Tissue plasminogen activator and its receptor in the human amnion, chorion, and decidua at preterm and term.

The plasminogen activator system consists of two proteins: tissue plasminogen activator (tPA) and urokinase plasminogen activator (uPA), which act upon their specific receptors to generate plasmin from plasminogen located on the cell surface. Plasmin then acts directly and indirectly to degrade the components of the extracellular matrix (ECM). This process is likely to be important in the normal turnover of the ECM of fetal membranes and in its premature weakening in preterm premature rupture of the fetal membranes. Quantitative Northern analysis and in situ hybridization have shown that the decidua expresses mRNA for tPA. However, the immunolocalized tPA protein was most strongly associated with the amnion and chorion, as was its receptor annexin II, suggesting that the amnion and chorion are the targets for decidual tPA. At term, decidual tPA expression was unaffected by labor, and the tPA receptor was elevated both before and after labor. At preterm, the converse was found: decidual tPA expression was significantly (p < 0. 05) up-regulated by labor, but the tPA receptor was not. The results suggest that the generation of plasmin at term would be controlled by an increased concentration of the tPA receptor in the amnion and chorion, whereas at preterm a pathological increase in plasmin would be generated by an overexpression of tPA, initiated by labor.

An autocrine/paracrine role of human decidual relaxin. II. Stromelysin-1 (MMP-3) and tissue inhibitor of matrix metalloproteinase-1 (TIMP-1).

Interstitial collagen types I and III are the predominant collagens in the amniotic and chorionic connective tissues. However, this matrix also contains proteoglycans, fibronectin, laminin, and elastin, which together with the collagens may undergo partial degradation prior to fetal membrane rupture at term. In this study, stromelysin (MMP-3) and tissue inhibitor of metalloproteinases-1 (TIMP-1) were immunolocalized in fetal membranes obtained at term prior to labor. MMP-3 stained the cells of the amniotic epithelium, fibroblasts and macrophages of the amniotic and chorionic matrix, and those of the chorionic cytotrophoblast; there was no staining in the maternal decidua. TIMP-1 showed a similar staining pattern, except that the staining was darker in some amniotic epithelial cells and was present in the maternal decidua. The maternal decidua produces the two human relaxins H1 and H2; the latter, when incubated with explants of human fetal membranes, caused a dose-dependent and significant increase in expression of the MMP-3 gene and its secreted protein into the media. A significant effect of relaxin H2 on 92-kDa gelatinase (MMP-9) gene expression was also shown--an effect requiring poly(A)+ RNA rather than total RNA. Both relaxin H1 and H2 caused a significant increase in secretion of MMP-9 protein and its enzyme activity in the media. The magnitude of the effects of the two relaxins was similar, in contrast to findings from other biological studies in which relaxin H2 was shown to be more active. Neither of the relaxins had any effect on 72-kDa gelatinase (MMP-2) activity or on the TIMP-1 protein or its activity. This study suggests that local relaxins may be involved in the degradation of the complex fetal membrane extracellular matrix and may cause activation of an enzyme cascade resulting in fully activated MMP-9. Such effects could be important in the degradative pathways occurring in the amnion and chorion in the peripartal period.