ORIGINAL ARTICLE

The Persistence of Paternal Antigens in the Maternal Body is Involved in Regulatory T-Cell Expansion and Fetal-Maternal Tolerance in Murine Pregnancy
Maria Laura Zenclussen1*, Catharina Thuere1*, Nadja Ahmad1, Paul O. Wafula1, Stefan Fest2, Ana Teles1,3, Anne Leber1,3, Pablo A. Casalis1, Ingo Bechmann4, Josef Priller5, Hans-Dieter Volk1, Ana Claudia Zenclussen1,3
1 2

Reproductive Immunology Group, Institute of Medical Immunology, Charite, Universitatsmedizin, Berlin, Germany; Paediatric Immunotherapies Group, Paediatrics, Medical Faculty, Otto-von-Guericke University, Magdeburg, Germany; 3 Experimental Obstetrics & Gynaecology, Medical Faculty, Otto-von-Guericke University, Magdeburg, Germany; 4 Institute of Anatomy, University of Leipzig, Leipzig, Germany; 5 Laboratory of Molecular Psychiatry and Department of Experimental Neurology, Charite, Universitatsmedizin, Berlin, Germany

Keywords Fetal-maternal interface, male antigens, reproductive immunology, tolerance, Treg cells Correspondence Ana Claudia Zenclussen, Experimental Obstetrics and Gynecology, Medical Faculty, Otto-von-Guericke University, Magdeburg, Gerhard-Hauptmann-Str. 35, 39108, Magdeburg, Germany. E-mail: ana.zenclussen@med.ovgu.de *These authors contributed equally to this work. Submitted November 5, 2009; accepted November 5, 2009. Citation Zenclussen ML, Thuere C, Ahmad N, Wafula PO, Fest S, Teles A, Leber A, Casalis PA, Bechmann I, Priller J, Volk H-D, Zenclussen AC. 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: 200208 doi:10.1111/j.1600-0897.2009.00793.x

Problem Mammalian pregnancy is a state of immunological tolerance and CD4+ CD25+ regulatory T cells (Treg) contribute to its maintenance. Knowing that Treg act in an antigen-specic way during pregnancy, we hypothesized that they are generated after maternal immune cells encounter paternal antigens. Method of study We mated wild type females with transgenic green uorescent protein (GFP) males in an allogenic setting and killed them on different days of pregnancy. Results Presence of paternal and maternal MHC class II+ cells in vaginal lavage on day 0.5 of pregnancy was conrmed. Thus, antigen presentation may take place early during pregnancy in the periphery either by the direct or indirect pathways. Foxp3+ cells known to have regulatory activity could be detected on day 2 of pregnancy in lymph nodes and shortly after implantation at the fetal-maternal interface. Conclusion Our data suggest that paternal antigens are processed early during pregnancy, which leads to the generation of Treg. The continuous release of placental antigens into the maternal circulation allows the maintenance of a Treg population which is specic for paternal antigens and mediates tolerance toward the semi-allogeneic fetus until the time point of birth.

Introduction Tolerance of paternal alloantigens without affecting anti-infectious immune responsiveness is the key for a successful pregnancy. However, mechanisms associated with natural tolerance during pregnancy are
200

far from being well-known. Somerset et al. proposed regulatory T cells (Treg) as crucial players during human gestation.1 Depletion of CD25+ cells negatively affected murine pregnancy outcome.2 Accordingly, we recently demonstrated the important consequences of a decient Treg activity during
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

