Placenta 35 (2014) 241e248

Contents lists available at ScienceDirect

Placenta journal homepage: www.elsevier.com/locate/placenta

Classics revisited

Potentiating maternal immune tolerance in pregnancy: A new challenging role for regulatory T cells J. Alijotas-Reig a, b, *, E. Llurba c, J.Ma. Gris d a Systemic Autoimmune Disease Unit, Department of Internal Medicine I, Vall d’Hebron University Hospital, Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain b Department of Medicine, Faculty of Medicine, Universitat Autonoma, Barcelona, Spain c High Obstetric Risk Unit, Obstetric Department, Vall d’Hebron University Hospital, Universitat Autonoma, Barcelona, Spain d Reproductive Medicine Unit, Obstetric Department, Vall d’Hebron University Hospital, Universitat Autonoma, Barcelona, Spain

a r t i c l e i n f o

a b s t r a c t

Article history: Accepted 4 February 2014

The maternal immune system needs to adapt to tolerate the semi-allogeneic conceptus. Since maternal allo-reactive lymphocytes are not fully depleted, other local/systemic mechanisms play a key role in altering the immune response. The Th1/Th2 cytokine balance is not essential for a pregnancy to be normal. The immune cells, CD4þCD25þFoxp3þ, also known as regulatory T cells (Tregs), step in to regulate the allo-reactive Th1 cells. In this review we discuss the role of Tregs in foeto-maternal immune tolerance and in recurrent miscarriage as well as their potential use as a new target for infertility treatment. Animal and human experiments showed Treg cell number and/or function to be diminished in miscarriages. Murine miscarriage can be prevented by transferring Tregs from normal pregnant mice. Tregs at the maternalefetal interface prevented fetal allo-rejection by creating a “tolerant” microenvironment characterised by the expression of IL-10, TGF-b and haem oxygenase isoform 1 (HO-1) rather than by lowering Th1 cytokines. Tregs increase placental HO-1. In turn, HO-1 may lead to up-regulation of TGF-b, IL-10 and CTLA-4. In vivo experiments showed Tregs sensitisation from paternal antigens to be essential for maternalefetal tolerance. Tregs increase throughout pregnancy and diminish in late puerperium. Recent data also support the capacity of Tregs to block maternal effector T cells, thereby reducing the maternalefetal pathological responses to paternal antigens. These findings also permit us to consider new strategies for improving pregnancy outcomes, i.e., anti-TNF blockers and granulocyte-colony stimulating factors as well as novel approaches to therapeutically exploiting Treg þ cell memory. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Cytokines Maternalefetal tolerance Miscarriages Regulatory-T lymphocytes NK cells/KIR Treatment

1. Introduction One of the foremost enigmas of reproductive biology is that healthy women with a normal functional immune system may successfully carry a semi-allogeneic gestation to term without apparent immune rejection, since the maternal immune system can undergo changes for the fetus, which acts as a semi-allogeneic implant, to be tolerated [1]. Human reproduction is a low efficiency process. Generally, almost 70% of embryos are lost in healthy women. Miscarriage occurs in approximately 15% of cases, and recurrent miscarriages in 2e5% [2]. Fetal alloantigens encoded by polymorphic genes * Corresponding author. Systemic Autoimmune Disease Unit, Department of Internal Medicine I, Vall d’Hebron University Hospital, Passeig Vall d’Hebron 119-129, 08035 Barcelona, Spain. Tel.: þ34 93 489 41 94; fax: þ34 93 489 30 39. E-mail addresses: [email protected], [email protected] (J. Alijotas-Reig). http://dx.doi.org/10.1016/j.placenta.2014.02.004 0143-4004/Ó 2014 Elsevier Ltd. All rights reserved.

inherited from the father ought to provoke a maternal immune response leading to fetal rejection; however, this does not normally occur [3]. Acceptance of a semi-allogeneic conceptus supports the involvement of systemic regulatory processes during a normal pregnancy [4,5]. In recent years, early non-chromosome-related miscarriages have been associated with rejection of the fetus by the maternal immune system, which has led to the use of immunosuppressant drugs such as glucocorticoids, lipid emulsions, intravenous immunoglobulin or paternal lymph-mononuclear cell therapy, with disappointing results according to a recent meta-analysis [3,6]. The exact mechanisms by which the maternal immune system tolerates a semi-allogeneic fetus without immunological rejection remain poorly understood. For many years, pregnancy was considered to be a temporary immune-suppressed state. However, researchers are beginning to consider this situation as a fetale

242

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

Nomenclature APC antigen presenting cells ART assisted reproductive techniques CD25 high-affinity a-chain of the interleukin 2 receptor CD4þCD25þFoxp3þ regulatory-T lymphocytes CNS1 conserved noncoding sequence 1 CTLA4 cytotoxic T-Lymphocyte Antigen 4, also known as CD152 Foxp3 forkhead box protein 3 also known as scurfin DCs dendritic cells IDO indoleamine 2,3-dioxygenase IL interleukin LIF leukaemia inhibitor factor HMOX1 haem oxygenase decycling1 (human gene that encodes for the enzyme OH-1)

maternal tolerant symbiosis [7,8]. Since maternal alloreactive lymphocytes are not systemically depleted, local mechanisms have to play a key role in evading the immune response [7,9]. The special characteristics of fetal trophoblasts in contact with maternal tissue have been considered to aid immune tolerance via mechanisms such as low tryptophan levels, lack of accurate activation of NK cells through HLA-G/C expression, high progesterone levels, absence of classical HLA Class I and Class II trophoblast expression, and antiidiotype network modulation [4,7,9e12] (Table 1). However, none of these proposed mechanisms can fully explain maternal tolerance of the fetus during pregnancy. An altered Th1/Th2 cytokine balance with Th2 predominance and T-cell transient anergy has been suggested as a possible mechanism for determining the survival of the fetus in the womb [9,13]. However, studies conducted in genetically-deficient mice with an inability to secrete Th2 cytokines did not always yield miscarriages, which would suggest that Th2 cytokines are not essential for normal pregnancy [7,14,15]. Thus, alloreactive Th1 cells must be differently blocked or regulated, for example, by regulatory T cells (Tregs) [7,13]. Recently, Rowe et al

