Review
Progesterone as an immunomodulatory molecule

https://doi.org/10.1016/S1567-5769(01)00035-2Get rights and content

Abstract

Increased progesterone sensitivity of pregnancy lymphocytes is due to activation-induced appearance of progesterone binding sites in the lymphocytes. Following recognition of fetally derived antigens γ/δ TCR+ cells develop progesterone receptors. Progesterone binding results in the synthesis of a mediator protein named the progesterone-induced blocking factor (PIBF). PIBF by acting on the phospholipase A2 enzyme interferes with arachidonic acid metabolism, induces a Th2 biased immune response, and by controlling NK activity exerts an anti-abortive effect.

Introduction

Pregnancy is characterized by overall hormonal changes. Steroid hormones are first produced by the corpus luteum following stimulation by human chorionic gonadotropin. From the seventh week of gestation, the placenta takes over steroid production and becomes the main source of steroid hormones until the end of pregnancy. In humans, progesterone production gradually rises during gestation to reach a level of 3000 ng/g of placental tissue [1]. The serum concentrations of progesterone range from approximately 100 to 500 nM during pregnancy. Blocking of progesterone binding sites by a progesterone antagonist RU 486 causes abortion in various mammalian species [2], [3].

Among steroids displaying immunomodulatory effects, progesterone has received much attention. Ovarectomized animals treated with progesterone, but not with estrogen, develop uterine abscesses after injection with Escherichia coli. The number of leukocytes, as well as immunoglobulin production in the rodent uterus, has been shown to depend on the hormonal status [4], [5]. High local concentrations of progesterone prolong the survival of xenogeneic and allogeneic skin grafts as well as of xenogeneic tumor cells implanted in hamster uteri [6]. In vitro treatment of T cells with pharmacological concentrations of progesterone blocks T cell activation and killing [7], [8], [9]. It is noteworthy that in all of the early human studies, nonpregnancy lymphocytes were used as targets for progesterone action and supra-physiologic concentrations of progesterone were needed for achieving a significant effect.

Data from our laboratory revealed a 100-fold higher progesterone sensitivity of pregnancy lymphocytes compared to that of lymphocytes from nonpregnant individuals [10]. Based on the increased progesterone responsiveness of pregnancy lymphocytes, it seemed plausible that lymphocytes of pregnant women express binding sites for progesterone.

A key component of hormone responsiveness is the available concentration of functional receptors. Alternatively, a direct inhibition of K+ channels in T cells by progesterone might contribute to progesterone-induced immunosuppression [11]. Early investigations failed to demonstrate progesterone receptors (PR) in normal lymphocytes by radioactive ligand-binding techniques [12]. Our own ligand-binding experiments on pregnancy lymphocytes suggested the presence of specific binding sites in human pregnancy lymphocytes, although at a lower density than in other progesterone receptor-containing tissues. Staining with a panel of progesterone receptor-specific monoclonal antibodies revealed a nuclear reactivity in pregnancy, but not in nonpregnancy, peripheral blood lymphocytes [13]. Similar differences were found between healthy pregnant and nonpregnant donors in the concentration of the cytosolic binding sites, which represent a part of the unoccupied receptors [14]. Furthermore, the synthesis of a progesterone-dependent protein (PIBF) was inhibited by RU 486, an antagonist of progesterone and glucocorticoid receptors, but not by a specific glucocorticoid antagonist [15].

The existence of lymphocyte progesterone receptors is a much debated issue. Kimoto et al. [16] demonstrated the mRNA for PR in human lymphocytes. The PR protein has been identified in human multiple myeloma cells B cells, plasma cells and macrophages as well as in murine hybridomas [17], [18]. Chiu et al. [19] found an increased PR expression on peripheral pregnancy lymphocytes after immunotherapy for habitual abortion. King et al. [20] and Steward et al. [21] failed to demonstrate PR or ER expression in endometrial lymphocytes. The latter group, however, found an increased ER expression on the lymphocytes in endometriosis [22], suggesting that the expression or detectability of steroid receptors might largely depend on the biological situation. On the other hand, van den Heuvel et al. [23] detected progesterone receptor-positive uterine NK cells at day 8 of gestation in mice. Furthermore, Roussev et al. [24] demonstrated an enrichment of progesterone receptor-positive lymphocytes in the pregnant uterus compared to peripheral blood. These contradictory data suggest that expression of the PR protein depends on the actual condition of the lymphocyte.

In our hands, both radioactive ligand binding and enzyme immunoassay revealed a substantially lower number of binding sites in the lymphocytes than in the classical target or mammary tumor tissues [14].