EARLY TOLERANCE TO PATERNAL ANTIGENS

pregnancy, namely immunological spontaneous abortion.3 It is known that Treg expand during pregnancy.1,2,4,5 We have also successfully shown that Treg are specic for paternal antigens5 and that tolerance can be transferred by injecting abortionprone mice with antigen-specic Treg.5,6 However, the events leading to the generation of antigen-specic Treg, especially how, when and at which places paternal antigens are recognized by maternal immune cells is not clear yet and this article should contribute to this knowledge. By mating wild type females with GFP+ males (homozygote), we show that paternal antigens are present in vaginal lumen immediately after fecundation. Besides, cells expressing paternal antigens are found in different immune and non-immune organs already at very early pregnancy stages. Our data suggest that the rst site of antigen presentation would be the vaginal mucus, after which Treg are generated in lymph nodes, where an augmentation in their number can be documented beginning from day 2 of pregnancy. Treg are present at the fetal-maternal interface shortly after implantation. The continuous release of placental antigens into the maternal circulation would allow the maintenance of a Treg population which is specic for paternal antigens and mediates tolerance toward the semi-allogeneic fetus until the time point of birth. Materials and methods Animals and Pregnancy Outcome Male DBA 2J (H2d) and female CBA J (H2k) mice were purchased from Charles River, Sulzfeld, Germany. C57 BL6 (H2b) males as well as BALB c females and males (H2d) were acquired from Harlan Winkelmann (Borchen, Germany). Vasectomized BALB c males were also obtained from Charles River, Germany. Homozygote C57 BL6 mice expressing GFP under the inuence of the betaactin promoter7 were maintained in our facility. All animals were housed in a barrier facility and received food and water ad libitum. Animal care and experimental procedures were followed according to institutional guidelines and conformed to the requirement of the state authority for animal research conduct (LaGeSo, Berlin, 0070 03). Twomonth-old females were paired with 2- to 4-month-old males, checked twice a day for vaginal plugs and separated from the males if mated. The
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

day of the vaginal plug was considered as day 0 of pregnancy. On different days of pregnancy, the females were killed, the uteri removed and the implantation sites were documented when appropriated. For conrming the presence of paternal fetal antigens in maternal organs throughout pregnancy, BALB c females (Harlan Winkelmann) were paired with GFP-transgenic males expressing GFP under the control of the beta-actin promoter (C57BL 6 background).7 Maternal organs (spleen, thymus, lungs, and liver) as well as other immunologically relevant material (lymph nodes, vaginal lavage, blood, and decidua) were collected from pregnant females on day 0, 2, 4, 6, 8, 10, 12, and 14 of pregnancy (n = 5 animals time point). Samples were frozen, processed for ow cytometry or kept overnight in paraformaldehyde (PFA) for microscopical analysis of GFP. Samples Collection Blood was obtained by cardiac puncture in dead animals and plasma was separated by centrifugation. Mucus samples were collected as described elsewhere.8 The uterine horns were opened longitudinally and the feto-placental unit separated from the uterine implantation sites. Up to day 5 the whole uterus was taken. Otherwise, the whole placental and decidual unit was separated individually from the respective embryo and its implantation site. Further, placenta and decidua were carefully separated from each other for ow cytometry and molecular biology studies. Spleen, lung, liver, lymph nodes, and thymus samples were kept in RPMI at 4C until processing for ow cytometry. For ow cytometry, decidual samples were cut in small pieces and collected in HBSS containing no Ca2+ and no Mg2+ (Sigma, Taufkirchen, Germany). For RNA isolation or Western Blot, placental and decidual tissues were carefully washed with cold sterile phosphate-buffered saline (PBS) pH = 7.40, snap frozen, and kept in )80C till use. Flow Cytometry We isolated mononuclear cells from spleen, lung, liver, lymph nodes, thymus or decidua as described before.3 Thereafter, the cells were stained using our standard protocol3 and read on a FACScan ow cytometer from Becton Dickinson, Heidelberg, Germany. Mucus samples were obtained rinsing the
201

ZENCLUSSEN ET AL.