Table 1 Proposed mechanism involved in the maternalefetal tolerance. 1. Maternal alloreactive T-lymphocyte depletion 2. Alloreactive Th1 cell blocking 3. Increased of both, progesterone and progesterone-induced blocking factor (PIBF) levels 4. Low tryptophan levels 5. High indoleamine 2,3-dioxygenase (IDO) levels 6. Increase of local levels of leukaemia inhibitory factor (LIF) 7. High heme oxygenase isoform (HO-1) levels 8. Increasing apoptosis of activated maternal lymphocytes (Fas/FasL) 9. Increase activity of Programmed Cell Death 1 protein (PD-1) 10. Specific profile of NK cells during pregnancy 11. Change in the balance of NK3/NK1 cells 12. Intrinsic resistance to NK cell mediated lysis 13. Lack of trophoblast HLA I/II expression 14. Up-regulation of trophoblastic HLA-G/E expression 15. Presence of cytotoxic, asymmetric antibodies directed against paternal classical HLA 16. Anti-idiotype network modulation 17. Diminishing complement system activity 18. Predominance of Th2 cytokine balance 19. [Transforming Growth Factor beta (TGF-b) isoforms 20. Increased of the regulatory T-cells (CD4þCD25þFoxp3þ) From: Somersen et al. [4]; Zenclussen et al. [5]; Aluvihare et al. [8]; Tafuri et al. [9]; Nardi et al. [10]; Nermine et al. [11]; Alijotas-Reig J [12]; Raghupathy R [13]; Zenclussen et al. [14]; Rowe et al. [16]; Miwa et al. [35]; Sollwedel et al. [37]; Stewartet al, [39].

HO-1 Haem oxygenase isoform 1 RM recurrent miscarriages TGF-b transforming growth factor-beta Th1 cytokine pro-inflammatory cytokines (mainly, IL-2, IFNgamma, TNF-a) Th2 cytokine anti-inflammatory cytokines (mainly, IL-4, IL-5, IL13) Tr1 cells deliver mainly IL-10 cytokine Th3 cells deliver mainly transforming growth factor beta (TGFb) TNF-a tumour necrosis factor-alpha Tregs regulatory T lymphocytes pTregs extrathymic or peripheral regulatory T lymphocytes tTregs thymic regulatory T lymphocytes

[16] at the Dr. Sing Sing Way laboratory showed that pregnancy imprints regulatory memory through the specific tolerogenic Tregs in the mother which sustain anergy to fetal antigens. In this review we discuss the role of Tregs in foeto-maternal immune tolerance and secondarily in recurrent miscarriage/implantation failure, as well as their potential use as a new target for infertility treatment. 2. A brief on regulatory T cells CD4þCD25þTregs are an important subset of T cells. Approximately 5e10% of peripheral CD4þ T cells in humans constitutively express CD25, the high-affinity a-chain of the interleukin 2 receptor [7,17]. Two main Treg subsets are now considered: those naturally occurring or thymic Tregs (tTregs) and those induced or extrathymic or peripheral Tregs (pTregs) (Table 2, Fig. 1) [18,19]. The thymus gives rise to CD4þCD25þFoxp3þ cells that function as Tregs. These cells suppress immune responses in a cell contact-dependent manner; however, the mechanism of suppression has not been fully established. In the periphery, some T cells are induced to become Tregs by antigen and either IL-10 or TGF-b cooperation. Tregs induced by IL-10 are CD4þCD25Foxp3- and are referred to as Tr1 cells. These cells suppress immune responses by secretion of IL10. Tregs induced by TGF-b are CD4þ/CD25þ/Foxp3þ and are referred to as induced Tregs. Finally, Some CD8þ cells can also be induced by antigens and IL-10 to become Tregs. These cells are CD8þ/Foxp3þ and suppress effector T cells by a cell contactdependent mechanism or by secretion of cytokines. The CD8þ/ Foxp3þ cells have been demonstrated in vitro, whether they exist in vivo remains unknown [18e20]. Paust and Cantor [21] showed a Foxp3 enhancer element, conserved noncoding sequence 1 (CNS1), to be essential for pTregs but not tTregs generation, thereby suggesting different types of biological properties. Tregs play a major role in preventing autoimmunity [21] and tolerating allogeneic organ grafts [22]. Tregs seems also play a key role maintaining the maternalefetal immune tolerance [7]. 3. Tregs during normal pregnancy Aluvihare [8] first showed in an animal model, the increase in this specific T cell subpopulation during normal pregnancy. Conversely, other authors [5] first observed an insufficient

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

243

Table 2 Subsets and function of natural and induced regulatory T cells.a Cell type

Subset markers

Regulatory mechanism

TFb

Target cells

Natural Tregs Peripheral Tregs Tr1cells Th3 cells pNKregs

CD4þCD25þ CD4þCD25 CD4þCD25þ CD4þCD25þ CD4CD25þCD56þ

Cell contact-dependent IL-10 Cytokines APCs IL-10 TGF-b IL-4, IL-10, TGF-b

Foxp3þ Foxp3 Foxp3 Foxp3þ/?c ?c

T cells

CD8þCD28

Fosp3þ?c

T cells T cells T cells

CD8þCD25 CD8þCD28þ Qa 1 þ restricted CD8þ

Cell contact-dependent Cytotoxixity, cytokines Induction ILT3/ILT4 (DCs) IL12; IL15; IL-21?c Cytokines: IL-15

