Identification of prometalloproteinase-3 as a major protein secreted by human endometrial fibroblasts and inhibited by coculture with trophoblast cells.

To assess endometrial fibroblast-cytotrophoblast interactions, we used a coculture system allowing analysis of the potential cell morphology modifications and protein secretion variations possibly involved in endometrial invasion arrest. Stromal cells and cytotrophoblasts were isolated from endometrial biopsies and first-trimester placental villi, respectively. In our culture conditions, a 57-kDa protein that was secreted by cultured fibroblasts but was absent in the 4-day coculture medium was found to be identical to prometalloproteinase-3 (proMMP-3) through determination of amino acid sequences of NH2-terminal and internal peptides. Northern blotting analysis of endometrial fibroblast total RNA showed a 38.6% metalloproteinase-3 (MMP-3) mRNA inhibition by 4-day 10(-6) M R5020 treatment. Inhibition of proMMP-3 secretion was weak when cytotrophoblasts were cultured for 4 days in a polycarbonate membrane insert over cultured fibroblasts without possible cell contact in spite of high levels of progesterone produced by cytotrophoblasts. Furthermore, cytotrophoblasts cultured on a monolayer of endometrial fibroblasts became syncytia, and most of the fibroblasts were decidualized. The closeness of the two cell types allowed paracrine relationships that might facilitate the progesterone action. Since MMP-3 is known to activate collagenases, inhibition of its secretion by cell contact might be a mechanism of invasion arrest for trophoblast cell migration.

Estrogens increase the expression of fibulin-1, an extracellular matrix protein secreted by human ovarian cancer cells.

Ovarian cancers have a high ability to invade the peritoneal cavity and some are stimulated by estrogens. In an attempt to understand the mode of action of estrogens on these cancer cells and to develop new markers, we have characterized estrogen-regulated proteins. This study was aimed at identifying a protein secreted by ovarian cancer cells whose level was increased by estradiol [Galtier-Dereure, F., Capony, F., Maudelonde, T. & Rochefort, H. (1992) J. Clin. Endocrinol. Metab. 75, 1497-1502]. By using microprotein sequencing, the 110-kDa protein was identified as fibulin-1, a protein of the extracellular matrix that binds to fibronectin, laminin, and nidogen. The amount of immunoprecipitated fibulin-1 secreted into the medium and present in the cell extract was increased up to 10-fold by estradiol in three estrogen-responsive ovarian cancer cell lines. By immunohistochemistry fibulin-1 was located in the stroma of several ovarian cancers and cysts. The findings highlight a potential role for fibulin-1 in the spread of ovarian cancer in the peritoneal cavity and/or in distal metastases.

In vivo stimulation by tamoxifen of cathepsin D RNA level in breast cancer.

We have previously shown that 3 weeks of treatment with tamoxifen, of patients with primary breast carcinomas, increased cytosolic cathepsin D protein in oestrogen receptor (ER) positive tumours [Maudelonde et al., Cancer 1989, 63, 1265-1270]. In order to investigate the mechanism of this increase and to eliminate a transient flare-up effect, we semi-quantified cathepsin D RNA levels by in situ hybridisation in 32 breast carcinomas from patients treated with tamoxifen for 3 weeks prior to surgery and in 35 breast cancer patients receiving no tamoxifen. We found that tamoxifen increased cathepsin D RNA level regardless of the ER status of the tumours. In ER positive tumours, tamoxifen increased the cathepsin D RNA level to the same extent as cytosolic cathepsin D protein but not in ER negative tumours. The induction of cathepsin D RNA by tamoxifen in ER positive tumours was probably due to its agonist activity, also observed in vitro in breast cancer cell lines. These results suggest that the cathepsin D gene is inducible by oestrogens in ER positive breast cancer as it is in breast cancer cell lines.

Cathepsin D assay in primary breast cancer and lymph nodes: relationship with c-myc, c-erb-B-2 and int-2 oncogene amplification and node invasiveness.

In breast cancer, axillary lymph node invasiveness is the major prognostic factor in predicting relapse and metastasis. Nevertheless, since 30% of node-negative tumors also relapse, it is necessary to develop other independent prognostic factors. Oncogene amplification and the level of cathepsin D (cath-D), an acidic lysosomal protease produced and secreted in excess by breast cancer cells, have been proposed as additional prognostic factors. We have compared the cytosolic cath-D level and the amplification of three oncogenes: c-myc, neu-erb-B-2 and int-2 in 140 primary breast carcinomas and 64 axillary lymph nodes collected in 1987 and 1988 at the Cancer Center of Montpellier (Centre Paul Lamarque). None of the patients had previously received hormonal or chemotherapy. The cath-D concentration was measured with an immunoradiometric assay using monoclonal antibodies. DNA purified from the same samples was analyzed by a standard Southern blotting technique to estimate oncogene amplification. No correlation was found between the level of cath-D in the tumor and node invasiveness. Using a cut-off level of 60 pmol/mg protein, the status of cath-D was not correlated with neu-erb-B-2 and int-2 amplification and only correlated with c-myc amplification (P = 0.011). Both c-myc and cath-D are associated with cell proliferation, induced by estrogens in ER+ breast cancer, and constitutively produced in ER- breast cancer. The level of cath-D was significantly higher in the invaded lymph nodes (P = 0.04) than in the histologically non-invaded ones. Nevertheless, some non-invaded lymph nodes contained a high level of cath-D, as confirmed by immunoperoxidase staining. In conclusion, in breast cancer, a high cytosolic cath-D concentration is more frequent in tumors with c-myc amplification but is dissociated from neu-erb-B-2 or int-2 amplification, suggesting that the determination of these three markers will have an additional prognostic value.

Cathepsin D: an independent prognostic factor for metastasis of breast cancer.

122 patients with primary breast cancer were followed-up for a median of 4.6 years after surgery. The concentration of cathepsin D in tumour cytosol was strongly related to both metastasis-free survival and disease-free survival and was independent of nine conventional prognostic indices. Cathepsin D assay may prove particularly useful in identifying women who, though without lymph node involvement at presentation, are at high risk of metastatic disease.

Differential regulation of cathepsin D by sex steroids in mammary cancer and uterine cells.

The precursor of cathepsin D, a lysosomal acidic protease, is secreted by human breast cancer cells, where its synthesis is specifically induced by estrogens and growth factors. In this study, we investigated the hormonal regulation of cathepsin D and its mRNA in uterine cells. In the Ishikawa endometrial cancer cell line, epidermal growth factor (EGF) increased the level of cathepsin D and its mRNA 2- to 3-fold. Although expression of the transiently transfected estrogen-responsive recombinant (Vit. tk. CAT) and the endogenous progesterone receptor was markedly increased by estradiol in Ishikawa cells, estradiol did not alter the level of cathepsin D or its mRNA. The progestin R5020 induced the expression of the LTR sp65 CAT, which contains the progesterone-responsive element of the MMTV but it too was without effect on cathepsin D. By contrast, the expression of cathepsin D gene, in normal rat uterus, was increased by R5020 but not by estradiol. We conclude that cathepsin D gene expression is regulated differently by sex steroid hormones in endometrial and breast cancer cell lines, whereas it is similarly induced by EGF in these cells.