vagina several times with 50 lL of sterile PBS. For the staining of extracellular markers, samples were stained with the respective antibodies for 10 min at 4C following our standard protocol. The following Abs were used: uorescein isothiocyanate (FITC)conjugated rat anti-mouse CD11c, and anti-major histocompatibility complex (MHC) class II. FITCCy5-or PE-conjugated rat IgG1 or Ig2b were used as negative control in separate tubes. All these antibodies were purchased from Becton Dickinson. For intracellular FoxP3 staining, a PE-labeled anti-mouse FoxP3 antibody was used (eBioscience, Kranenburg, Germany), and the staining was performed for 30 min at 4C following manufacturers instructions. For gating and or quadrant setting, when adequate, samples from either wild type or GFP-reporter mice served as controls. For antibody staining, respective isotype controls were included. GFP Detection by Flow Cytometry, Microscopy and Quantitative PCR We checked GFP protein expression during pregnancy (days 0, 2, 4, 6, 8, 10, 12 and 14, n = 5 day) in maternal organs by ow cytometry, uorescence microscopy and GFP DNA expression by quantitative PCR. For ow cytometry, cells were isolated from the different organs and xed in PFA 1% overnight at 4C before measurement. Pregnant BALB c females served to dene negative controls in the cytometer while positive controls were cells from pregnant GFP transgenic C57 BL6 females. For uorescence microscopy, PFA-xed samples were cut (vibrotome) and analyzed under uorescence microscope for the presence of GFP+ cells. Organs from GFP transgenic females served as positive controls. We also analyzed the origin of paternal cells antigens by investigating their MHC class II and CD11c expression by ow cytometry using our standard protocol. The antibodies employed were purchased from BD Pharmingen (San Diego, CA, USA). For semi-quantitative PCR, DNA was isolated using the E.Z.N.A. Tissue DNA Mini Kit (Peqlab, Erlangen, Germany), following manufacturers instructions. Amplications reactions (25 lL) were performed by mixing 5 lL DNA, 12,5 lL Master mix (Stratagene, Heidelberg, Germany) containing PCR Buffer, dNTPs, MgCl2, Ampli-Taq DNA Polymerase and SYBR-Green, 6 lL of the primer mix, 1 lL uorescein and 0.5 lL water. PCR reaction was performed as follows: an initial denaturation step of
202

3 min at 95C, followed by 30 s at 95C and 30 s at the appropriate annealing temperature (60C), followed by 30 s at 72C for 45 cycles. Apolipoprotein b (Apob) was used as housekeeping gene as described elsewhere.3 Sequences of primers are available upon request. Data Analysis and Statistics After conrming a non-normal distribution of the samples, all data are presented as medians, in box plots. Data for GFP DNA amplication by PCR are shown as medians (n = 5 for each time point). Please note that box plots show 75%, minimum and maximum and not standard deviations. Outliers and extreme values are included in the statistics and shown as stars or circles. Analysis of the differences between all groups was performed using the nonparametric KruskalWallis test. The MannWhitney U-test was applied for analyzing differences between two particular groups. In all cases, P < 0.05 was considered a statistically signicant difference. Results Levels of Treg are Augmented on Day 2 of Pregnancy in Lymph Nodes and Blood from Normal Pregnant Animals but not in AbortionProne Mice From our previous investigations, we knew that normal pregnant animals presented elevated Treg levels when compared with abortion-prone animals on day 14 of pregnancy.3 Besides, augmented Treg levels were observed by other authors and us very early in pregnancy also in other mating combinations.2,4 To conrm the expansion of Treg in early pregnancy, we determined the percentage of CD4+ Foxp3+ in lymph nodes from virgin, normal pregnant and abortion-prone animals on day 2 of pregnancy. We observed that the levels of CD4+ Foxp3+ cells (Treg) in the lymph nodes from the CBA J BALB c normal pregnancy combination went up beginning as early as day 2 of pregnancy (Fig. 1) and remained high as long as analyzed (last time point: day 12 of pregnancy, data not shown). Notably, CBA J DBA 2J abortion-prone mice presented diminished levels of Foxp3+ cells in blood and lymph nodes when compared with normal pregnant animals at all time points (Fig. 1 and data not shown). This revalidates the data we already reported.4 As we
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

EARLY TOLERANCE TO PATERNAL ANTIGENS

7 % CD4+ FoxP3+ cells in LN 6 5 4 3 2 1 0 Virgin *

*
CBA/J CBA/J x BALB/c CBA/J x DBA/2J

(a) 45 40 35 30 25 20 15 10 5 0
% GFP+ cells

Vaginal lavage Lymph nodes

0.5

6 8 10 Days of pregnancy

12

14

(b)

(c)

Day 2 of pregnancy

GFP DNA in spleen (2Ct)

Fig. 1 Treg expand very early in normal pregnancy. Figure shows the levels of CD4+ foxp3+ cells in lymph nodes of normal pregnant or abortion-prone mice on day 2 of pregnancy when compared with virgin mice. Please note that box plots show 75%, minimum and maximum and not standard deviations. Outliers and extreme values are included in the statistics and shown as stars or circles. Statistical analysis was carried out using KruskalWallis test followed by MannWhitney U-test. *P < 0.05.

