High iNOS mRNA and protein levels during early third trimester suggest a role for NO in prelabor cervical ripening in the bovine

Nitric oxide (NO) plays a key role in the processes leading to cervical softening prior to labor. Inducible nitric oxide synthase (iNOS) contributes most to the increased production of NO during labor, as demonstrated in the rat cervix, or at term pregnancy in women. Changes in expression of iNOS during late gestation have not yet been studied longitudinally in any species, because repeatedly taking biopsies could not be performed. iNOS mRNA (n = 6) and protein expression (n = 3) in serial cervical biopsies of pregnant pluriparous cows taken around days 225, 250, and 275 of pregnancy and within 1.5 hr after calving (d225, d250, d275 and parturition biopsies, respectively) were measured using quantitative RT-PCR and Western blotting. iNOS mRNA expression decreased from the d225 biopsy onwards, differences being significant between the d250 and d275 (P < 0.05) and between the d275 and parturition biopsies (P < 0.05). iNOS protein expression decreased from d225 to d250 onwards. Immunohistochemical analysis of biopsies showed, besides positive staining in endothelium and epithelium, which remained unchanged at different time points, that iNOS expressing cells in the connective tissue cells of early biopsies were predominantly spindle shaped (mostly smooth muscle cells and some fibroblasts). In the parturition biopsies, iNOS reactivity was mainly found in mononuclear leucocytes. These results lead us to suggest that iNOS from spindle shaped cells is involved in prepartum cervical ripening, while iNOS in mononuclear inflammatory cells may be important for local tissue repair mechanisms during postpartum cervical involution.     

MMP-2 expression precedes the final ripening process of the bovine cervix

Collagen is denatured in the gradual cervical ripening process during late pregnancy, already before the onset of final cervical ripening at parturition. Matrix Metallo Proteinases (MMPs) might be responsible for this process. To investigate the presence and potential function of MMPs at the different stages of the ripening process, serial cervical biopsies were obtained from 10 cows at Days 185 and 275 of pregnancy (approximately 5 days before calving), at parturition and at 30 days after parturition. The mRNA and protein expression of MMP-1, MMP-2, and MMP-9 and of the tissue inhibitors of MMPs (TIMP)-1 and TIMP-2 were semi-quantitatively determined using RT-PCR, respectively, zymography, Westernblot, and ELISA techniques and the localization of MMP-2 protein and presence of granulocytes by immunohistochemistry and Luna staining. At parturition compared to 185 days pregnancy the MMP-1 protein expression and the numbers of granulocytes were significantly increased by 3 and 26-fold respectively. MMP-2 mRNA and protein expression had already increased 2.5 (P < 0.05) and twofold (P < 0.05) at 5 days before parturition, prior to final ripening. At that time, MMP-2 was present in smooth muscle cells and extra cellular matrix. TIMP-1 mRNA expression was significantly increased at parturition and TIMP-2 mRNA expression peaked at 5 days before parturition. The increased expression of MMP-2 at 5 days before parturition, suggests that in the cow MMP-2 is responsible for collagen denaturation in the last part of gradual cervical ripening, while MMP-1 and MMP-9 are only active during the final cervical ripening process at parturition.     

High iNOS mRNA and protein localization during late pregnancy suggest a role for nitric oxide in mouse pubic symphysis relaxation

