. S. Swings, S. A. Scherjon and F. H. J. Claas. The possible role of virus-specific CD8+ memory T cells in decidual tissue. Journal of Reproductive Immunology 113:1-8, 2016.

82. T. Meuleman, E. van Beelen, R. Kaaja, J. van Lith, F. Claas and K. Bloemenkamp. HLA-C antibodies in women with recurrent miscarriage suggests that antibody mediated rejection is one of the mechanisms leading to recurrent miscarriage. Journal of Reproductive Immunology 116:28-34, 2016.

83. B. W. Mol, C. T. Roberts, S. Thangaratinam, L. A. Magee, C. J. De Groot and G. J. Hofmeyr. Pre-eclampsia. The

Lancet 387:999-1011, 2016.

01

84. L. Rieger, S. Segerer, T. Bernar, M. Kapp, M. Majic, A. K. Morr, J. Dietl and U. Kämmerer. Specific subsets of immune cells in human decidua differ between normal pregnancy and preeclampsia - a prospective observational study. Reproductive Biology and Endocrinology 7:132, 2009.

85. F. Ning, H. Liu and G. E. Lash. The role of decidual macrophages during normal and pathological pregnancy.

American Journal of Reproductive Immunology 75:298-309, 2016.

86. C. Coulam. Epidemiology of recurrent spontaneous abortion. American Journal of Reproductive Immunology 26:23-27, 1991.

87. A. S. Bansal. Joining the immunological dots in recurrent miscarriage. American Journal of Reproductive Immunology 64:307-315, 2010.

88. P. M. Emmer, W. L. D. M. Nelen, E. A. P. Steegers, J. C. M. Hendriks, M. Veerhoek and I. Joosten. Peripheral natural killer cytotoxicity and CD56posCD16pos cells increase during early pregnancy in women with a history of recurrent spontaneous abortion. Human Reproduction 15:1163-1169, 2000.

89. N. Gomez-Lopez, D. StLouis, M. A. Lehr, E. N. Sanchez-Rodriguez and M. Arenas-Hernandez. Immune cells in term and preterm labor. Cellular and Molecular Immunology 11:571-581, 2014.

90. M. Arenas-Hernandez, R. Romero, Y. Xu, B. Panaitescu, V. Garcia-Flores, D. Miller, H. Ahn, B. Done, S. S. Hassan, C.-D. Hsu, A. L. Tarca, C. Sanchez-Torres and N. Gomez-Lopez. Effector and activated T cells induce preterm labor and birth that is prevented by treatment with progesterone. The Journal of Immunology 202:2585-2608, 2019.

91. T. Nagashima, Q. Li, C. Clementi, J. P. Lydon, F. J. DeMayo and M. M. Matzuk. BMPR2 is required for

postimplantation uterine function and pregnancy maintenance. Journal of Clinical Investigation 123:2539-2550, 2013.

Three types of functional regulatory T cells control T cell responses at the human maternal-fetal interface

Maria Salvany-Celades*, Anita van der Zwan*, Marilen Benner*, Vita Setrajcic-Dragos, Hannah Ananda Bougleux Gomes,Vidya Iyer,

Errol R. Norwitz, Jack L. Strominger, Tamara Tilburgs

*These authors contributed equally Cell Rep. 2019 May 28;27(9):2537-2547

Chapter 02

Summary

During pregnancy, maternal regulatory T cells (Tregs) are important in establishing immune tolerance to invading fetal extravillous trophoblasts (EVTs). CD25

FOXP3+ Tregs are found at high levels in decidual tissues and have been shown to suppress fetus-specific and non-specific responses. However, limited data are available on additional decidual Treg types and the mechanisms by which they are induced. This study investigated three distinct decidual CD4+ Treg types in healthy pregnancies with a regulatory phenotype and the ability to suppress T cell responses: CD25

FOXP3+, PD1

IL-10+, and TIGIT+

FOXP3

. Moreover, co-culture of HLA-G+ EVTs or decidual macrophages with blood

CD4+ T cells directly increased the proportions of CD25

FOXP3+ Tregs compared to

T cells cultured alone. EVTs also increased PD1

Tregs that could be inhibited by HLA-C and

CD3 antibodies, suggesting an antigen-specific induction. The presence of distinct Treg

types may allow for the modulation of a variety of inflammatory responses in the placenta.

02

Introduction

During pregnancy, CD4+CD25

FOXP3+ regulatory T cells (Tregs) are found at high levels in decidual tissue and have the ability to suppress fetus-specific and non-specific responses (1, 2). Most interestingly, HLA-C mismatched pregnancies (where the fetus and extravillous trophoblasts (EVTs) express an HLA-C allotype that the mother does not have) had increased levels of functional CD4+CD25

Tregs in decidua, compared to HLA-C matched pregnancies (3). Furthermore, in vitro co-culture of naive CD4+ T cells with EVTs directly increased the proportion of CD4+FOXP3+ Tregs, compared to CD4+ T cells cultured alone (4, 5, 6). This suggests that maternal T cells may specifically recognize fetal HLA-C, but its expression on EVTs promotes immune tolerance. The importance of maternal immune tolerance for fetal HLA-C is further illustrated by a recent study suggesting that HLA-C antibodies may contribute to the etiology of miscarriage (7). The proportion of circulating FOXP3+ Tregs was shown to be diminished in maternal blood obtained after spontaneous preterm birth (SPTB) (8, 9, 10), preeclampsia (PE) (11, 12) and in decidual tissue obtained after recurrent spontaneous miscarriage (13, 14). Furthermore, clonally expanded CD4+ CD25