T cells; APCs T and B cells T cells T cells T cells, APCs Tumor cells T cells

Foxp3þ Foxpþ?c STAT5? Foxp3

DCs/APCs T cells T/B cells

Abbreviations: APCs: antigen presenting cells; DCs: dendritic cells; Foxp3: forkhead box protein 3(scurfin); IL: interleukin; ILT: immunoglobulin transcript; pNKregs: peripheral regulatory natural killer cells; STAT5: Signal Transducers and Activators of Transcription protein 5; TCR: T cell receptor; Th3: T helper type 3; Tr1 cell: Tr1 regulatory cells; Treg: regulatory T cell; TGF-b: transforming growth factor b. a Subtypes detected in rodents and humans. b Transcription factor. c Issue uncertain; : suppression.

production of pregnancy-induced Tregs in abortion-prone mice. Sasaki et al. have reported an increased number of CD4þCD25þT cells in human deciduas during pregnancy [23]. In addition, Somerset et al. [4] demonstrated that the median percentage of Tregs was higher in pregnant women than in non-pregnant. The CD4þCD25þT cells increased throughout pregnancy and diminish in the late puerperium [4,21,24]. Somerset et al. [4] also revealed that CD4þCD25þ cells were highly enriched in Foxp3 and that these type of cells isolated from pregnant women are capable of suppressing thymidine uptake in allogenically stimulated lymphocytes by a 50%. Similarly, has also been reported an in vitro analysis showing an inhibitory capacity of murine CD4þCD25þT cells on maternal alloresponses [5]. In animal models, by using 2photon imaging, Teles et al. [25] have shown that Foxp3þ cells

accumulated in the mouse uterus during receptive phase of the oestrus cycle. In this way, they also showed that seminal fluid further fostered Tregs expansion. Homing of Tregs to the uterus seems to be mediated by CCR7. Finally, it have also been demonstrated that during the preimplantation period, factors in the seminal fluid deliver at coitus cause expansion of a CD4þCD25þputative Tregs population, with a further increase of Foxp3þ. Thus, Treg cells accumulate in the uterus prior to embryo implantation and seminal fluid is a key regulator of the uterine Tregs population [26]. In conclusion, CD4þCD25þFoxp3þ T cells are increased in normal pregnancies and seem to play a critical role in embryo implantation and in the maintenance of the maternal immune tolerance in front of the semi-allogeneic fetal antigens (Fig. 2).

Fig. 1. Differentiation and development of regulatory T cells. Natural CD4þCD25þ Tregs develop in the thymus as a result of positive selection between TCR and host antigens. The autoreactive T cells undergo negative selection and are depleted by apoptosis. The acquired Tregs develop in the periphery from naïve precursors, and their specificity lies in an antigen other than those that come into contact with the thymus. Tr1 cells are exposed to an antigen in the presence of IL-10. They are not generated by naïve CD4þ T cells but result from exposure to low doses of antigen via the oral route and secrete TGF-b. All in all, these cells confer a tolerogenic immune response whereas Th1, Th2 and Th17 cells mediate protective immunity. The tolerogenic properties of Tregs play a major role in maternal tolerance to paternally-derived fetal antigens, thereby favouring implantation and the whole pregnancy. IL: interleukin; þve: positive; TGF-b: transforming growth factor beta; Tregs: regulatory T-cells; Th3: T-lymphocyte producing and secreting TGF-b; Tr1: T-lymphocyte producing and secreting IL-10.

Fig. 2. Relationship between Tregs, cytokines and fetalematernal tolerance. Maternal tolerance to paternal allo-antigens expressed by the fetus is essential to avoid infertility, implantation failures, miscarriages or even late pregnancy complications such as pre-eclampsia. Adequate Tregs number and function act to suppress Th1/Th17maternal attack on the semi-allogeneic conceptus. Dendritic cells (DCs), and specially the so-called modified or tolerogenic DCs, help Tregs to develop this antigenspecific immune tolerance. Modified DCs, or tolerogenic DCs, are characterised by being relatively immature cells, with scant expression of CD80 and IL-12, high expression of TNF-a and G-CSF receptors and high capacity to synthesise indoleamine 2,3 dioxygenase. RIF: recurrent implantation failure; T-cell; Th1, Th17: helper type I, and interleukin 17-producing Th cell; Tregs: regulatory T cells.

244

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

4. Tregs and pregnancy failure 4.1. Tregs are diminished in pregnancy loss It has been demonstrated that Th1 specific-paternal-alloantigen cells are recruited at the fetalematernal interface of abortion-prone mice [5]. Those authors also compared the number of CD4þCD25þT cells in different tissues and deciduas from normal pregnant and abortion-prone mice and found an increase of Treg in normal pregnant mice when compared to abortion-prone mice. They concluded that in normal pregnancy, high levels of CD4þCD25þT cells can be generated in the thymus, but not in the abortion-prone animals. Similarly, levels of Foxp3 and neuropilin-1 are diminished in the placentas of these mice, suggesting low Tregs activity [5,7]. Jasper et al. [27] shown that the expression of Foxp3þ mRNA was reduced approximately two-fold in the tissue of infertile women, particularly in those experiencing repeated failed cycles of IVF. Overall, these studies suggest that the absence of Tregs can lead to pregnancy failure. Similar data were observed from human pregnancies; women with spontaneous abortion were found to have low levels of CD25þ cells within the CD4þ subpopulation [23]. Cytotoxic T-lymphocyte antigen 4 (CTLA4), also known as CD152, and a member of the immunoglobulin superfamily, is expressed on the surface of helper T cells and transmits an inhibitory signal to T cells. CTLA4 is similar to the T-cell co-stimulatory protein CD28, and both molecules bind to CD80 and CD86, also called B7-1 and B7-2, respectively, on antigen-presenting cells. CTLA4 transmits an inhibitory signal to T cells, whereas CD28 transmits a stimulatory signal [28]. Sasaki et al [23] hypothesised that CTLA-4 expressed on human placental Tregs can induce, in decidual and in peripheral dendritic cells, the expression of indoleamine 2,3-dioxygenase (IDO) by an interaction with its ligand CD80-CD86, especially during early phases of pregnancy. In turn, expression of IDO promotes maternalefetal tolerance. 4.2. Miscarriage can be prevented by transfer of Tregs from normal pregnant mice Based on the evidence that Tregs could inhibit the decidual effector cells, several authors have further demonstrated that the transfer of CD4þCD25þT cells of normal pregnant mice into 0e2days pregnant abortion-prone mice prevented spontaneous abortion, but not when Tregs were transferred from non-pregnant mice [5,29]. This situation is very similar to that observed in transplantation, where T regulatory cells are necessary before grafting to avoid rejection [30]. Thus, it seems that Tregs are necessary in situ to prevent early abortion, but only those Treg that have been previously exposed to paternal alloantigens, similar to the situation observed in transplantation [5].