Remodeling and relaxation of the mouse pubic symphysis (PS) are central events in parturition. The mouse PS remodels in a hormone-controlled process that involves the modification of the fibrocartilage into an interpubic ligament (IpL), followed by its relaxation prior to parturition. It is recognized that nitric oxide synthase (NOS) and consequently nitric oxide (NO) generation play important roles in extracellular matrix modification, and may promote cytoskeleton changes that contribute to the remodeling of connective tissue, which precedes the onset of labor. To our knowledge, no studies thus far have investigated inducible nitric oxide synthase (iNOS) expression, protein localization, and NO generation in the mouse PS during pregnancy. In this work, we used a combination of the immunolocalization of iNOS, its relative mRNA expression, and NO production to examine the possible involvement of iNOS in remodeling and relaxation of the mouse IpL during late pregnancy. The presence of iNOS was observed in chondrocytes and fibroblast-like cells in the interpubic tissues. In addition, iNOS mRNA and NO production were higher during preterm labor on Day 19 of pregnancy (D19) than NO production on D18 or in virgin groups. The significant increase in iNOS mRNA expression and NO generation from the partially relaxed IpL at D18 to the completely relaxed IpL at D19 may indicate that NO plays an important role in late pregnancy during relaxation of the mouse IpL.     

Endocrine regulation of cervical ripening in humans--potential roles for gonadal steroids and insulin-like growth factor-I

Cervical softening is crucial for a normal parturition and corresponds to remodeling of the dominating cervical extra cellular matrix (ECM). The onset of labor as well as cervical ripening is under hormonal control. To get further information about the endocrine regulation of term cervical ripening the following study was undertaken: cervical biopsies were obtained vaginally at elective caesareans, after normal vaginal delivery and after PGE2 or antiprogestin RU486. Biopsies from non-pregnant women served as controls. The concentrations of estrogen receptor (ER) and progesterone receptor (PR) protein were quantitated by EIA and the mRNA levels by solution hybridization. The ERalpha and beta were localized by immunohistochemistry, identified by RT-PCR and quantitated by solution hybridization. The co-localizations of CD45 (leukocyte antigen) and CD68 (macrophage antigen) were studied by immunohistochemistry. The cervical concentrations of ER and PR proteins decreased at term to 15 and 25%, respectively, compared to the non-pregnant levels. A further decrease was measured in the maximal ripened cervix at parturition. The mRNA levels were unchanged but IGF-I mRNA reached a maximum at term. ERalpha mRNA was significantly decreased until delivery, whereas ERbeta mRNA, like IGF-I; was maximum at term. By immunostaining ERbeta could be co-localized with CD45 leukocyte antigen and CD68 macrophage specific antigen. Oral administration of RU486 induced a significant increase in ER protein concentration, whereas PGE2 and spontaneous ripening did not. These findings indicate that cervical ripening is related to significant local hormonal changes.     

Immunolocalization of adipocytes and prostaglandin E2 and its four receptor proteins EP1, EP2, EP3, and EP4 in the caprine cervix during spontaneous term labor

The mechanisms of cervical ripening and dilation in mammals remain obscure. Information is lacking about the localization of prostaglandin E(2) (PGE(2))-producing cells and PGE(2) receptors (EP) in intrapartum cervix and whether cervical dilation at parturition is an active process. To reveal these mechanisms, immunolocalization of EP1-EP4 (official gene symbols PTGER1-PTGER4) and PGE(2)-producing cells in caprine cervix during nonpregnancy, pregnancy, and parturition was assayed by immunohistochemistry (IHC); the mRNA expression levels of PTGS2, PTGER2 (EP2), and PTGER4 (EP4) were determined using quantitative PCR; and the existence of adipocytes in the cervix at various stages was demonstrated with Oil Red O staining and IHC of perilipin A. The results suggested that in intrapartum caprine cervix staining of the PGE(2) was observed in the overall tissues, for example, blood vessels, canal or glandular epithelia, serosa, circular and longitudinal muscles, and stroma in addition to adipocytes; EP2 was detectable in all the tissues other than glandular epithelia; EP4 was strongly expressed in all the tissues other than serosa; EP1 was detected mainly in arterioles and canal or glandular epithelia; and EP3 was poorly expressed only in stroma, canal epithelia, and circular muscles. Little or no expression of EP2, EP3, and EP4 as well as PGE(2) in all cervical tissues was observed during nonpregnancy and pregnancy except for the strong expression of EP1 in canal or glandular epithelia during pregnancy. The mRNA expression levels of PTGS2, PTGER2, and PTGER4 were significantly higher in intrapartum than nonpregnant and midpregnant cervices (P < 0.01). Adipocytes appear only in the intrapartum cervix. These results support the concept that PGE(2) modulates specific functions in various anatomical structures of the caprine cervix at labor and the appearance of adipocytes at labor is likely related to caprine cervical dilation.     

