Moreover, excessive CSF1-driven accumulation of decidual macrophages at the fetal-maternal interface can be prevented by the high constitutive decidual MIF levels, resulting in tight control of the decidual macrophage population that could have an impact on fertility and pregnancy maintenance.
Previous studies have shown that pre-eclampsia is associated with augmented expression of TNF and IL1B and excess decidual macrophage infiltration.
Decidual macrophages bind to and phagocytose bacteria, regulate placental apoptosis, and are involved in clearance of placental apoptotic cells. Conversely, macrophages are also a source of factors that may lead to adverse effects on the conceptus. Tissue macrophages synthesize and secrete the Th1 cytokines interferon-gamma and TNF, as well as the effector molecules nitric oxide and prostaglandins. Consequently, mechanisms that ensure the regulation of macrophage numbers at the fetal-maternal interface are likely to play crucial roles in the establishment and maintenance of pregnancy.
Although two recent reports have demonstrated that exposure to TNF and IL1B increases MIF protein and mRNA levels in stromal cells derived from cycling human endometrium, the current results indicate that while MIF is synthesized and secreted at high concentrations by term decidual cells, its expression is not regulated by TNF or IL1B. Previous studies have demonstrated that MIF can induce, either directly or indirectly, the production of several proinflammatory molecules, including cytokines, nitric oxide, and prostaglandins.
The present study immunolocalized CSF1 and MIF to decidual cells in sections of term decidua, and established that leukocyte-free, cultured decidual cells from first trimester and term placentas synthesize and secrete both chemokines. These findings extend previous observations of CSF1 and MIF expression in early decidua to include term decidua, and suggest that these chemokines are synthesized and secreted by decidual cells throughout pregnancy. Several lines of evidence prompted us to evaluate the effects of TNF and IL1B on term decidual cells. TNF and IL1B are expressed at the fetal-maternal interface.
Corresponding results were obtained when: 1) the incubation period with TNF and IL1B was extended to 48 h; 2) the cytokine concentration was increased to 10 ng/ml; and 3) cultures were not treated with steroids.
To determine whether this differential response to inflammatory cytokines was present throughout gestation, TNF and IL1B were tested on cultured first trimester decidual cells.
In E2 + MPA-treated cultures, CSF1 output was increased from the basal level of 0.2 6 0.05 ng/ml/mg protein to 0.7 6 0.14 (P < 0.05) and 0.79 6 0.13 (P < 0.05) ng/ml/ mg protein by TNF and IL1B, respectively. In contrast to the marked elevation of CSF1 elicited by these cytokines, the addition of MPA plus E2 did not significantly alter the levels of CSF1 compared to cultures treated with E2 alone, and did not alter the responses to the cytokines.
CSF1 and MIF Expression in Term Decidua
The distributions of the CSF1 and MIF proteins in term decidua were analyzed by immunohistochemistry. Strong immunostaining for CSF1 (Fig. 1, A and B) and MIF (Fig. 1, C and D) were observed in the decidua. Observations at higher magnification revealed that both proteins were primarily distributed in the cytoplasm. Intense immunostaining was also detected in the syncytiotrophoblast (CSF1) and cytotrophoblast (MIF) of the chorionic villi. In addition, MIF immunoreactivity was consistently found in extravillous trophoblasts (data not shown). Staining for both cytokines was eliminated by substituting non-immune serum for the primary antibodies (data not shown).
Real-Time Quantitative RT-PCR
Total RNA was extracted from cultured cells with Tri Reagent (Sigma). Reverse transcription was carried out with an AMV reverse transcriptase kit (Invitrogen) on an Eppendorf Mastercycler (Eppendorf, Westbury, NY). To perform quantitative real-time RT-PCR, a standard curve was created for 500 pg to 250 ng cDNA using the Roche Light Cycler (Roche, Indianapolis, IN), with monitoring of the increasing fluorescence of the PCR products during amplification.
Since the circulating levels of E2 and progesterone are high during the third trimester, E2 was employed with MPA to mimic the gestational steroidal milieu. MPA was used instead of native progesterone, which is rapidly metabolized in vitro. The cultures were washed twice with PBS, to remove residual serum components, and switched to a serum-free defined medium (DM) that consisted of BM plus ITS+ premix (BD Biosciences, Bedford, MA), 5 im FeSO4, 50 im ZnSO4, 1 nm CuSO4, 20 nm Na2SeO3, trace elements (Invitrogen), 50 ig/ml ascorbic acid (Sigma), and 50 ng/ml epidermal growth factor (BD Biosciences).
The decidual cells were seeded onto polystyrene tissue culture dishes that were precoated with 2% type B gelatin (Sigma Chemical Co.). The cultures were grown to confluence in 5% CO2 in air at 37°C, and passaged three times. Fluorescent antibody cell sorting for the presence of CD45 + demonstrated that unpassaged cultures contained 12-15% CD45+ cells, while the passaged cultures were >99% negative for this common leukocyte marker. The latter cell populations were used for the subsequent experimental cell incubations. The cultured cells were vimentin-positive and cytokeratin-negative.