implantation is also necessary for tolerating male antigens, perhaps due to immunomodulatory TGF-b present in seminal plasma [32]. In the same way, Sharma el al [33] and Chen et al. [34] have shown IDO expression by dendritic cells (DCs). In these cells, IDO confers the ability to directly activate resting Tregs for potent suppressor activity, and is implicated in converting CD4þCD25þFoxp3þ Tregs from CD4þCD25- T-cells, both in mice and in humans. However, the positive role played to IDO on pregnancy is still matter of discussion [10,35]. Haem oxygenase isoform (HO-1) represents inducible form of Haem oxygenase (HO) activated by Th1 cytokines, oxidative stress and hypoxia. HO-1 is an important player in maternalefetal tolerance. Abortion-prone mice express low HO-1 levels in placental tissues when compared to normal pregnant female [5]. Besides, Treg-foxp3þ are capable of markedly increasing the placental HO-1 levels [36]. Interestingly, the HO-1 up-regulation was related to increases of neuropilin-1, a Treg marker indicating a cross-link between both systems during pregnancy [37]. At last, HO-1 induction may lead to up-regulation of TGB-b, IL-10 and CTLA-4, suggesting that tolerogeneic effect of HO-1 might be CD4þCD25þ Tregs-dependent [38]. In the same context of modifications of maternalefetal microenvironment, normal leukaemia inhibitory factor (LIF) levels have been related to successful implantation. Animals lacking LIF gene fail to implant blastocysts into LIF-deficient uterus [39]. In the same way, it has found an important increase of local levels of LIF after Tregs therapy [5]. In the end, all these data suggest that Tregs at the maternalefetal interface are important players in avoiding fetal allo-rejection by means of the creation a “tolerant” microenvironment characterized by expression of IL-10, TGF-b, HO-1, IDO and LIF, rather than diminishing the Th1 cytokines at the fetalematernal place [7,31] Fig. 3. 4.4. Pregnancy-functional Tregs need to be stimulated by malespecific antigens As stated previously, some reports using genetic animal models demonstrated that, in vivo, alloantigens are required for optimal Tregs activity and function during pregnancy [5,24,36]. Thus, Tregs may act in an antigen-specific manner during pregnancy as suggested by the fact that Tregs from normal pregnant (CBA/JxBALB/c) but not non-pregnant females (CBA/J) could prevent abortion in the CBA/JxDBA/2 combination [5]. Hence, Tregs sensitisation from paternal antigens seems to be essential to achieve maternalefetal tolerance [16,40]. To confirm this hypothesis, Zenclussen et al [7,] vaccinated CBA/J females with BALB/c male antigens before mating took place. The abortion rates in CBA/J females matched with DBA/2J males could be strongly reduced by vaccination with BALB/c male splenocytes. Those authors confirmed that the transfer of BALB/c male splenocytes generated Tregs. Thus, in vivo alloantigen priming of Tregs before pregnancy is required for optimal Tregs activity during pregnancy.

4.3. Role of Tregs in creating fetalematernal tolerant microenvironment

4.5. Time of Tregs generation is important for pregnancy outcome

It seems Tregs may collaborate to achieve a state of privilege in the tissue microenvironment. It has been presumed associations between Tregs, some Th2 and Th3 cytokine production and enhanced IDO expression as a protective immune-mechanism [31]. IDO is an intracellular enzyme that mediates degradation of tryptophan to N-formyl-kynurenine. This hypothesis was studied in an animal model, finding an increase TGF-b protein levels, but not TGF-b mRNA after Treg therapy [7]. This dissociation could be explained because TGF-b is mostly post-transcriptionally regulated. As we will comment later, the immune activation necessary for

Exactly where and when male antigen presentation and consequent Tregs up-regulation take place remains unknown. Male antigens may migrate to the thymus and stimulate Tregs (nTregs), or paternal allo-antigens may stimulate peripheral CD4þCD25 T cells, converting them also into pTregs [5]. Peripheral CD4þCD25 T cells become suppressive if low antigen dose is present during chronic antigen exposure. Under these conditions, and with a previous induction by APCs, CD4þCD25 T cells are converted into CD4þCD25þ peripheral Treg [19]. The last two mechanisms seem to be crucial in semi-allogeneic pregnancy since naïve T cells may

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

245

Fig. 3. Regulatory T cells and maternalefetal immune-tolerance. After encountering paternally-derived fetal antigens in the periphery or in the genital tract, antigen specific Tregs are generated. This specific Treg population could expand later, as paternally-derived fetal alloantigens are continuously released to the periphery. Peripheral Tregs would then migrate into fetalematernal interface where they would aid to create a site of immune privilege characterized by high levels of protective molecules, ie., HO-1, LIF, IDO, TGFb, IL-10, PIBF. In turn, these pTregs are able to block maternal effector T cells, thereby reducing the maternalefetal pathologic responses to paternally-derived antigens. Furthermore, Tregs maintain adequate uNK cell subset preventing trophoblast injury and permits a sufficient angiogenesis. APC: antigen presenting cells; HO-1: Heme oxygenase isoform 1; IDO: indoleamine 2,3-dioxygenase; LIF: leukaemia inhibitor factor; PIBF: progesterone-induced blocking factor; pTregs: extrathymic or peripheral regulatory-T lymphocyte; tTregs: thymic regulatory-T lymphocytes; Tcyt: maternal T cytotoxic lymphocytes; uNK: uterine NK cells; TGF-b: transforming growth factor beta. : NKG2-CD94: lectin-like receptors (mainly in NK cells). They recognise HLA-E molecules. :KIR: killer-cell immunoglobulin-like receptor (mainly in NK cells). They recognise HLA-C molecules. : step blocking.

