No abstract

BACKGROUND Enhanced proliferation and survival of eutopic endometrial cells from patients with endometriosis compared with healthy women is associated with abnormal activation of extra-cellular signal-regulated kinases 1 and 2 (ERK1/2). Given the role of Ras/Raf/mitogen-activated protein kinase (MAPK) and RhoA/ROCKII signalling pathways in the regulation of cell proliferation and migration, we analysed their possible roles in endometriosis. METHODS Primary eutopic endometrial stromal cells of patients with endometriosis (Eu-hESC, n= 16) and endometriosis-free controls (Co-hESC, n= 14) were harvested and subjected to proliferation and migration assays as well as kinase activity assays and immunoblot analysis of proteins from the Ras/Raf/MAPK and RhoA/ROCKII signalling pathways. Effects of ROCKII (Y-27632) and MAPK (U0126) inhibitors or siRNA knockdown of ROCKII, Raf-1 and B-Raf were analysed. RESULTS The proliferation rate of Eu-hESC was 54% higher than Co-hESC. Eu-hESC also displayed a 75% higher migration rate than Co-hESC. Eu-hESC displayed higher levels of ERK phosphorylation (83%) and p27 expression (61%) and lower levels of Raf-1 protein (47%) compared with controls. In addition to an inhibitory effect on cell proliferation, ROCKII knockdown led to significant down-regulation of cyclinD1 and p27 but did not affect ERK phosphorylation. Down-regulation of Raf-1 by siRNA was dispensable for cell proliferation control but led to an increase in ROCKII activity and a decrease in cell migration. B-Raf was shown to act as a regulator of hESC proliferation by modulating cellular ERK1/2 activity and cyclinD1 levels. Eu-hESC displayed 2.4-fold higher B-Raf activity compared with Co-hESC and therefore exhibit abnormally activated Ras/Raf/MAPK signalling. CONCLUSIONS We show that the same molecular mechanisms operate in Co- and Eu-hESC. The differences in cell proliferation and migration between both cell types are likely due to increased activation of Ras/Raf/MAPK and RhoA/ROCKII signalling pathways in cells from endometriosis patients.

The mechanisms that ultimately regulate cervical ripening during parturition remain largely unknown. A possible role for nitric oxide (NO) has recently emerged; however, the expression of NO synthase (NOS) within the human cervix in the ripening process has not been investigated. The purpose of this study was to identify cell types in the human cervix that contain NOS isoforms and to examine changes in their expression during the ripening process and the nonpregnant state. Inducible NOS (iNOS) immunoreactivity was observed in the epithelial cells and stromal spindle cells in 17 of 20 biopsies from cervices obtained within 10 min postpartum, but in only 4 of 12 nonpregnant controls (p = 0.03). Endothelial NOS (eNOS) immunoreactivity was restricted to vascular endothelia in all sections, whereas neuronal NOS was not detectable. Inducible NOS activity in the postpartum group was 3.2 times that of the control group (p = 0.0005), whereas constitutive NOS activity remained unchanged in both groups (p = 0.222). Competitive reverse transcription-polymerase chain reaction revealed no differences in the expression of iNOS (p = 0.443) or eNOS mRNA (p = 0.409). The existence of iNOS in the human postpartum cervix suggests that increased production of NO, probably induced by cytokines, may be relevant to the process of natural cervical ripening in humans.

Expression of inducible nitric oxide synthase (iNOS) by tumor cells has been suggested to abrogate metastasis in several tumor models, whereas constitutive NOS expression correlated positively with tumor grade in human breast carcinoma. Whether or not expression of one of the various NOS isoforms could predict the prognosis of breast cancer, however, has not been established. In the present report we investigated the cellular distribution of NOS isoforms in a series of benign and malignant breast tumors and in normal breast tissue. Immunohistochemistry revealed that in samples of benign disease the number of iNOS+ epithelial cells or total epithelial cells was 69+/-16% (n = 50). In samples of grade II invasive ductal breast carcinomas the number of iNOS+ tumor cells or total tumor cells was 62+/-20% (n = 40), compared to 12+/-9% (n = 40) in samples of grade III carcinomas (P<0.0001). iNOS protein was also identifiable in most of the epithelial cells of normal breast tissue (n = 4). In contrast, eNOS protein was restricted to vascular endothelial cells in all of the specimens studied. Since the presence of tumor cell iNOS protein is inversely related to the tumor's metastatic potential, we conclude that endogenous tumor cell mediated iNOS expression might have an inhibitory effect on the metastatic process in breast cancer.

Estrogen-induced loss of estrogen receptor (ER) alpha expression limits estrogen responsiveness in many target cells. However, whether such a mechanism contributes to changes in vascular endothelial ER alpha and/or ER beta levels is unclear. Using RT-PCR assays, we did not find any regulation of ER alpha or ER beta mRNA expression in human uterine artery endothelial cell (HUAEC) nuclear extracts on stimulation with 17 beta-estradiol for 1 or 2 h. By contrast, Western analysis on HUAEC extracts revealed that 17 beta-estradiol was capable of down-regulating both ER alpha and ER beta protein starting 1 h after treatment, an effect that can be blocked by pretreatment with tamoxifen as well as with the proteasome inhibitor lactacystin. The proteolysis inhibitors insulin, cycloheximide, and puromycin impede ER alpha, but not ER beta, turnover. Ubiquitin, but not its competitive inhibitor methyl-ubiquitin, induces rapid turnover of both ERs in a cell-free system of MCF-7 and HUAEC extracts. We, thus, propose the existence of estrogen-induced ER degradation that serves to control physiological responses in an estrogen target tissue, i.e. human vascular endothelium, by down- regulating ER alpha as well as ER beta through different proteasomal uptake mechanisms.