(d)
4x104 3x104 2x104 1x104 0 2 4 6 8 10 Days of pregnancy 12 14

conrmed a very early raise of Treg in lymph nodes and knowing that they are specic for paternal antigens,5 we wondered whether these cells are generated earlier in response to paternal antigens. Detection of Paternal Fetal Cells and Antigens in the Maternal Body Throughout Pregnancy Knowing that Treg are present at such an early time point of pregnancy and knowing that they act in antigen-specic fashion5 suggests the possibility that paternal antigens are released to the periphery immediately after copulation. The migration of cells from the fetus to the mother is well documented, being microchimerism a common feature during human and murine pregnancy.914 To elucidate to what extend and in which period paternal antigens are present in maternal organs in mice and most important, to know to which extent and how paternal cells are recognized by maternal immune cells, we moved to another model and mated BALB c females with transgenic males expressing GFP under the control of the b-actin promoter (homozygous, C57BL 6 background).7 We then analyzed the percentage of GFP+ cells and checked the GFP DNA expression during pregnancy (days 0, 2, 4, 6, 8, 10, 12 and 14) in maternal organs (n = 5 mice for each time point). GFP+ cells could be detected in lymph
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

Fig. 2 Paternal fetal antigens can be detected throughout the whole pregnancy in maternal organs (a) shows the distribution of GFP+ cells analyzed by ow cytometry in vaginal lavage and lymph nodes during pregnancy. The data are represented as % of GFP+ cells (medians) as analyzed in the total cell population (without a gate). BALB c mice previously mated with C57 BL6 negative for GFP served as a negative control while GFP transgenic C57 BL6 females mated with GFP transgenic males served as positive controls for setting the gates. (b) depicts GFP+ cells found in lymph node of a mouse on day 10 of pregnancy (pointed out by arrows), a lymph node from a GFP+ animal (c) served as positive control. (d) shows the expression of GFP DNA found in spleen samples of BALB c females previously mated with GFP+ males. Data are presented as medians (n = 5 for each time point).

nodes by both, ow cytometry and uorescence microscopy (Fig. 2a,b respectively) with a lymph node from a GFP+ animal serving as positive control (Fig. 2c). GFP+ cells could also be conrmed in spleen (Fig. 2d) and blood as well as in various organs (Table I) of mothers beginning on day 0.5 by ow cytometry and uorescence microscopy. This conrms the presence of paternal antigens early during pregnancy in maternal tissues and may explain
203

ZENCLUSSEN ET AL.

Table I Number of Organs with Presence of GFP+ Cells Under Fluorescence Microscope Day of pregnancy 0.5 2 4 6 8 10 12 14

Spleen 03 03 03 03 03 13 03 03

Decidua 13 13 03 03 14 15 03 08

Liver 02 03 03 04 13 03 03 15

Lung 02 03 13 03 03 03 03 15

Heart 02 03 03 04 05 04 04 19

Placenta 13 23 23 33

Kidney 02 03 03 13 03 13 03 14

Lymph nodes 12 04 23 13 03 33 13 35

GFP+ cells were found in different maternal organs at all stages of murine pregnancy. The rst number indicates the number of samples where GFP+ cells were found with respect to the number of samples analyzed (second number) for each tissue and time of pregnancy.

why Treg, known to act antigen-specically, also begin to expand on day 2 of pregnancy. Our data further conrm recently published data about the presence of paternal antigens in maternal organs throughout all stages of pregnancy in mouse.13 Very interestingly, high numbers of GFP+ cells were detected in vaginal mucus at even earlier pregnancy stages (day 0.5 and day 2, continuing throughout pregnancy) (Fig. 2a). Especially the novel nding of paternal antigens in vaginal mucus on day 0.5 and 2 emphasizes the possibility of local antigen presentation at very early stages. Paternal antigen presentation may take place either by the direct or indirect pathway. In direct pathway paternal antigen presenting cells (APCs) can be recognized by maternal T cells while in the indirect presentation stimulation of APCs of the mother by paternal proteins presented by self-MHC antigens takes place. The fact that maternal APCs are present in this uid and we could also observed MHC-II+ of paternal origins (Fig. 3a,b) strongly suggests paternal alloantigen presentation already at this very early stage which can take place by both direct and indirect antigen presentation as it will be discussed in the next section. Paternal Antigen Presentation Pathways During Pregnancy We could conrm the presence of cells and antigens of paternal (days 05) as well as fetal (day 5 until birth) antigens of paternal origin in the maternal body as to the presence of GFP. Moreover, the generation of Treg with antigen specicity3,5 conrms that paternal antigens are being presented to the maternal immune system. The antigen presentation
204