have the potential to be converted into cells with regulatory and/or immunosuppressive properties, thereby ensuring fetal maintenance. After formation of the placenta, maternal immune tolerance to semi-alloantigens may be maintained by the continuous release of fetal-paternal alloantigens. These antigens would be processed by APCs from the maternal immune system and induce a noninflammatory, non-dangerous state, as proposed for other in vivo models [41]. In summary, Tregs increased throughout pregnancy and diminished in late puerperium. Tregs in situ appear to be are essential to prevent early abortion, but only those Tregs that have been previously exposed to specific paternal alloantigens (see later on). 5. New data on tregs and maternalefetal tolerance Recent interesting advances have been made in the understanding of the role of Tregs in maternalefetal tolerance. Samstein et al. [42] hypothesised that pTregs developed throughout the evolution of the mammalian placenta to attenuate the maternale fetal response. In support of this theory, those authors showed CNS1 to be present in all 14 mammalian placentas analysed and absent in non-placental mammals and non-mammals. They postulated that the retrotransposition of CNS1 into the Foxp3 locus permitted the emergence of pTregs. In diverse animal experiments, those authors showed that pregnancy-induced maternal pTregs recognised paternal antigens. These pTregs are able to block maternal effector T cells by different suppressive mechanisms which most likely operate synergistically, thereby reducing the maternalefetal pathological responses to paternal antigens. In brief, Tregs can act as an IL-2 “sponge”, competitively depriving

adjacent T cells of autocrine IL-2, thereby impeding IL-2-dependent T-cell activation. Additionally, Tregs may suppress maternal effector T cells by altering paracrine cytokine production or by cell-to-cell contact directly between Tregs and effector T cells [43]. By contrast, absence of this kind of Tregs in the mother resulted in the infiltration of activated T cells into the placentas, leading to spontaneous abortions. A further work meriting comment is that reported by Rowe [16]. That group stressed the key role played by Tregs since even transient partial ablation triggers fetal-specific effector T-cell activation and pregnancy loss. They showed that pregnancy selectively stimulates the accumulation of maternal Foxp3þ CD4þCD25þ cells with fetal specificity using tetramerbased enrichment that allows the identification of rare endogenous T cells. Interestingly, after delivery, fetal-specific Tregs persist at elevated levels, maintain tolerance to pre-existing fetal antigen, and rapidly re-accumulate during subsequent pregnancy. The accelerated expansion of Treg cells during secondary pregnancy was driven almost exclusively by proliferation of fetal-specific Foxp3þ cells retained from prior pregnancy, whereas induced Foxp3 expression and proliferation of pre-existing Foxp3þ cells each contribute to Tregs expansion during primary pregnancy. Thus, pregnancy imprints Foxp3þCD4þ cells that sustain protective regulatory memory to fetal antigen. In the same way, researchers have found in animal models, high levels of Foxp3þ cells in vaginal secretions at very stages of normal pregnancy but not in abortionprone animals. These data support the hypothesis that there is local generation of Tregs during early normal pregnancy [24]. In this way, Kallikourdis et al., [44] stressed the role played by uterine Treg cells, and their capacity to express a highly suppressive CCR5þ, which represent the effector arm of regulatory T cells. Interestingly, the accumulation of CCR5þ regulatory T cells in the maternalefetal

246

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

interface appears to be enhanced by paternal alloantigens. In some mouse models, continuous but very low-level maternal “sensitisation” from paternal antigens through seminal proteins could be a non-sufficient stimulus to maintain allogeneic tolerance [45]. The Programmed Death-1 (PD1) receptor with its ligand (PDL1) is a negative co-stimulatory pathway that has been shown to play a major role in tolerance maintenance at the uteroplacental interface [46]. As stated above, naturally-occurring Tregs are characterised by the surface expression of CD4, CD25 and Foxp3. In addition, CTLA4, glucocorticoid-induced TNFR-related protein and PDL1 are also expressed in those cells. In an animal model, blocking PDL1 resulted in a loss of regulatory function and reduction in the fetal survival rate ([47]). Lack of PDL1 function seems to be able to favour Th1 and Th17 expansion, thereby directly affecting Th1/Th2/Th17 balance, leading to defective foeto-maternal tolerance [48]. Overall, these findings indicate that fetal-maternal-l alloimmune tolerance to paternal antigens is a dynamic and active process in which Tregs, and particularly pTregs, specifically respond to paternal antigens. 6. New therapeutic opportunities related to treg cells These findings also permit us to consider new strategies for improving pregnancy outcomes and novel approaches for therapeutically exploiting Tregs þ cell memory. Although a recently proposed way of managing recurrent miscarriages was via the use of biologic therapy (anti-TNF-a blockers) to mitigate the inflammatory maternalefetal response related to the Th1 cytokine profile, particularly TNF-a [49], the findings regarding Tregs led us to think that new therapies should aim not to suppress the maternalefetal immune system but rather to enhance maternal tolerance. The evidence previously commented also adds biologic plausibility to the paradoxical observation that, in women suffering from unexplained recurrent miscarriages or in ART-patients with repetitive implantation failure syndrome, the administration of recombinant human granulocyte colony-stimulating factors such as r-metHuGCSF or Filgrastim (NeupogenÒ), rHuGM-CSF or Molgramostin (LeucomaxÒ) and rHuG-CSF or Lenograstim (GranocyteÒ) dramatically improved fetal outcomes [50,51]. G-CSF is synthesised by mononuclear, endometrial, trophoblast and NK cells. In turn, G-CSF receptors exist in macrophages, endometrial, granulocyte and dendrititc cells [51]. G-CSF is induced by ovarian steroid hormones and constituents of seminal plasma. Insufficient G-CSF may impair optimal MHC class-II and class-Imediated indirect presentation of reproductive antigens. G-CSF signalling through the low-affinity G-CSF receptors in blastocytes is associated with an increase in glucose uptake and enhanced proliferation of blastomeres [52]. Finally, G-CSF also plays a major role in dendritic cell maturation. Conversely, these “tolerogenic” dendritic cells contribute actively to the selective Tregs maturation to male-derived antigens. As expected, G-CSF administration appears to be associated with an increase in Tregs and DCs [51]. Thus, it is possible that the effectiveness of those granulocyte-stimulating factors is related to certain killer-cell immunoglobulin-like receptors (KIR) expressed on NK cells [51]. It has been postulated that the absence of certain activating receptors, particularly 2DS1, 2DS3 and 3DS5, is accompanied by high implantation/pregnancy rates in ART patients with repetitive implantation failure syndrome [51]. Obviously, further studies are required to conclusively show the effectiveness of granulocyte colony-stimulating factor in this field. Taking these considerations into account, Tregs up-regulation could be hypothesized as a possible future therapeutic strategy in humans, leading to the design of vaccines acting on the specific suppressor Tregs that could mitigate maternal alloimmune fetal

