Recurrent spontaneous abortion (RSA), defined as two or more consecutive pregnancy losses prior to 20 weeks of gestation, may seriously compromise not only the physical but also psychological well-being of women. Although parental or embryonic karyotype anomalies, infection-associated factors, endocrine disorders, uterine anatomic abnormalities and anti-phospholipid syndrome have been acknowledged as the common causes of RSA, approximately half of the cases still remain unexplained (1). The development of the allogeneic fetus in the maternal uterus represents an immunological paradox. A successful pregnancy requires the maternal immune system to accept the fetus expressing allogeneic paternal antigens and provide competent responses to infections. Thus, maternal-fetal immune abnormalities might have an important role in the development of RSA (2).

The modulation of the maternal CD4⁺ T-cell response to fetal antigens has been considered an important component of maternal-fetal tolerance during pregnancy. Upon encountering antigens on the surface of antigen-presenting cells or being driven by a set of cytokines, naive CD4⁺ T cells are able to differentiate into distinct subsets, including type-1 T-helper (Th1), Th2, Th17 and T-regulatory (Treg) cells (3). While a pro-inflammatory microenvironment is required for trophoblastic growth and invasion (4), Th1 cytokines, including tumor necrosis factor-α and interferon-γ, have deleterious effects on pregnancy (5). For numerous years, it has been thought that a shift in the maternal immune response toward a Th2 bias is crucial for maintaining a successful pregnancy (6).

Th17 cells have a critical role in inducing inflammation, while abnormal Th17-cell levels have been associated with the pathogenesis of autoimmune diseases (7). Treg cells express anti-inflammatory cytokines, including interleukin (IL)-10 and transforming growth factor (TGF)-β1, which dampen an excessively effective immune response (8). Recent studies have indicated that the balance between Treg and Th17 cells is also important for maintaining a normal pregnancy, while a shift in the Th17/Treg ratio toward Th17 cells has been suggested to cause several pregnancy-associated diseases, including RSA, pre-eclampsia and gestational diabetes mellitus (9-11).

In addition to the signal from antigen recognition, T-cell activation requires a second signal from co-stimulatory molecules, which maybe either positive or negative (12,13). Among the inhibitory co-stimulatory molecules, cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), T-cell immunoglobulin mucin-3 (Tim-3) and programmed cell death-1 (PD-1) have attracted much attention regarding their roles in regulating Th17- and Treg-cell function to maintain a balanced immune response (14-17). For instance, Tim-3 has been associated with a shift in the balance from Th17 and Treg cells to Treg dominance (16). The PD-1 signaling pathway has been demonstrated to have an important role in Th17-cell function during pregnancy (18). Furthermore, PD-1 and CTLA-4 are critical for the suppressive activity of Treg cells (19,20), while PD-1⁺Tim-3⁺Treg cells exhibit an increased effector function (21). Previous studies by our group revealed that the expression of these three molecules was lower in human miscarriages than in normal pregnancies, while their blockade during the implantation window increased maternal rejection of the fetus in CBA/Jx BALB/c matings (22,23).

Progesterone (P4), the most common medication for fetal protection in RSA, is a multifunctional female hormone that promotes breast development, ovulation and pregnancy maintenance; it is mainly secreted by ovarian granulosa cells and the corpus luteum during non-pregnancy and by the placenta during pregnancy (24). While the mechanisms remain to be fully elucidated, P4 has been reported to dampen immune responses to fetal and maternal antigens (25). As an immunosuppressant, cyclosporine A (CsA) has been widely used in organ transplantation (26). CsA not only induces Th2 immuno-dominance at the maternal-fetal interface, but also promotes trophoblastic proliferation and invasion, further promoting placental formation (27). This immunosuppressive agent, categorized as pregnancy category C by the Food and Drug Administration, has been utilized for the treatment of RSA at our hospital (Hospital of Obstetrics and Gynecology, Fudan University Shanghai Medical College) since 2005.

To the best of our knowledge, studies on the effects of P4 or CsA on the proportion of Th17 and Treg cells within the peripheral blood of patients with RSA are rare. Therefore, the present study aimed to investigate the effects of P4 or CsA therapy on the Th17/Treg cell balance and the expression of associated molecules in the peripheral blood of pregnant females with a history of RSA.

As a semi-allograft, the human fetus is antigenically foreign to the mother. Accordingly, a successful pregnancy relies on the fine maternal-fetal immune balance between immune tolerance, allowing allogeneic fetal trophoblasts to invade maternal tissues and immune defense against a variety of pathogens. Disruption of this immune balance is thought to be associated with several pregnancy-associated complications, including RSA, preeclampsia and fetal intrauterine growth restriction (2). The present study also confirmed that imbalance of Th17/Treg cells and associated molecule is linked to RSA.