In Vivo Evaluation of Cervical Stiffness Evolution during Induced Ripening Using Shear Wave Elastography,Histology and 2 Photon Excitation Microscopy: Insight from an Animal Model

Prematurity affects 11% of the births and is the main cause of infant mortality. On the opposite case, the failure of induction of parturition in the case of delayed spontaneous birth is associated with fetal suffering. Both conditions are associated with precocious and/or delayed cervical ripening. Quantitative and objective information about the temporal evolution of the cervical ripening may provide a complementary method to identify cases at risk of preterm delivery and to assess the likelihood of successful induction of labour. In this study, the cervical stiffness was measured in vivo in pregnant sheep by using Shear Wave Elastography (SWE). This technique assesses the stiffness of tissue through the measurement of shear waves speed (SWS). In the present study, 9 pregnant ewes were used. Cervical ripening was induced at 127 days of pregnancy (term: 145 days) by dexamethasone injection in 5 animals, while 4 animals were used as control. Elastographic images of the cervix were obtained by two independent operators every 4 hours during 24 hours after injection to monitor the cervical maturation induced by the dexamethasone. Based on the measurements of SWS during vaginal ultrasound examination, the stiffness in the second ring of the cervix was quantified over a circular region of interest of 5 mm diameter. SWS was found to decrease significantly in the first 4–8 hours after dexamethasone compared to controls, which was associated with cervical ripening induced by dexamethasone (from 1.779 m/s ± 0.548 m/s, p < 0.0005, to 1.291 m/s ± 0.516 m/s, p < 0.000). Consequently a drop in the cervical elasticity was quantified too (from 9.5 kPa ± 0.9 kPa, p < 0.0005, to 5.0 kPa ± 0.8 kPa, p < 0.000). Moreover, SWE measurements were highly reproducible between both operators at all times. Cervical ripening induced by dexamethasone was confirmed by the significant increase in maternal plasma Prostaglandin E2 (PGE2), as evidenced by the assay of its metabolite PGEM. Histological analyses and two-photon excitation microscopy, combining both Second Harmonic Generation (SHG) and Two-photon Fluorescence microscopy (2PF) contrasts, were used to investigate, at the microscopic scale, the structure of cervical tissue. Results show that both collagen and 2PF-active fibrillar structures could be closely related to the mechanical properties of cervical tissue that are perceptible in elastography. In conclusion, SWE may be a valuable method to objectively quantify the cervical stiffness and as a complementary diagnostic tool for preterm birth and for labour induction success.     

Sensory nerves and neuropeptides in uterine cervical ripening

At the time of parturition (fetal delivery) the uterine cervix must "ripen," becoming soft, pliable, and dilated to accommodate the fetus' delivery. The fundamental processes underlying cervical ripening remain poorly understood. Knowledge that abundant autonomic and sensory nerves supply the uterine cervix, that transection of afferent nerves supplying the cervix blocks parturition, and that some of the changes in the cervix resemble those seen in inflammatory reactions suggests nerves may have a role in the cervical ripening changes. The present study utilized immunohistochemistry, plasma extravasation, and solution hybridization-nuclease protection assay to elucidate the complement of primary afferent nerves and some receptors in the rat cervix during pregnancy, and to determine if they may have roles in the ripening process at term. This study revealed an abundance of nerves associated with the cervical vasculature and myometrial smooth muscle containing immunoreactivity for substance P, calcitonin gene-related peptide, secretoneurin, and nitric oxide synthase throughout pregnancy. Many of these are small unmyelinated capsaicin-sensitive C-fibers. Substance P- (NK1-) and calcitonin gene-related peptide receptors were apparent on uterine cervix vasculature from pregnant, parturient, and postpartum rats. NK1 receptor mRNA was maximal at 20 days of pregnancy. Plasma extravasation of i.v. administered Evans Blue or Monastral Blue was most pronounced at parturition (shortly after NK1 mRNA is maximal); this was similar to plasma extravasation evoked by i.v. administration of substance P or capsaicin-treatment. This study revealed new data about the nervous system of the rat uterine cervix and that these nerves and their transmitters could very well be part of a neurogenic inflammatory process involved in cervical ripening.     