via the direct pathway is possible15 as it has been suggested that exposure to sperm is capable of initiating a protective immune response. We aimed to further characterize the nature of GFP+ cells present in vaginal lavage by ow cytometry. Our data show that some of the GFP+ cells were also MHC II+ (Fig. 3a), and some of the MHC class II+ cells were also positive for CD11c (Fig. 3b). Therefore, paternal APCs may be directly recognized by maternal T cells. Interestingly, cells with DC morphology could be found in other maternal organs as lungs (Fig. 3c). Fig. 3d shows a lung slide from a GFP+ mother, which served as control. Alternatively, the indirect pathway might be initiated after self APCs process paternal antigens present in seminal plasma.16 One can also speculate that the observed GFP+ DCs are of maternal origin which recently engulfed GFP+ paternal cells. Intriguingly, Foxp3+ cells are not present within the lymphocyte population in vaginal lavage at days 0.5 or day of pregnancy as analyzed by ow cytometry. The presence of paternal antigens and maternal as well as paternal APCs accompanied by the absence of Foxp3+ cells in vaginal mucus immediately after fecundation suggests that maternal or paternal APCs which have participated in antigen recognition migrate to lymph nodes, where Treg are rst observed on day 2 of pregnancy. Treg Expansion as Result of Continuous Presence of Paternal Cells in Maternal Body? As to the maintenance of the Treg population during pregnancy, we have previously shown that hormones can not count for this effect as similar levels of progesterone, estradiol and estrone were found in
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

EARLY TOLERANCE TO PATERNAL ANTIGENS

(a) 104
MHC class II PE
103 102 101 100

3.97%

8.06%

(b)104
MHC class II PE
103 102 101

8.22%

3.81%

transient expansion of Treg in lymph nodes during pregnancy.4 Discussion

6.46%

100 101

102 103 GFP

104

100 100 101 102 103 104 CD11c APC

0.31%

(c)

(d)

(e)

Fig. 3 MHC class II+ cells from maternal and paternal origin can be found in the mother. (a) depicts a representative dot plot of a vaginal mucus sample from a pregnant mice on day 4 in which it is possible to see that some of the GFP+ cells are MHC class II+, conrming that APCs present in vaginal lavage are from both origin, maternal and paternal (b) depicts a representative dot plot of a vaginal mucus sample from a pregnant mice on day 4, where it is possible to see that some of the MHC class II+ cells are also positive for CD11c (c) shows 2 GFP+ cells (indicated with arrows) with DC morphology in lung tissue from a 14-day pregnant BALB c mice mated with a GFP male (d) shows a representative picture taken from a lung of a GFP female which served as positive control (e) depicts a representative picture of a placenta showing that the outer placental cells are GFP+.

normal pregnant and abortion-prone animals despite their different pregnancy outcome5 and Treg levels.4 It is known that, after implantation, the placenta continuously releases allo-antigens17 which can be processed by APCs from the maternal immune system under non-inammatory conditions. This would promote the expansion of Treg, which play a crucial role for the maintenance of a successful pregnancy. In our experiment, some of the GFP+ cells detected in the mother can, in fact, be originally from the placenta, as we could see that the outer cells from the placenta (the ones expected to migrate) are green (Fig. 3e). The augmentation of paternal cells in the maternal organs is in accordance with the
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