rejection. To this end, paternal antigens obtained from sensitisating membrane vesicle extracts, seminal plasma proteins [53] or from peripheral blood mononuclear cells (PBMC) [54] have been used. Similarly, i.v immunogloubulin immunotherapy has also been tried [55,56]. In addition, given the action of progesterone on Tregs, it could be used to stimulate them to achieve positive modulation of the maternalefetal immunologic microenvironment [12]. Unfortunately, a recent meta-analysis showed that none of these therapeutic options were significantly effective in women suffering from unexplained recurrent miscarriages [57]; however, differences in the populations analysed do not permit definitive conclusions to be drawn. Focussing on the possible role of paternal antigen-inducing maternal-specific Tregs sensitisation, and methodological differences apart, the source and number of these antigens should be taken into account. Moreover, in view of better understanding of the mechanisms controlling induction of maternal tolerance in pregnancy, it seems likely that the administration of paternal antigens without suitable tolerance-inducing molecules might worsen Th1 or Th17-mediated immunity in some women, rather than induce Tregs [58]. Maternal immune sensitization from paternal antigens would provoke the expansion of pTregs. In turn, Tregs may migrate to the fetalematernal interface and contribute to creating a “tolerant” microenvironment, as previously mentioned. Therefore, paternally-derived allogeneic sensitization to maternal immune cells appears to provide a rationale for addressing these challenging situations, especially if a tolerance-inducing agent, i.e G-CSF is added [58]. Finally, several questions arise: what source of paternal antigens would be better for treating these women? When should treatment be started? How long should it be maintained? Similarly, how many mononuclear paternally-derived cells (or paternally-derived antigens) should be transferred to the mother? Although better explanations exist for unexplained recurrent miscarriages that open up new therapeutic opportunities, i.e., maternal sensitization by means of paternal mononuclear cells, future data resulting from laboratory and clinical studies will be welcome to improve our knowledge of this interesting and challenging topic.

7. Conclusion Since maternal allo-reactive lymphocytes are not fully depleted, and the Th1/Th2 cytokine balance alone cannot prevent miscarriage, new immune cells such as CD4þCD25þFoxp3þ, also known as regulatory T cells (Tregs), emerge as a cornerstone of maternale fetal tolerance. Peripheral regulatory T cells step in to regulate allo-reactive Th1/Th17 cells. Animal and human experiments showed Tregs cell number and/or function to be diminished in miscarriages. Tregs at the maternalefetal interface are able to impede fetal allorejection, thereby favouring a “tolerant” microenvironment characterised by the expression of diverse immune-modulating cytokines, such as IL-10, TGF-b and HO-1, rather than by lowering Th1 cytokines. In vivo experiments showed Tregs sensitisation from paternal antigens to be essential for maternalefetal tolerance. These findings also permit us to consider new strategies for improving pregnancy outcomes and novel approaches to therapeutically exploiting Tregs þ cell memory. The best source of paternal-derived antigens, the time and way to administer them and the effective dose to be injected, and its association with other Tregs modulating drugs, i.e. G-CSF cytokine family, are challenges that need to be addressed as soon as possible through welldesigned clinical studies.

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

Authors’ contributions to the manuscript JAR, contributed to 1) conception and design of this revision, 2) drafting the article and revising it critically, and 3) final approval of the version to be published. EL and JMG contributed equally to the conception and design of this revision, the critical revision and final approval of the version to be published.

[23]

[24]

[25]

Disclosure of interest

[26]

The author also state that they do not have any commercial, financial, or any other type of interest that may influence the drawing up and the results of this paper.

[27]

[28]

Acknowledgements [29]

The authors thank Ms. Christine O’Hara for reviewing and correcting the style and grammar of the manuscript.

[30]

References

[31]