A high proportion of peripheral pregnancy lymphocytes react with the PR-specific monoclonal antibodies, whereas resting nonpregnancy lymphocytes are largely nonreactive [14]. The rate of receptor-positive cells increases throughout gestation [13]. The percentage of progesterone receptor-expressing cells among lymphocytes of recurrent aborters, or among those of women with clinical symptoms of threatened preterm delivery, is significantly lower than that in peripheral blood from pregnant women of corresponding gestational ages [13].

These findings suggest a relationship between lymphocyte progesterone receptor expression and the outcome of pregnancy; thus; the regulation of lymphocyte progesterone receptors might be of biological importance. In contrast to the classical endocrine tissues, in lymphocytes estrogens do not induce PRs [14]. This might be due to the absence of estrogen receptors in peripheral human lymphocytes. Though E2 receptors have been demonstrated in mouse spleen, rat thymus, guinea pig fetal thymus, human T cells human mononuclear cells and macrophages [25], [26], [27], [28], we could not detect estrogen receptors in resting human lymphocytes, by an enzyme immunoassay utilizing a rat monoclonal antibody to the human estrogen receptor [14].

On the other hand, exposure of nonpregnancy lymphocytes to PHA or alloantigenic stimuli was followed by increased lymphocyte PR expression [14], [29], suggesting that PR induction is an activation-related phenomenon.

Allogeneic stimulation of maternal lymphocytes by paternal-type antigens during pregnancy is a real possibility.

Transplantation seemed to be an ideal natural model for answering the question, whether allogeneic stimulation by itself is the only condition required for inducing progesterone receptors, or if hormonal and other pregnancy-specific aspects are also involved.

The expression of progesterone receptors was equally high in transplanted patients and in pregnant women, showing that pregnancy is not the major factor for progesterone receptor induction [30].

This observation suggests that PR expression is not the consequence of pregnancy per se, but that of in vivo allogeneic stimulation by foetal antigens. The above assumption gains support from the data of Chiu et al. [19], who found an increased PR expression in lymphocytes from recurrent aborters after immunization with paternal leukocytes. These data clearly show that activation is a major factor in lymphocyte progesterone receptor induction.

Therefore, the setting in of a progesterone-dependent immunomodulatory mechanism depends on the mode of fetal antigen presentation and, at the other end, on the immunocompetent cell that will recognize the antigen and determine the quality and the extent of the response.

Polymorphic MHC is absent from the trophoblast; thus, presentation of fetally derived antigens is unlikely to be MHC-restricted [31]. γ/δ T cells recognize a distinct group of ligands with a smaller receptor repertoire than α/β T cells and, therefore, play a different role in immunity than α/β T cells [32], [33]. Most γ/δ T cells recognize unprocessed foreign antigens without MHC.

In peripheral blood of healthy pregnant women, γ/δ TCR-positive cells are significantly increased in number and express progesterone receptors [34]. On the other hand, more than 90% of PR-positive peripheral blood pregnancy lymphocytes express a γ/δ TCR [34], [35], [36]. In the decidua where there is a high rate of overlapping between NK and γ/δ T cells [37], van den Heuvel et al. [23] identified PR-positive lymphocytes with NK cells.

In the decidua, the ratio of γ/δ TCR-positive cells is significantly higher than that in peripheral blood [38], [39]. The number of γ/δ T cells in the uterus is higher in allogeneic than in syngeneic pregnancy and the expression of the γ/δ TCR in the pregnant uterus has been shown to be hormonally controlled [40]. Therefore, it seems likely that this population might play a role in recognition of fetal antigens.

The majority of decidual γ/δ T cells are in an activated form [40], [41] and peripheral γ/δ TCR-positive lymphocytes of pregnant women express progesterone receptor, which is an activation marker [34], [35].

What is the nature of the antigen that activates these lymphocytes? Different classes of the trophoblast express tissue-specific antigens, oncofetal antigens; however, the expression of classical polymorphic Class I molecules is markedly reduced or absent [31]. Extravillous cytotrophoblast cells, on the other hand, express nonconventional HLA molecules [42], [43], [44], [45].

γ/δ T cells react with nonpolymorphic Class I or Class I-like molecules [46], [47]. Heyborne et al. [47] showed that trophoblast recognition is TCR-dependent, but not MHC-restricted and is mediated by the Hsp60 reactive Vγ1 subset of γ/δ T lymphocytes.