Interleukin-8 release from human gestational tissue explants: effects of gestation, labor, and chorioamnionitis.

Interleukin-8 (IL-8) is a chemotactic cytokine that has been implicated in the process of human parturition, including the processes of cervical ripening and rupture of fetal membranes. In this study, the in vitro release of IL-8 from human amnion, choriodecidua, and placenta tissues obtained before and after spontaneous labor onset both at term and preterm, was assessed. The effect of chorioamnionitis on IL-8 release was also established. All tissue explants examined released IL-8; however, IL-8 release from choriodecidual explants was significantly (p < 0.02) greater than that observed from amnion and placenta. Furthermore, choriodecidual IL-8 release was significantly (p < 0.001) greater from term tissues (850 +/- 134.4 ng/mg DNA, n = 18) than from preterm tissues (458 +/- 68.8 ng/mg DNA, n = 17). Spontaneous onset of labor, irrespective of the eventual mode of delivery, was not associated with any significant changes in IL-8 release from human gestational tissues compared to not-in-labor tissues, both at term and preterm. IL-8 release from gestational tissues was not significantly different in the absence or presence of chorioamnionitis. These data are in contrast to the previously reported stimulatory effects of bacterial endotoxin on IL-8 release from human gestational tissues. The data are consistent, however, with the suggestion that IL-8 release is an early event in chorioamnionitis that precedes the appearance of clinically overt symptoms.     

Progesterone represses interleukin-8 and cyclo-oxygenase-2 in human lower segment fibroblast cells and amnion epithelial cells

Labor is preceded by cervical ripening through upregulation of interleukin (IL)-1beta, IL-8, and increased prostaglandin synthesis via inducible type 2 cyclooxygenase (COX-2). Progesterone maintains myometrial quiescence during pregnancy. In this study, we examined the effects of IL-1beta and progesterone on IL-8 and prostaglandin E2 (PGE2) synthesis and IL-8 and COX-2 mRNA and promoter activity in amnion cells and lower segment fibroblast (LSF) cells. In both cell types, progesterone had no effect on basal IL-8 or PGE2 synthesis. In LSF cells, IL-1beta significantly increased IL-8 and PGE2 synthesis and COX-2 and IL-8 mRNA expression, but progesterone significantly attenuated these effects. In prelabor amnion cells, IL-1beta also increased IL-8 and PGE2 synthesis and both COX-2 and IL-8 mRNA and promoter expression; however, progesterone significantly attenuated these effects on IL-8 and PGE2 synthesis and COX-2 expression. In postlabor amnion cells, IL-1beta increased IL-8 and PGE2 synthesis and COX-2 expression, but progesterone did not attenuate the effect of IL-1beta upon IL-8 synthesis. Progesterone repression of IL-8 and COX-2 in LSF cells suggests that IL-8 and COX-2 have similar regulatory mechanisms in LSF cells and that progesterone may play a role in maintenance of cervical competence. The lack of effect of progesterone on IL-8 in postlabor cells may be the result of downregulation of the progesterone receptor during labor.     

