IL-11 expressed by endometrial stromal cells is crucial for normal pregnancy. IL-11 receptor alpha (IL-11Ralpha) null mice are infertile due to abnormal development of the placenta. In these mice, the mesometrial decidual tissue, which is the site of trophoblast invasion, thins and disappears at mid-pregnancy. Degeneration of the decidua is accompanied by uncontrolled trophoblast invasion. In this report, we show, using IL-11Ralpha null mice, that a defect in IL-11 signaling in the decidua leads to severe down-regulation of alpha(2)-macroglobulin (alpha(2)-MG), a metalloproteinase inhibitor crucial for limiting trophoblast invasion. We also present evidence, using uterine stromal cells that decidualize in culture, that IL-11 robustly stimulates the endogenous alpha(2)-MG expression and enhances alpha(2)-MG promoter activity. Serial 5' deletion and internal deletion of the promoter reveal two important signal transducer and activator of transcription (Stat) binding sites. Mutation of either one of these motifs decreases IL-11 stimulation, whereas double mutation prevents IL-11 action. We also found that IL-11 activates Janus kinase 2 (Jak2) and induces rapid phosphorylation, nuclear translocation, and promoter binding activity of Stat3 in decidual cells, whereas Jak1, Tyk2, and Stat5 activities are not affected. In addition, Jak2 inhibitor totally prevents alpha(2)-MG expression in decidual cells. Taken together, results of this investigation provide, at least in part, an explanation for the overinvasiveness of the trophoblast in IL-11Ralpha null mice and reveal, for the first time, that IL-11 signals through the Jak2/Stat3 pathway in decidual cells to stimulate the expression of alpha(2)-MG, a protease inhibitor essential for normal placentation in pregnancy.

Successful pregnancy requires profound differentiation and reorganization of the uterine tissues including, as pregnancy progresses, extensive apoptosis of decidual tissue to accommodate the developing conceptus. We have previously shown a positive correlation between expression of activin A and apoptosis in the decidua and have also shown that expression of activin A occurs at the time when prolactin (PRL) receptors disappear from decidual cells. The goals of this study were to examine whether activin A plays a role in decidual apoptosis and whether expression of activin A in the decidua is regulated by PRL and placental lactogens. Studies were carried out using primary rat decidual cells, a decidual cell line (GG-AD), and PRL null mice. Treatment of decidual cells with activin A significantly increased DNA degradation, caspase 3 activity, and caspase 3 mRNA expression. However, this effect was observed only in the absence of endogenous activin production by these cells. Addition of follistatin to decidual cells that were producing activin A decreased both caspase 3 activity and mRNA expression. Similarly, addition of activin-blocking antibodies to cultures of GG-AD cells, which also produce activin A, caused a reduction in both DNA degradation and caspase 3 activity. PRL and placental lactogens caused an inhibition of activin A mRNA expression in primary decidual cells. Even more convincingly, decidua of PRL null mice expressed abundant activin A at a time when no expression of this hormone is detected in wild-type mice and treatment of PRL null mice with PRL caused a profound inhibition of activin A mRNA expression. In summary, our investigations into the role and regulation of decidual activin have revealed that activin A can induce cell death in the decidua and that its expression is under tight regulation by PRL and placental lactogens.

Inflammatory cytokines secreted by the placenta and fetal membranes are believed to play an important role in the initiation of parturition. The suppressor of cytokine signaling (SOCS) proteins regulate signal transduction by several cytokines that have been reported to affect gestational tissues. The presence, distribution and roles of SOCS proteins, however, have not been described in human gestational tissues. Using reverse transcriptase (RT)-PCR and Western blot analysis we investigated the expression of SOCS1, SOCS2, and SOCS3 mRNA and protein, respectively, by human villous placenta, amnion and choriodecidua (n = 3-4). Tissues were obtained from uncomplicated pregnancies at term after either spontaneous labor and vaginal delivery or caesarean section (before labor). Messenger RNAs for SOCS1, SOCS2, and SOCS3 were expressed in all tissue types, irrespective of labor status. SOCS proteins were, however, only detectable in villous placenta and in one case in the choriodecidua. Labor was associated with abrogated expression of SOCS1 and SOCS3 proteins in villous placenta and the choriodecidua sample. Following labor the band for SOCS2 protein increased slightly in size which may indicate post-translational modification of SOCS2. Reduced expression of SOCS proteins in gestational tissues may provide a mechanism by which inflammatory cytokines enter into a positive feedback loop of inflammatory changes leading to delivery.

Decreased suppressors of cytokine signaling (SOCS) activity in human gestational tissues may play a part in the onset/progression of term labor. Since SOCS proteins negatively regulate cytokine-mediated inflammatory processes, we hypothesized that SOCS proteins are elevated in gestational tissues from spontaneous preterm deliveries with intrauterine infection. SOCS1, -2 and -3 mRNAs and proteins were detectable by RT-PCR and immunoblotting respectively, in preterm amnion, choriodecidua and placenta, irrespective of infection status. Immunoperoxidase staining localized SOCS1, -2 and -3 to all cell types of the gestational membranes, with infiltrating leukocytes reacting strongly in infected tissues. In villous placenta, SOCS was immunolocalized to the syncytiotrophoblast with marked staining of round mesenchymal cells, possibly Hofbauer cells. Nuclear SOCS staining was seen in amnion, chorion and placental syncytiotrophoblasts. SOCS proteins were, in general, significantly more abundant in placenta compared with amnion or choriodecidua. Placental SOCS1 and interleukin-1 concentrations were positively correlated (r 2 =0·47; P<0·05). However, no changes in SOCS levels in any tissues were observed with intrauterine infection. The relatively large amounts of SOCS proteins in the placenta may reflect a placenta-specific immuno-protective response to minimize the elaboration and effects of cytokines with potential to harm the placenta and fetus. Lack of labor-associated changes in SOCS levels suggests that the regulation of SOCS expression in preterm gestational tissues differs from those at term, perhaps reflecting roles in regulating placental somatotropic responses.