Objective: Labor is associated with 'decidual activation' with increased proteolysis and extracellular matrix degradation. The balance between plasminogen activator inhibitor-1 (PAI-1) and urokinase (uPA) and tissue-type plasminogen activator (tPA) is an important determinant of proteolytic activity at the maternal-fetal interface. Thrombin released at the time of placental abruption (decidual hemorrhage) is known to promote decidual proteolysis and uterine contractions. This study investigates the separate and interactive effects of steroid hormones and thrombin on PAI-1, uPA, and tPA expression by term decidual cells (DCs).
Study design: Term DCs were isolated by enzymatic digestion, purified, and depleted of leukocytes. Cells were treated with estradiol (10(-8) mol [E2]), medroxyprogesterone acetate (10(-7) mol [MPA]), both, or vehicle for 7 days. After 24-hour incubation with or without thrombin (0.1-2.5 U/mL), levels of PAI-1, uPA, and tPA in conditioned supernatant were measured by specific ELISA and Western blotting. Levels of PAI-1 and uPA mRNA were measured by quantitative RT-PCR.
Results: In the cultured term DCs, ELISA measurements indicated that basal output of PAI-1 was about 2 logs higher than that of either uPA or tPA (2.5 +/- 0.7 ng/mL per microg protein, 13.4 +/- 6.3 pg/mL per microg protein, and 25.4 +/- 10.8 pg/mL per microg protein, respectively). Although E2 alone did not affect PAI-1 output, MPA and E2+MPA significantly enhanced PAI-1 production (2.5 +/- 0.7 vs 8.2 +/- 2.0 ng/mL per microg protein for E2+MPA [3.3-fold]; P < .01). By contrast, uPA output was inhibited by exposure to MPA (13.4 +/- 6.3 vs 2.6 +/- 1.1 pg/mL per microg protein [0.2-fold]; P < .05), whereas tPA production was not affected by MPA. Thrombin did not significantly affect uPA and tPA production by term DCs. In contrast, in E2+MPA-treated term DCs, thrombin, a hemostatic proinflammatory cytokine, selectively increased PAI-1 output in a dose-dependent fashion, which could be blocked by the selective thrombin inhibitor, hirudin. Western blotting confirmed the effects of MPA and thrombin in elevating secreted levels of PAI-1. Unlike the increase in PAI-1 output elicited by thrombin, term DCs were unresponsive to either of the classic proinflammatory cytokines, TNFalpha or IL-1beta. Corresponding effects on PAI-1 mRNA levels were elicited by MPA and thrombin as seen for PAI-1 protein expression, suggesting that these up-regulatory effects are transcriptionally mediated.
Conclusion: Progestin enhanced PAI-1 and inhibited uPA expression by term DCs, which may explain in part the pregnancy-prolonging properties of progesterone as a consequence of inhibited proteolytic activity at the maternal-fetal interface. Thrombin augmented PAI-1 expression in the absence of increased uPA or tPA expression by term DCs, suggesting that abruption-associated decidual proteolysis and preterm labor is mediated primarily by thrombin-enhanced matrix metalloproteinase expression rather than an indirect effect on the plasminogen activator/inhibitor system.