Mono-(2-Ethylhexyl) Phthalate Promotes Pro-Labor Gene Expression in the Human Placenta

Women exposed to phthalates during pregnancy are at increased risk for delivering preterm, but the mechanism behind this relationship is unknown. Placental corticotropin-releasing hormone (CRH) and cyclooxygenase-2 (COX-2) are key mediators of parturition and are regulated by the non-canonical NF-kB (RelB/p52) signaling pathway. In this study, we demonstrate that one of the major phthalate metabolites, mono-(2-ethylhexyl)-phthalate (MEHP), increased CRH and COX-2 mRNA and protein abundance in a dose-dependent manner in primary cultures of cytotrophoblast. This was coupled with an increase in nuclear import of RelB/p52 and its association with the CRH and COX-2 promoters. Silencing of NF-kB inducing kinase, a central signaling component of the non-canonical NF-kB pathway, blocked MEHP-induced upregulation of CRH and COX-2. These results suggest a potential mechanism mediated by RelB/p52 by which phthalates could prematurely induce pro-labor gene activity and lead to preterm birth.     

Identification of 9 uterine genes that are regulated during mouse pregnancy and exhibit abnormal levels in the cyclooxygenase-1 knockout mouse

Background  

Preterm birth is the leading cause of all infant mortality. In 2004, 12.5% of all births were preterm. In order to understand preterm labor, we must first understand normal labor. Since many of the myometrial changes that occur during pregnancy are similar in mice and humans and mouse gestation is short, we have studied the uterine genes that change in the mouse during pregnancy. Here, we used microarray analysis to identify uterine genes in the gravid mouse that are differentially regulated in the cyclooxygenase-1 knockout mouse model of delayed parturition.     

Inhibition of cyclooxygenase-2 prevents inflammation-mediated preterm labor in the mouse

Prostaglandins (PGs) have proven important during parturition, but inhibition of PG production treating preterm labor (PTL) results in significant maternal and fetal side effects. We hypothesize that specific inhibition of either cyclooxygenase (COX)-1 or -2 may result in separation of therapeutic and toxic effects. We demonstrate that COX-2, but not COX-1, is induced during inflammation-mediated PTL caused by lipopolysaccharide (LPS) administration. A two- to threefold increase in uterine and ovarian PG concentrations coincides with this induction of COX-2. The COX-2-selective inhibitor SC-236 proved effective in stopping preterm delivery and the increases in PGs. The COX-1-selective inhibitor SC-560 also attenuated uterine and ovarian PG production after LPS but did not inhibit PTL as efficiently as SC-236. COX-1-deficient mice, which show delay in the onset of term labor, exhibited no delay in onset of PTL after LPS. These findings suggest that the mechanisms for initiation of inflammation-mediated PTL and term labor differ and that selective COX-2 inhibition may provide a means of stopping inflammation-induced PTL in humans.     

Fetal Fibronectin Signaling Induces Matrix Metalloproteases and Cyclooxygenase-2 (COX-2) in Amnion Cells and Preterm Birth in Mice

Fetal fibronectin (fFN) in cervical and vaginal secretions has been used as a predictor of preterm delivery. Here, we clarified the pathological function of fFN on cell type-specific matrix metalloproteinases (MMPs) and prostaglandin synthesis in fetal membranes. Treatment of amnion mesenchymal cells with fFN resulted in dramatic increases in MMP-1 and MMP-9 mRNA and enzymatic activity as well as COX-2 mRNA and prostaglandin E₂ synthesis, activating both NFκB and ERK1/2 signaling. Fetal FN-induced increases in MMPs and COX-2 were mediated through its extra domain A and Toll-like receptor 4 expressed in mesenchymal cells. Lipopolysaccharide and TNF-α increased the release of free FN in medium of amnion epithelial cells in culture. Finally, injection of fFN in pregnant mice resulted in preterm birth. Collectively, these results indicate that fFN is not only a marker of preterm delivery but also plays a significant role in the pathogenesis of preterm labor and premature rupture of fetal membranes.