Growth inhibition by danazol in a human endometrial cancer cell line with estrogen-independent progesterone receptors

Since we recently found that danazol, an isoxazol derivative of ethinyltestosterone, has a growth-inhibitory effect on human endometrial cancer cells in primary culture, the effects of danazol on a human endometrial cancer cell line (IK-90 cells), which contains estrogen-independent progesterone receptors (PR), were investigated in the present study. The addition of danazol (1 nM-1 microM) in culture medium caused a decrease in the growth of IK-90 cells in a dose-dependent manner. Competitive binding studies showed that danazol effectively binds to PR in IK-90 cells, and the binding affinity for PR was estimated to be 6.0% of that of R5020. The addition of 1 microM danazol in culture medium resulted in a rapid and significant increase in nuclear PR with a concomitant decrease in cytoplasmic PR in the cells. These findings suggest that danazol has a growth-inhibitory effect on human endometrial adenocarcinoma cells directly through PR system in the cells.     

RU486, a progestin antagonist, binds to progesterone receptors in a human endometrial cancer cell line and reverses the growth inhibition by progestins

The human endometrial cancer cell line, IK-90 cells, contains estrogen-independent progesterone receptors (PR) and is progestin sensitive. Accumulation of glycogen in the cytoplasm of IK-90 cells as well as growth inhibition of the cells in response to progestins are observed. In the present study, the effects of RU486, a progestin antagonist, on IK-90 cells were investigated in a serum-supplemented medium. Scatchard plot analysis of cytoplasmic binding data in the cells revealed a high affinity binding site for RU486 (Kd, 2.6 nM) with maximum binding sites of 169 fmol/mg protein. However, the binding ability to DNA-cellulose of heat activated [3H]RU486-PR complexes was lower when compared with that of the progestin agonist [3H]R5020-PR complexes, suggesting a decrease in progestin activity of RU486 in IK-90 cells. The addition of 1 microM RU486 to culture medium produced periodic acid-Schiff-positive granules in the cytoplasm of the cells. On the other hand, RU486 (1 nM-1 microM) did not significantly inhibit the growth of cells. However, RU486 (0.1-1 microM) totally prevented the growth-inhibitory effect of R5020 (0.1-1 microM) on IK-90 cells. In conclusion, RU486, an antiprogestin, had a dual activity both a progestin antagonist and weak agonist in human endometrial cancer cells, which was not mediated through the estrogen receptor system.     

Molecular characterization of the microsomal tamoxifen binding site

Tamoxifen is a selective estrogen receptor modulator widely used for the prophylactic treatment of breast cancer. In addition to the estrogen receptor (ER), tamoxifen binds with high affinity to the microsomal antiestrogen binding site (AEBS), which is involved in ER-independent effects of tamoxifen. In the present study, we investigate the modulation of the biosynthesis of cholesterol in tumor cell lines by AEBS ligands. As a consequence of the treatment with the antitumoral drugs tamoxifen or PBPE, a selective AEBS ligand, we show that tumor cells produced a significant concentration- and time-dependent accumulation of cholesterol precursors. Sterols have been purified by HPLC and gas chromatography, and their chemical structures determined by mass spectrometric analysis. The major metabolites identified were 5alpha-cholest-8-en-3beta-ol for tamoxifen treatment and 5alpha-cholest-8-en-3beta-ol and cholesta-5,7-dien-3beta-ol, for PBPE treatment, suggesting that these AEBS ligands affect at least two enzymatic steps: the 3beta-hydroxysterol-Delta8-Delta7-isomerase and the 3beta-hydroxysterol-Delta7-reductase. Steroidal antiestrogens such as ICI 182,780 and RU 58,668 did not affect these enzymatic steps, because they do not bind to the AEBS. Transient co-expression of human 3beta-hydroxysterol-Delta8-Delta7-isomerase and 3beta-hydroxysterol-Delta7-reductase and immunoprecipitation experiments showed that both enzymes were required to reconstitute the AEBS in mammalian cells. Altogether, these data provide strong evidence that the AEBS is a hetero-oligomeric complex including 3beta-hydroxysterol-Delta8-Delta7-isomerase and the 3beta-hydroxysterol-Delta7-reductase as subunits that are necessary and sufficient for tamoxifen binding in mammary cells. Furthermore, because selective AEBS ligands are antitumoral compounds, these data suggest a link between cholesterol metabolism at a post-lanosterol step and tumor growth control. These data afford both the identification of the AEBS and give new insight into a novel molecular mechanism of action for drugs of clinical value.     

Microsomal binding sites for antioestrogens in rat liver. Properties and detergent solubilization.