The mechanisms leading to tolerance toward the semiallogenic fetus are still unknown. Treg have been proposed to be crucial players in allowing this foreign tissue, the fetus, to survive within the maternal uterus, which contains immune cells with the potential to destroy it.1,2 It is widely shown that Treg augment in number during normal pregnancy.15 We could previously show that even at early pregnancy stages Treg can be observed in high numbers when compared with virgin specimens in pregnancy combinations with normal outcome, like CBA J BALB c or CBA J C57 BL6.4,5 An expansion of the Treg pool, rather than the amount of Treg before gestation, correlated with a good pregnancy outcome.4 Furthermore, we could show that Treg are specic for paternal antigens in a murine model5 as it is also reported for humans.18 It is, however, unknown how and at which time point(s) Treg are generated and if this in fact occurs after encountering paternal or fetal antigens locally or at the periphery. As Treg are augmented very early during murine pregnancy in lymph nodes,2,4 it is expected that they are either generated earlier and migrate to the lymph nodes around day 2 or that they are generated in the lymph nodes themselves. Several studies have failed to identify IL-2 and IL-2-R at the fetalmaternal interface.19,20 Moreover, the administration of IL-2 or IL-2-R is related to miscarriage.21 This may rule out the possibility that Treg are generated in uterus because of their dependence on IL-2. Intriguingly, the early administration of a mAb against the IL-2-R, CD25, results in implantation failure,3 suggesting by the diminution in the number of Treg due to blockage in the CD25 system, that Treg are necessary for implantation, thus very early. Therefore, we wondered whether Treg could be found in vaginal mucus immediately after copulation. We failed to detect Foxp3+ cells in vaginal mucus, reason as to why we postulated a possible scenario in which Treg are being generated in lymph nodes after APCs migrate there. The presence of immune cells with regulatory function at early time points and the fact that Treg act in an antigen-specic fashion had led us to hypothesize that paternal antigens are released to the periphery immediately after copulation. The bi-directional transfer of cells
205

ZENCLUSSEN ET AL.

between mother and fetus is now well documented.1113 Accordingly, fetal cells were found in maternal mouse brain 4 weeks post-partum.14 A very interesting yet unknown question is which kind of antigen is being released to the periphery at this early time point and how this antigen is being presented to the maternal immune system. By mating BALB c females with GFP+ transgenic males, we could conrm the presence of paternal antigens in the maternal body throughout all stages of pregnancy. Especially the nding of paternal antigens in vaginal mucus on day 0.5 and 2 emphasizes the possibility of local antigen presentation at very early stages. The antigen presentation via the direct pathway is possible as seminal uid mimics class II MHC proteins of other lymphocytes.15 Moreover, sperm antibodies specically precipitate antibodies against class II MHC mimics.16 These data additionally conrm that exposure to whole sperm components is capable of initiating a (protective) immune response. An earlier study in pigs suggests that seminal plasma from vasectomized boars deposited in the uterus can activate cells in the local draining lymph nodes, demonstrating that the immunological changes occurring in uterus are induced in vivo by seminal plasma.22 In the human seminal plasma, the presence of a glycoprotein capable of binding to CD4 may also be relevant to the modulation of maternal immunity at insemination.23 Alternatively, the indirect pathway might be initiated after self APCs process paternal antigens present in seminal plasma.16 Thus, exposure to sperm initiates a protective immune response. This was recently conrmed once more by an elegant study by Robertson et al. who showed that vasectomized male could not induce an augmentation of Treg in the paraaortical uterine draining lymph nodes in females after copulation as it was seen in combinations with nonvasectomized males.24 This is in line with our data. The observed changes took place, as we also described in our study, before fetus implantation. In the Robertson study, it was furthermore demonstrated that even when conception was inhibited by surgically ligating the uterine oviductal junctions prior to mating with intact males, the females showed an increased tolerance to male antigens.24 The fact that in our study some of the GFP+ paternal cells found in maternal organs were MHC class II+, and some of them expressed CD11c+ suggests that paternal APCs are directly recognized by maternal T cells and that indirect presentation is also possible.
206