Almost all γ/δ T cells in the decidua use the Vδ1 chain and only a small fraction uses the Vδ2 chain [48], [49]. The majority of human peripheral γ/δ lymphocytes expresses the Vγ9/Vδ2 TCR [50]. Vγ9/Vδ2 TCR-positive cells are able to respond to nonprocessed and nonpeptidic phosphoantigens in an HLA-unrestricted manner [51], [52].

In our hands, the ratio of Vγ1.4Vδ1 and Vγ9Vδ2 cells was eight times higher in peripheral blood of healthy pregnant women, than that of recurrent aborters. Thus, normal pregnancy seems to be characterized by an altered rate of these subpopulation; whereas in recurrent aborters, the ratio of Vγ9Vδ2 to Vγ1Vδ1 expressing cells is similar to that in nonpregnant individuals [35], [36]. Recent data from our laboratory suggest that γ1δ1 cells do not recognise nonconventional HLA molecules on the trophoblast, whereas γ9/δ2 cells preferentially recognise nonpolymorphic MHC via their KIR receptor (manuscript in preparation). This would result in activation of the γ1δ1 subset of decidual lymphocytes, which, in turn, would develop progesterone receptors.

Considering the high local concentration of progesterone at the feto-maternal interface, along with the fact that progesterone has been shown to modulate lymphocyte functions in vitro, this finding might be biologically meaningful.

Section snippets

Progesterone-induced blocking factor

Allogeneic stimulation occurring during pregnancy allows the binding of progesterone to specific lymphocyte-expressed receptors.

Steroid hormones act according to the “hit-and-run” mechanism. Each steroid receptor undergoes a structural alteration or “transformation”-upon-hormone exposure, which, in turn, enables DNA binding [53]. This might result in the induction of genes leading to protein synthesis. DiRosa and Persico [54] described a protein synthesis-dependent inhibition of prostaglandin

The effects of PIBF on cytokine synthesis

It is now evident that the protection of fetus from a harmful maternal immune response is based on a complicated mechanism, and the communication between the various steps in the cascade of events is accomplished via cytokines.

Many observations suggest that pregnancy is associated with an altered TH1/TH2 balance. Soon after fertilization, the fertilized ovum signals to the maternal immune system [67], and this signal will induce the shift in cytokine production towards Th2 [68].

Both clinical

References (115)

  • G. Par et al.

    Cyclooxygenase is involved in the effect of Progesterone Induced Blocking Factor (PIBF) on IL-12 production

    Am. J. Obstet. Gynecol.

    (2000)
  • J. Szekeres-Bartho et al.

    Immunoregulatory effects of a suppressor factor from healthy pregnant womens lymphocytes after progesterone induction

    Cell. Immunol.

    (1989)
  • J. Szekeres-Bartho et al.

    A progesterne-dependent immunomodulatory protein alters the Th1/Th2 balance

    J. Reprod. Immunol.

    (1996)
  • J. Szekeres-Bartho et al.

    A progesterone-induced immunologic blocking factor corrects high resorption rate in mice treated with anti-progesterone

    Am. J. Obstet. Gynecol.

    (1990)
  • J. Szekeres-Bartho et al.

    The anti-abortive effect of PIBF in mice is manifested by modulating NK activity

    Cell. Immunol.

    (1997)
  • M.J. Myers et al.

    Estradiol induced alterations of the immune system: I. Enhancement of IgM production

    Int. J. Immunopharmacol.

    (1985)
  • T.G. Wegmann et al.

    Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a Th2 phenomenon?

    Immunol. Today

    (1993)
  • M. Parant

    Possible mediators of endotoxin-induced abortions

    Res. Immunol.

    (1990)
  • Y. Wang et al.

    Sex hormones and dexamethasone modulate interleukin-5 gene expression in T lymphocytes

    J. Steroid Biochem. Mol. Biol.

    (1993)
  • K. Aoki et al.

    Preconceptional natural-killer-cell activity as a predictor of miscarriage

    Lancet

    (1995)
  • J. Szekeres-Bartho et al.

    Immunosuppressive effect of progesterone during pregnancy depends on the progesterone binding capacity of the lymphocytes

    J. Reprod. Immunol.

    (1983)
  • J. Szekeres-Bartho et al.

    Immunological factors contributing to the initiation of labour-lymphocyte reactivity in term labour and threatened pre-term delivery

    Am. J. Obstet. Gynecol.

    (1986)
  • B.A. Croy et al.

    Characterization of murine decidual natural killer (NK) cells and their relevance to the success of pregnancy

    Cell. Immunol.

    (1985)
  • B.A. Croy et al.