[1] Williams Z. Inducing tolerance to pregnancy. N Engl J Med 2012;367:1159e 61. [2] Branch DW, Gibson M, Silver RM. Recurrent miscarriage. N Engl J Med 2010;363:1740e7. [3] Medawar PB. Some immunological and endocrinological problems raised by the evolution of viviparity in vertebrates. Symp Soc Exp Biol 1953;7:320e38. [4] Somerset DA, Zheng Y, Kilby MD, Sansom DM, Drayson MT. Normal human pregnancy is associated with an elevation in the immune suppressive CD25þCD4þ regulatory T-cell subset. Immunol 2004;112:38e43. [5] Zenclussen AC, Gerlof K, Zenclussen ML, Sollwedel A, Bertoja AZ, Ritter T, et al. Abnormal T cell reactivity against paternal antigens in spontaneous abortion: adoptive transfer of pregnancy of pregnancy-induced CD4þCD25þ T regulatory cells prevent foetal rejection in a murine abortion model. Am J Pathol 2005;166:811e22. [6] Porter TF, LaCoursiere Y, Scott JR. Immunotherapy for recurrent miscarriage. Cochrane database Syst Rev; 2006:CD000112. [7] Zenclussen AC. Regulatory T cells in pregnancy. Semin Immunol 2006;28:31e 9. [8] Aluvihare V, Kallikourdis M, Betz A. Regulatory T cells mediate maternal tolerance to the foetus. Nat Immunol 2004;3:266e71. [9] Tafuri A, Alferink J, Moller P, Hammerling G, Arnold B. T cell awareness of paternal alloantigens during pregnancy. Science 1995;270:630e3. [10] Nardi Fda S, Slowik R, Wowk PF, da Silva JS, Gelmini GF, Michelon TF, et al. Analysis of HLA-G polymorphisms in couples with implantation failure. Am J Reprod Immunol 2012;68:507e14. [11] Nermine L, Rouas-Freiss, Dausset J, Carpentier A, Carosella ED. Soluble HLA-G protrein secreted by allo-specific CD4þ T cells suppresses the alloproliferative response: a CD4þ T cell regulatory mechanism. Proc Nat Acad Sci 2001;98:12151e6. [12] Alijotas-Reig J. Immunological puzzle related to recurrent miscarriage: overview. Curr Immunol Rev 2009;5:175e86. [13] Raghupathy R. Th1 immunity is incompatible with successful pregnancy. Immunol Today 1997;18:478. [14] Zenclussen AC, Joachim R, Hagen E, Peiser C, Klapp BF, Arck PC. Heme oxygenase is downregulated in stress-triggered and interleukin-12-mediated murine abortion. Scand J Immunol 2002;55:560e9. [15] Chaouat G, Lédée-Bataille N, Zourbas S, Ostojic S, Dubanched S, Martal J, et al. Cytokines, implantation and early abortion: re-examining the Th1/Th2 paradigm leads to question the single pathway, single therapy concept. Am J Reprod Immunol 2003;50:177e86. [16] Rowe JH, Ertelt JM, Xin L, Way SS. Pregnancy imprints regulatory memory that sustains anergy to foetal antigen. Nature 2012;490:102e6. [17] Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science 2003;299:1057e61. [18] Beissert S, Schwarz A, Schwarz T. Regulatory T cells. J Invest Dermatol 2006;126:15e24. [19] Lohr J, Knoechel B, Abbas AK. Regulatory T cells in the periphery. Immunol Rev 2006;212:149e62. [20] Saito S, Shima T, Inada K, Nakashima A. Which types of regulatory T cells play important roles in implantation and pregnancy maintenance? Am J Reprod Immunol 2013;69:340e5. [21] Paust S, Cantor H. Regulatory T cells and autoimmune disease. Immunol Rev 2005;204:195e207. [22] Sakaguchi S, Sakaguchi N, Asano M, Itoh M, Toda M. Immunlogic selftolerance maintained by activated T cells expressing IL-2 receptor alpha-

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

[46]

[47]

[48]