P4, the traditional treatment for RSA, inhibits uterine contraction and improves luteal-phase support and utero-placental circulation (29). Chien et al (30) indicated that P4 may also regulate immunity and inhibit T-cell activation through a non-classical steroid receptor pathway during pregnancy. In early 1995, Piccinni et al (31) proposed that P4 was beneficial to Th2 differentiation of CD4⁺ T cells. Furthermore, Mjosberg et al (32) revealed that P4 regulated Treg-cell functions in human pregnancy, while Lee et al (33) reported that P4 inhibited Th17-cell differentiation and reduced the expression of associated functional molecules, including IL-17A. However, the present study revealed no significant differences in the proportion of Treg and Th17 cells in the peripheral blood between the P4 and RSA groups. This may be due to differences in the gestational stages tested and the fact that certain studies were performed using established cell clones.

While certain studies have suggested the involvement of Th17 cells in pregnancy-associated pathologies (34), a study by our group (35) and Wu et al (36) suggested that Th17 cells serve important roles in regulating trophoblast function. It has also been indicated that not all Th17 cells are pro-inflammatory, given that Th17 subpopulations with diverse functions may exist (37). IL-10 has been previously associated with non-pathogenic Th17 cells (38), while TGF-β1 may be produced mainly by Th17 cells (39). Although P4 had no effect on IL-10 expression in all three indicated subsets of T cells and TGF-β1 expression in pCD4⁺ T cells and pTreg cells, it significantly increased the expression of TGF-β1 and CTLA-4 in pCD4⁺ T cells and pTh17 cells, as well as on Tim-3 expression in all three indicated subsets of T cells and PD-1 expression in pTreg cells. As co-stimulatory signals have important roles in T-cell function and pregnancy maintenance (18,28), despite the lack of evidence regarding improved outcomes following first-trimester progesterone therapy in females with a history of RSA (40,41), the results of the present study suggested that P4 may also have an immunomodulatory role through certain co-inhibitory molecules (including Tim-3, PD-1 and CTLA-4) in the treatment of RSA.

The observed association between immunologic abnormalities and RSA has led to the development and use of various types of immune therapy with the objective of restoring normal immune homeostasis during pregnancy. CsA is a previously known immunosuppressive drug that has been used for the treatment of RSA based on previous studies (27,42-48). Previous studies report that CsA: i) Induced Th2 dominance and Treg-cell proliferation during the implantation window and significantly reduced the embryo absorption rate (27,42); ii) downregulated CD28 expression at the maternal-fetal interface and upregulated CTLA-4 expression in abortion-prone mice (43); iii) improved adhesion and invasion of mouse preimplantation embryos (44); and iv) promoted trophoblastic invasion and proliferation at low dosages (0.0001-1 µmol/l) (45,46). Small-scale clinical observations (47,48), in addition to unpublished data by our group, have indicated that CsA significantly increased the live birth rate in cases of RSA without increasing complications during pregnancy and among neonates.

The present study suggested that CsA reduced the proportion of Th17 cells and promoted Treg-cell dominance among pregnant females with a history of RSA. CsA decreased the production of IL-17A (the major effector cytokine of Th17 cells) and also upregulated the expression of TGF-β1, IL-10, CTLA-4 and Tim-3. These results suggest that CsA, as a trophoblast function promoter, also restores the Th17/Treg-cell balance, while inducing Th2 bias during pregnancy. Accordingly, future research group will be aimed at studying the effect of P4 and CsA on other pro-inflammatory cytokines in Th17 cells, including IL-6, IL-21 and IL-23. Further investigations are required to explore whether non-pathogenic Th17 subsets and pro-inflammatory Th17 subpopulations co-exist at the maternal-fetal interface, and which mechanisms may be used to modulate Th17 subsets during pregnancy.

To date, RSA with its underlying causes and mechanisms has remained to be fully elucidated. Studies have indicated that the more abortions occur, the higher the likelihood of abortion during the next pregnancy. For instance, after two spontaneous abortions, the risk of spontaneous abortion during the third pregnancy is ~30%, whereas after >3 spontaneous abortions, the risk of spontaneous abortion during the next pregnancy maybe as high as 45% (49). Early diagnosis and treatment of RSA has been of great significance to families. Although previous studies have focused on the maternal-fetal interface, non-invasive examination is of great importance in clinical practice. The present study investigated the imbalance between Th17 and Treg cells in the peripheral blood to improve the current understanding of the pathogenesis of RSA. The results confirmed that the percentage of Treg cells and the expression of TGF-β1, IL-10, CTLA-4 and Tim-3 decreased in the peripheral blood of RSA patients, which may provide an experimental basis for non-invasive monitoring indicators of pregnancy. However, while the experiments were performed using flow cytometry with PMA stimulation, determining the serum levels of cytokines in peripheral blood or cytokine production in different T-cell subpopulations through ELISA or PCR may have been more representative of the in vivo situation. Of course, further follow-up and larger-scale studies are required to confirm whether such molecules maybe used as monitoring indicators of pregnancy. The present results revealed that, while CsA reduced the proportion of Th17 cells and promoted Treg-cell dominance, P4 may also have an immunomodulatory role through certain co-inhibitory molecules to exert its treatment effect in RSA. However, the small sample size and lack of follow-up of pregnancy outcomes have limited the assessment of the effects of P4 and CSA on fetal protection in the present study. However, follow up of the subjects of the present study regarding pregnancy outcomes will continue and, in the future, the sample size will be expanded.