The presence of Treg specic for paternal antigens in lymphoid organs has been conrmed during pregnancy. Treg themselves are able to induce CD4+ CD25) cells to become suppressor cells in the periphery during the so-called infectious tolerance,25 which may involve cellcell contact26 or operate through an indirect route, i.e. through the generation of a suppressive local microenvironment.27,28 This mechanism is extremely important in the context of a semi-allogenic pregnancy, as nave allo-specic cells would have the potential to be converted to cells with regulatory properties and re-enforce the tolerant state, allowing pregnancy maintenance. After placenta formation (in mice beginning on day 5), this tolerant tissue continuously releases allo-antigens16,29 which would be processed by APCs from the maternal immune system under noninammatory, non-dangerous circumstances as proposed for other in vivo models.27 This danger-free antigen presentation would provoke the expansion of Treg and enable them to become an important population as long as the pregnancy status lasts. Once the antigen disappears (at birth), Treg diminish in number and consequently, their effects disappear which would explain the non-acceptance of paternal grafts after pregnancy.30 In conclusion, our study provides evidences that paternal antigens can be detected by maternal immune cells very early during pregnancy and supports the antigen-specic expansion of Treg. Our data further give clues about how the maternal immune system is aware of the presence of these antigens and yet does not reject them. Acknowledgments The authors are very grateful to Brigitte Wegner and Dr. Werner Hopfenmuller for her advice in statistics and Jacqueline Mahlo for her assistance with the vibratome. The authors also thank Varuna Aluvi hare, Peter M. Johnson, Ignacio Anegon and Christiane Kling for their critical reading of the manuscript and helpful suggestions. The present work was supported by grants from the DFG (ZE526 4-1, ZE526 4-2) and the Charite (UFF 2003-109) to ACZ. POW is a supported by a PhD grant from the DAAD and AL by a PhD grant from the Medical Faculty of Otto-von-Guericke University Magdeburg. JP is a member of the SFB507 A5 (DFG).
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

EARLY TOLERANCE TO PATERNAL ANTIGENS

References
1 Somerset DA, Zheng Y, Kilby MD, Sansom DM, Drayson MT: Normal human pregnancy is associated with an elevation in the immune suppressive CD25+ CD4+ regulatory T-cell subset. Immunology 2004; 112:3843. 2 Aluvihare VR, Kallikourdis M, Betz AG: Regulatory T cells mediate maternal tolerance to the fetus. Nat Immun 2004; 5:266271. 3 Zenclussen AC, Gerlof K, Zenclussen ML, Sollwedel A, Bertoja AZ, Ritter T, Kotsch K, Leber J, Volk HD: Abnormal T-cell reactivity against paternal antigens in spontaneous abortion: adoptive transfer of pregnancyinduced CD4+ CD25+ T regulatory cells prevents fetal rejection in a murine abortion model. Am J Pathol 2005; 166:811822. 4 Thuere C, Zenclussen ML, Schumacher A, Langwisch S, Schulte-Wrede U, Teles A, Paeschke S, Volk HD, Zenclussen ML: Kinetics of regulatory T cells during murine pregnancy. Am J Reprod Immunol 2007; 58:514523. 5 Schumacher A, Wafula PO, Bertoja AZ, Sollwedel A, Thuere C, Wollenberg I, Yagita H, Volk HD, Zenclussen AC: Mechanisms of action of regulatory T cells specic for paternal antigens during pregnancy. Obstet Gynecol 2007; 110:11371145. 6 Wafula PO, Teles A, Schumacher A, Pohl K, Yagita H, Volk HD, Zenclussen AC: PD-1 but not CTLA-4 blockage abrogates the protective effect of regulatory T cells in a pregnancy murine model. Am J Reprod Immunol 2009; 62(5):283292. 7 Okabe M, Ikawa M, Kominami K, Nakanishi T, Nishimune Y: Green mice as a source of ubiquitous green cells. FEBS Lett 1997; 407:313319. 8 Blois S, Zenclussen AC, Roux ME, Olmos O, di Conza J, Arck PC, Margni RA: Asymmetric antibodies (AAb) in the female reproductive tract. J Reprod Immunol 2004; 64:3143. 9 Kirby DR, Billington WD, Bradbury S, Goldstein DJ: Antigen barrier of the mouse placenta. Nature 1964; 204:548549. 10 Billington WD: The normal fetomaternal immune relationship. Baillieres Clin Obstet Gynaecol 1992; 6:417 438. 11 Andrassy J, Kusaka S, Jankowska-Gan E, Torrealba JR, Haynes LD, Marthaler BR, Tam RC, Illigens BM, Anosova N, Benichou G, Burlingham WJ: Tolerance to noninherited maternal MHC antigens in mice. J Immunol 2003; 171:55545561. 12 Yan Z, Lambert NC, Guthrie KA, Porter AJ, Loubiere LS, Loubiere LS, Madeleine MM, Stevens AM, Hermes HM, Nelson JL: Male microchimerism in
American Journal of Reproductive Immunology 63 (2010) 200208 2010 John Wiley & Sons A/S