247

chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol 1995;155:1151e64. Sasaki Y, Sakai M, Miyazaki S, Higuma S, Shiozaki A, Saito S. Decidual and peripheral blood CD4þCD25þ regulatory T cells in early pregnancy subjects and spontaneous abortion cases. Mol Hum Reprod 2004;10:347e53. Zenclussen ML, Thuere C, Ahmad N, Wafula PO, Fest S, Teles A, et al. The persistence of paternal antigens in the maternal body is involved in regulatory T-cell expansion and fetal-maternal tolerance in murine pregnancy. Am J Reprod Immunol 2010;63:200e8. Teles A, Schumacher A, Kühnle MC, Linze N, Thuere C, Reichardt P, et al. Control of uterine microenvironment by Foxp3þ cells facilitates embryo implantation. Front Immunol 2013;4:1e12. Guerin LR, Moldenhauer LM, Prins JR, Bromfield JJ, Hayball JD, Robertson SA. Seminal fluid regulates accumulation of FOXP3þ regulatory T cell in the preimplantation mouse uterus through expanding the FOXP3þ cell pool and CCL19-mediated recruitment. Reprod Biol 2011;85:397e408. Jasper MJ, Tremellen KP, Robertson SA. Primary unexplained infertility is associated with reduced expression of the T-regulatory cell transcription factor Foxp3 in endometrial tissue. Mol Hum Reprod 2006;12:301e8. Magistrelli G, Jeannin P, Herbault N, Benoit De Coignac A, Gauchat JF, Bonnefoy JY, et al. A soluble form of CTLA-4 generated by alternative splicing is expressed by non-stimulated human T cells. Eur J Immunol 1999;29:3596e602. Yin Y, Han X, Shi Q, Zhao Y, He Y. Adoptive transfer of CD4þCD25þ regulatory T cells for prevention and treatment of spontaneous abortion. Eur J Obstet Gynecol Reprod Biol 2012;161:177e81. Karim M, Kingsley CI, Bushell AR, Wood K. Alloantigen-induced CD25þCD4þ regulatory T cells can develop in vivo from CD25-CD4þ precursors in a thymus-independent process. J Immunol 2004;172:923e8. Waldmann H, Graca L, Cobbold S, Adams E, Tone M, Tone Y. Regualtory T cells and organ transplantation. Semin Immunol 2004;16:119e26. Remes Lenicov F, Rodriguez Rodrigues C, Sabatté J, Cabrini M, Jancic C, Ostrowski M, et al. Semen promotes the differentiation of tolerogenic dendritic cells. J Immunol 2012;189:4777e86. Sharma MD, Baban B, Chandler P, Hou DY, Singh N, Yagita H, et al. Plasmacytoid dendritic cells from mouse tumour-draining lymph nodes directly activate mature T-regs via indoleamine-2,3deoxyhenase. J Clin Invest 2007;117:2570e82. Chen W, Liang X, Petersen AJ, Muun DH, Blazan BR. The indoleamina 2,3dioxygenase pathway is essential for human plasmocytoid dendritic cellinduced adaptive T regulatory cell generation. J Immunol 2008;181:5396e 404. Miwa N, Hayakawa S, Miyazaki S, Myojo S, Sasaki Y, Sakai M, et al. IDO expression on decidual and peripheral blood dendritic cells and monocytes/ macrophages after treatment with CTLA-4 or interferon-gamma increase but decrease in spontaneous abortion. Mol Hum Reprod 2005;11:865e70. Choi BM, Pae HO, Jeong YR, Kim YM, Chung HT. Critical role of heme oxygenase-1 in Foxp3-mediated immune suppression. Biochem Biophys Res Commun 2005;327:1066e71. Sollwedel A, Bertoja AZ, Zenclussen ML, Gerlof K, Lisewski U, Wafula P, et al. Protection from abortion by HO-1 up-regulation is associated with increased levels of Bag-1 and neuropilin-1 at the foetal-meternal interface. J Immunol 2005;175:4875e85. Zenclussen AC, Fest S, Busse P, Joachim R, Klapp BF. Arck PC.Questioning the Th1/Th2 paradigm in reproduction: peripheral levels of IL-12 are downregulated in miscarriage patients. Am J Reprod Immunol 2002;48:245e51. Stewart CL, Kaspar P, Brunet LJ, Bhatt H, Gadi I, Köntgen F, et al. Blastocyst implantation depends on maternal expression of leukaemia inhibitory factor. Nature 1992;359:76e9. Schumacher A, Wollenberg I, Bertoja AZ, Sollwedel A, Zenclussen AC. Pregnancy-protective Treg are antigen-specific. Am J Reprod Immunol 2005;54 (abstract):147. Schumacher A, Wafula PO, Bertoja AZ, Sollwedel A, Thuere C, Wollenberg I, et al. Mechanisms of action of regulatory T cells specific for paternal antigens during pregnancy. Obstet Gynecol 2007;110:1137e45. Samstein RM, Josefowicz SZ, Arvey A, Treuting PM, Rudensky AY. Extrathtymic generation of regulatory T cells in placental mammals mitigates maternal-fetal conflict. Cell 2012;150:29e38. Mjöberg J, Berg G, Ernerudh J, Ekerfelt C. CD4þCD25þ regulatory T cells in human pregnancy: development of a Treg-MLC-ELIOSPOT suppression assay and indications of paternal specific Tregs. Immunology 2007;120:456e66. Kallikourdis M, Anderson KG, Welch KA, Betz AG. Alloantigen-enhaced accumulation of CCR5þ “effector” regulatory T cells in the gravid uterus. Proc Nat Acad Sci 2007;104:594e9. Seavey MM, Mosmann T. Paternal antigen-bearing cells transferrd during insemination do not stimulate antipaternal CD8þ T cells: role of estradiol in locally inhibiting CD8þ T cells responses. J Immunol 2006;177:7567e78. Taglauer ES, Yankee TM, Petroff MG. Maternal PD-1 regulates accumulation of fetal antigen-specific CD8þ T cells in pregnancy. J Reprod Immunol 2009;80: 12e21. Habicht A, Dada S, Jurewicz M, Fife BT, Yagita H, Azuma M, et al. A link between PDL1 and T regulatory cells in fetomaternal tolerance. J Immunol 2007;179:5211e9. D’Addio F, Riella LV, Mfarreg BG, Chabtini L, Adams LT, Yeung M, et al. The link between the PDL1 costimulatory pathway and TH17 in fetamaternal tolerance. J Immunol 2011;187:4530e41.

248

J. Alijotas-Reig et al. / Placenta 35 (2014) 241e248

[49] Winger EE, Reed JL. Treatment with tumor necrosis factor inhibitors and intravenous immunoglobulin improves live births rates in women with recurrent spontaneous abortion. Am J Reprod Immunol 2008;60:8e16. [50] Scarpellini F, Sbracia M. Use of granulocyte colony-stimulating factor for the treatment of unexplained recurrent miscarriage: a randomized controlled trial. Humanit Reprod 2009;24:2703e8. [51] Würfel W, Santjohanser C, Hirv K, Bühl M, Meri O, Laubert I, et al. High pregnancy rates with administration of granulocyte colony-stimulating factor in ART-patients with repetitive implantation failure and lacking killer-cell immunoglobulin-like receptors. Humanit Reprod 2010;25:2151e3. [52] Robertson SA, Sjöblom C, Jasper MJ, Norman RJ, Seamark RF. Granulocyte-macrophage colony-stimulating factor promotes glucose transport and blastomere viability in murine preimplantation embryos. Biol Reprod 2001;64:1206e15. [53] Stern JJ, Coulam CB. Current status of immunologic recurrent pregnancy loss. Curr Opin Obstet Gynecol 1993;5:252e9.

[54] Taylor C, Faulk WP. Prevention of recurrent abortion with leukocyte transfusions. Lancet 1981;2:68e70. [55] Mueller-Eckhardt G, Heine O, Neppert J, Kunzel W, Mueller-Eckhardt C. Prevention of recurrent spontaneous abortion by intravenous immunoglobulin. Vox Sang 1989;56:151e544. [56] Christensen OB, Pedersen B, Rosgaard A, Husth M. A randomized, doubleblind, placebo-controlled trial of intravenous immunoglobulin I the prevention of recurrent miscarriage: evidence for a therapeutic effect in women with secondary recurrent miscarriage. Humanit Reprod 2002;17: 809e16. [57] Scott JR. Imunotherapy for recurrent miscarriage. Cochrane database Syst Rev; 2006:CD000112. [58] Guerin LR, Prins JR, Robertson SA. Regulastory T-cell and immune tolerance in pregnancy: a new target for infertility treatment? Hum Reprod Update 2009;15:517e35.