CD127–CD45RA– Treg populations were observed in healthy term pregnancy decidua, and failure of this clonal expansion may be related to development of preeclampsia (15). The importance of Tregs was also demonstrated in murine pregnancy models (16, 17, 18, 19). Depletion of CD25+ Tregs during allogeneic matings, but not syngeneic matings, resulted in an increased resorption rate (16). Besides highlighting the role for Tregs, this also demonstrated that in the absence of Tregs, effector cells cause immunologic rejection of allogeneic fetal or placental tissues. A more recent murine study demonstrated that FOXP3+ Tregs with specificity to paternal antigens were generated extrathymically and accumulated in the placenta. In this study, females with impaired ability for extrathymic Treg induction showed increased fetal resorption rates and had increased influx of immune cells to the placenta in allogeneic matings, but not syngeneic matings (20). Other pathways and molecules have been demonstrated to play a role in Treg induction during pregnancy, including the blockade of the PD1-PDL1 pathway that led to reduced decidual CD25+FOXP3+ Treg numbers and increased embryo resorption in mice, which could be abrogated by adoptive CD25+FOXP3+ Treg transfer (21). This demonstrates the importance of PD1-PDL1 for decidual Treg induction. However, specific factors that contribute to diminished decidual Treg numbers or function during pregnancy complications have not been identified in human pregnancy.

Thus far, research on the role of Tregs in human pregnancy has mainly focused on FOXP3+

Tregs (22, 13, 14, 8, 9, 23, 25, 10, 2,15), while other types of FOXP3- Tregs have not been studied in as much detail. Most importantly, only a handful of studies provide functional analysis of peripheral blood (11, 12) and decidual Tregs (2, 3) during human pregnancy.

FOXP3, in combination with high expression of CD25 and HELIOS and the absence of

CD127 expression, primarily identifies natural Tregs (nTregs), although whether HELIOS is a defining marker for human nTregs remains controversial. The nTregs are generated in the thymus, are specific for self-antigens, and are responsible for preventing anti-self (autoimmune) responses (26, 27). In contrast, induced Tregs (iTregs) are generated in the periphery and can be specific for a large variety of antigens, including allo-antigens and viral-antigens (28, 29, 30). A well-characterized type of iTregs are Tr1 cells that secrete high levels of IL-10; express PD1; co-express CD49b and LAG3, but do not express FOXP3; and are important in the control of alloimmune responses (31, 32). Other iTregs include TIGIT+

cells that modulate antigen-presenting cells (APC) through interaction with CD155 on APCs and Tr35 cells that function through secretion of IL-35, an immune suppressive cytokine (33, 34). A large variety of other markers have been used to identify distinct iTreg populations (including but not limited to FOXP3, CD25, GITR, TIM3, CD39, LRRC32 (also known as GARP), LAP, and CCR8) (35, 36, 37, 38). None of these markers are truly specific for iTregs, as they can also be expressed on activated T cells. Thus, to identify iTregs, functional assays are required to demonstrate their capacity to suppress immune responses such as proliferation, cytokine secretion, and cytotoxicity (39, 30). In this study, we provide extensive phenotypic and functional characterization of three types of decidual CD4+

Tregs in uncomplicated human pregnancies and investigate the ability of HLA-G+ HLA-C+

EVTs and decidual macrophages, the main APCs at the maternal-fetal interface, to increase Treg proportions.

Results

Distinct CD4+ T cell types with a regulatory phenotype are present in decidual tissue FACS analysis on freshly isolated peripheral blood CD4+ T cells (CD4+ pTs) and decidual CD4+ T cells (CD4+ dTs) isolated from first-trimester decidua (gestational age 6–12 weeks) and term placenta decidua basalis (d.basalis) and decidua parietalis (d.parietalis) (gestational age > 37 weeks) was performed to determine cell surface expression of CD45, CD4, CD25, PD1, TIGIT, CD127, CD45RA, CD49b, and LAG3 and intracellular expression of FOXP3 and HELIOS. A clear population of activated nTregs was identified in all tissues based on the high expression of CD25, FOXP3, and HELIOS and the lack of CD45RA and CD127 (Fig. 1; Fig. S1A-S1D) (26). While the percentage of FOXP3+ and HELIOS+ cells within this CD25

population significantly decreased in term pregnancy decidua, the proportion of HELIOS+ cells within CD25

FOXP3+ cells remained relatively stable (Fig. 1D, right panel). A second T cell population was identified based on the high expression of PD1, the lack of FOXP3 and HELIOS, and low CD25. LAG3 and CD49b were not expressed by CD4+ dTs. While both CD25

and PD1

cells co-expressed high levels of TIGIT, a third population of CD4+ dTs also expressed high levels of TIGIT and low levels of FOXP3, HELIOS, PD1, and CD25 (Fig. 1; Fig. S1A-S1D). FOXP3 expression in TIGIT+

cells was significantly lower than in CD25

cells (Fig. S1E). The t-Distributed Stochastic

Neighbor Embedding (t-SNE) analysis confirmed the separation of these three T cell

02

Figure 1. Three distinct CD4+ T cell populations express Treg-associated markers.

(A and B) Representative FACS plots (A) and percentages (B) of CD25

, PD1

, and TIGIT+ cells within CD4+

T cells in blood, decidua 6–12 weeks, d.basalis >37 weeks, and d.parietalis >37 weeks. (C and D) Representative FACS plots (C) and percentages (D) of FOXP3+ and HELIOS+ cells within the four CD4+ T cell types and HELIOS+

cells within CD25

FOXP3+ cells in blood, decidua 6–12 weeks, d.basalis >37 weeks, and d.parietalis >37 weeks (n = 8–10). Bars represent median and interquartile range; *p < 0.05, **p < 0.01 and ***p < 0.001.

See also Figures S1-S3.