13

14

15

16

17

18

19

20

21

22

23

24

25

women without sons: quantitative assessment and correlation with pregnancy history. Am J Med 2005; 118:899906. Khosrotehrani K, Johnson KL, Guegan S, Stroh H, Bianchi DW: Natural history of fetal cell microchimerism during and following murine pregnancy. J Reprod Immunol 2005; 66:112. Tan XW, Liao H, Sun L, Okabe M, Xiao ZC, Dawe GS: Fetal microchimerism in the maternal mouse brain: a novel population of fetal progenitor or stem cells able to cross the blood-brain barrier? Stem Cells 2005; 23:14431452. Root-Bernstein RS, DeWitt SH: Semen alloantigens and lymphocytotoxic antibodies in AIDS and ICL. Genetica 1995; 95:133156. Robertson SA: Seminal plasma and male factor signalling in the female reproductive tract. Cell Tissue Res 2005; 322:4352. Sargent IL, Borzychowski AM, Redman CW: Immunoregulation in normal pregnancy and preeclampsia: an overview. Reprod Biomed Online 2006; 13:680686. Tilburgs T, Roelen DL, van der Mast BJ, de GrootSwings GM, Kleijburg C, Scherjon SA, Claas FH: Evidence for a selective migration of fetus-specic CD4+ CD25bright regulatory T cells from the peripheral blood to the decidua in human pregnancy. J Immunol 2008; 180:57375745. Bulmer JN, Johnson PM: The T-lymphocyte population in rst-trimester human decidua does not express the interleukin-2 receptor. Immunology 1986; 58:685687. Athanassakis I, Iconomidou B: Cytokine production in the serum and spleen of mice from day 6 to 14 of gestation: cytokines placenta spleen serum. Dev Immunol 1996; 4:247255. Tezabwala BU, Johnson PM, Rees RC: Inhibition of pregnancy viability in mice following IL-2 administration. Immunology 1989; 67:115119. Bischof RJ, Lee CS, Brandon MR, Meusen E: Inammatory response in the pig uterus induced by seminal plasma. J Reprod Immunol 1994; 26:131146. Autiero M, Abrescia P, Guardiola J: Interaction of seminal plasma proteins with cell surface antigens: presence of a CD4-binding glycoprotein in human seminal plasma. Exp Cell Res 1991; 197:268271. Robertson SA, Guerin LR, Bromeld JJ, Branson KM, Ahlstrom AC, Care AS: Seminal uid drives expansion of the CD4+ CD25+ T regulatory cell pool and induces tolerance to paternal alloantigens in mice. Biol Reprod 2009; 80:10361045. Onodera K, Lehmann M, Akalin E, Volk HD, Sayegh MH, Kupiec-Weglinski JW: Induction of infectious

207

ZENCLUSSEN ET AL.

tolerance to MHC-incompatible cardiac allografts in CD4 monoclonal antibody-treated sensitized rat recipients. J Immunol 1996; 157:19441950. 26 Scully R, Qin S, Cobbold S, Waldmann H: Mechanisms in CD4 antibody-mediated transplantation tolerance: kinetics of induction, antigen dependency and role of regulatory T cells. Eur J Immunol 1994; 24:23832392. 27 Waldmann H, Graca L, Cobbold S, Adamas E, Tone M, Tone Y: Regulatory T cells and organ transplantation. Semin Immunol 2004; 16:119126. 28 Zenclussen AC, Gerlof K, Zenclussen ML, Ritschel S, Zambon Bertoja A, Fest S, Hontsu S, Ueha S,

Matsushima K, Leber J, Volk HD: Regulatory T cells induce a privileged tolerant microenvironment at the fetal-maternal interface. Eur J Immunol 2006; 36:82 94. 29 Goswami D, Tannetta DS, Magee LA, Fuchisawa A, Redman CW, Sargent IL, von Dadelszen P: Excess syncytiotrophoblast microparticle shedding is a feature of early-onset pre-eclampsia, but not normotensive intrauterine growth restriction. Placenta 2006; 27:56 61. 30 Tafuri A, Alferink J, Moller P, Hammerling GJ, Arnold B: T cell awareness of paternal alloantigens during pregnancy. Science 1995; 270:630633.

American Journal of Reproductive Immunology 63 (2010) 200208

208

2010 John Wiley & Sons A/S

Copyright of American Journal of Reproductive Immunology is the property of Blackwell Publishing Limited and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.