    Normal murine and porcine embryos recruit NK cells to the uterus

    Cell. Immunol.

    (1988)
  • R.S. Parhar et al.

    PGE2-mediated immunosuppression by first trimester human decidual cells blocks activation of maternal leucocytes in the decidua with potential anti-trophoblast activity

    Cell. Immunol.

    (1989)
  • J. Visser et al.

    Differential regulation of interleukin-10 (IL-10) and IL-12 by glucocorticoids in vitro

    Blood

    (1998)
  • F. Ferre et al.

    Influence of the oral administration of micronized progesterone on plasma and tissue steroid levels in human pregnancy

    Acta Physiol. Hung.

    (1985)
  • D. Philibert et al.

    Pharmacological profile of RU 486 in animals

  • C.P. Puri et al.

    Effect of the antiprogestin ZK.98734 on the ovarian cycle, early pregnancy and on its binding to progesterone receptors in the myometrium of marmoset Callithrix jacchus

    Biol. Reprod.

    (1988)
  • C.R. Wira et al.

    Hormonal regulation of immunoglobulins: influence of estradiol in immunoglobulins A and G in the rat uterus

    Endocrinology

    (1980)
  • I. Moriyama et al.

    Progesterone facilitates implantation of xenogeneic cultured cells in hamster uterus

    Nat. New Biol.

    (1972)
  • F.A. Wyle et al.

    Immunosuppression by sex steroid hormones: I. The effect upon PHA and PPD stimulated lymphocytes

    Clin. Exp. Immunol.

    (1977)
  • G.R. Ehring et al.

    A nongenomic mechanism for progesterone-mediated immunosuppression: inhibition of K+ channels, Ca2+ signaling, and gene expression in T lymphocytes

    J. Exp. Med.

    (1998)
  • J. Szekeres-Bartho et al.

    Progesterone suppression of pregnancy lymphocytes is not mediated by a gluccorticoid effect

    Am. J. Reprod. Immunol.

    (1990)
  • Y. Kimoto

    A single human cell expresses all messenger ribonucleic acids: the arrow of time in a cell

    Mol. Gen. Genet.

    (1998)
  • L. Daniel et al.

    Estrogen and progesterone receptors in semen human myeloma cell lines and murine hybridomas

    J. Steroid Biochem.

    (1988)
  • L. Chiu et al.

    Enhancement of the expression of progesterone receptor on progesterone-treated lymphocytes after immunotherapy in unexplained spontaneous abortion

    Am. J. Reprod. Immunol.

    (1996)
  • A. King et al.

    Evaluation of oestrogen and progesterone receptor expression in uterine mucosal lymphocytes

    Hum. Reprod.

    (1996)
  • J.A. Steward et al.

    Endometrial leucocytes: expression of steroid hormone receptors

    J. Clin. Pathol.

    (1998)
  • R.K. Jones et al.

    Phenotypic and functional studies of leukocytes in human endometrium and endometriosis

    Hum. Reprod. Update

    (1998)
  • M.A. Defletsen et al.

    A high affinity low-capacity receptor for estradiol in normal and anaemic mouse spleen cytosols

    Biochem. Biophys. Res. Commun.

    (1977)
  • I. Screpanti et al.

    The fetal thymus of guinea pig as an estrogen target organ

    Endocrinology

    (1982)
  • J.H.M. Cohen et al.

    Sex steroid receptors in peripheral T cells: absence of androgen receptor and restriction of estrogen receptors to OKT8 positive cells

    J. Immunol.

    (1983)
  • W.H. Stimson

    Oestrogen and human T lymphocytes: presence of specific receptors in the T suppressor cytotoxic subset

    Scand. J. Immunol.

    (1988)
  • A. Paldi et al.

    Expression of the gene coding for the progesterone receptor in activated human lymphocytes

    Endocr. J.

    (1994)
  • J. Szekeres-Bartho et al.

    Progesterone receptors in lymphocytes of liver-transplanted and transfused patients

    Immunol. Lett.

    (1989)
  • W.D. Billington

    The nature and possible functions of MHC antigens on the surface of human trophoblast

  • B.C. Weintraub et al.

    γ/δ T cells can recognize nonclassical MHC in the absence of conventional antigenic peptides

    J. Immunol.

    (1994)
  • W. Born et al.

    Recognition of a peptide antigen by heat shock-reactive γ/δ T lymphocytes

    Science

    (1990)
  • B. Polgar et al.

    The role of γ/δ TCR positive cells in pregnancy

    Am. J. Reprod. Immunol.

    (1999)
  • Cited by (0)

    View full text