The properties of an antioestrogen binding site (AEBS), which has high affinity and specificity for nonsteroidal antioestrogens and structurally related compounds, have been studied in rat liver microsomes. When subcellular organelles were separated on Percoll density gradients the distribution of the AEBS paralleled that of NADPH-cytochrome c reductase, indicating that the AEBS is associated with the endoplasmic reticulum. Saturation analysis showed that [3H]tamoxifen was bound to a single class of saturable binding sites in liver microsomes with a KD of 0.9 +/- 0.1 nM at 0 degrees C. The equilibrium KD was not significantly different at 22 degrees C. The KD calculated from the association and dissociation rate constants for [3H]tamoxifen binding at 0 degrees C and 22 degrees C was compatible with the KD measured at equilibrium. Ligand specificity studies using tamoxifen analogues showed qualitatively similar structure-affinity relationships for the AEBS from both rat liver and the MCF 7 breast cancer cell line. In general structural modifications caused correspondingly greater changes in affinity for rat liver AEBS than for MCF 7 AEBS. The AEBS was solubilized from microsomal membranes with sodium cholate. This was the only detergent of nine tested that solubilized the site in high yield without loss of activity. Solubilization using cholate was more effective in the presence of 1 M-NaCl. In the solubilized state there was an apparent loss of [3H]tamoxifen binding activity which could be restored by dilution of the detergent. Gel filtration indicated an Mr of 440,000-490,000 for the AEBS-cholate complex. These studies demonstrate that rat liver contains high concentrations of a microsomal AEBS which has similar properties and specificity to the AEBS previously described in human breast cancer cells. This site can be solubilized by sodium cholate to supply material suitable for further purification.     

Mechanisms of growth inhibition by nonsteroidal antioestrogens in human breast cancer cells

Treatment of MCF7 human mammary carcinoma cells with the nonsteroidal antioestrogens, tamoxifen and clomiphene, leads to a concentration-dependent decrease in cellular proliferation rate which can be resolved into oestrogen-reversible and oestrogen-irreversible components. This became more clearly apparent when cells were treated with the 4-hydroxylated derivatives of these compounds where, because of enhanced affinity for the oestrogen receptor (ER), the dose-response curves for the two components could be separated. Thus treatment with 4-hydroxyclomiphene resulted in a distinct biphasic effect on cell growth. In the concentration range 10(-10)-10(-8) M, cell proliferation was inhibited in a concentration-dependent manner to a maximum of 60-70%, there was no further effect between 10(-8) and 10(-6) M, but at concentrations greater than 10(-6) M there was another concentration-dependent decrease in cell growth. Studies with a series of vinyl-substituted hydroxytriphenylethylenes revealed that in the nanomolar concentration range, where the effects of the drugs could be completely negated by the simultaneous addition of oestradiol, the potency for growth inhibition was highly correlated with affinity for ER. Such data provide strong evidence that in this concentration range the growth inhibitory effects of nonsteroidal antioestrogens are mediated by the intracellular ER. In the micromolar concentration range the effects of antioestrogens are not completely reversed by oestradiol, potency is not well correlated with affinity for either ER or the antioestrogen binding site (AEBS) but the effect is cell cycle phase-specific. Furthermore, the disparity between the affinity for AEBS (0.8-3.3 nM) and the concentration of drug needed for oestrogen-irreversible growth inhibition (greater than or equal to 2.5 microM) argue against a central role for AEBS in mediating this effect. The observation that triphenylethylene antioestrogens are calmodulin antagonists may provide some insight into potential mechanisms for this oestrogen-irreversible effect. Indeed, in identical experiments two phenothiazine calmodulin antagonists inhibited MCF 7 cell proliferation at concentrations greater than or equal to 2.5 x 10(-6) M. Growth inhibition following administration of fluphenazine, perphenazine and triphenylethylene antioestrogens was accompanied by qualitatively similar changes in the cell cycle kinetic parameters, i.e. accumulation in G1 phase at the expense of S phase cells. These data suggest triphenylethylene antagonism of calmodulin activated cellular processes as a potential mechanism for the oestrogen-irreversible effects of the nonsteroidal antioestrogens.     

Src Drives Growth of Antiestrogen Resistant Breast Cancer Cell Lines and Is a Marker for Reduced Benefit of Tamoxifen Treatment

The underlying mechanisms leading to antiestrogen resistance in estrogen-receptor α (ER)-positive breast cancer is still poorly understood. The aim of this study was therefore to identify biomarkers and novel treatments for antiestrogen resistant breast cancer. We performed a kinase inhibitor screen on antiestrogen responsive T47D breast cancer cells and T47D-derived tamoxifen and fulvestrant resistant cell lines. We found that dasatinib, a broad-spectrum kinase inhibitor, inhibited growth of the antiestrogen resistant cells compared to parental T47D cells. Furthermore western blot analysis showed increased expression and phosphorylation of Src in the resistant cells and that dasatinib inhibited phosphorylation of Src and also signaling via Akt and Erk in all cell lines. Immunoprecipitation revealed Src: ER complexes only in the parental T47D cells. In fulvestrant resistant cells, Src formed complexes with the Human Epidermal growth factor Receptor (HER)1 and HER2. Neither HER receptors nor ER were co-precipitated with Src in the tamoxifen resistant cell lines. Compared to treatment with dasatinib alone, combined treatment with dasatinib and fulvestrant had a stronger inhibitory effect on tamoxifen resistant cell growth, whereas dasatinib in combination with tamoxifen had no additive inhibitory effect on fulvestrant resistant growth. When performing immunohistochemical staining on 268 primary tumors from breast cancer patients who had received tamoxifen as first line endocrine treatment, we found that membrane expression of Src in the tumor cells was significant associated with reduced disease-free and overall survival. In conclusion, Src was identified as target for treatment of antiestrogen resistant T47D breast cancer cells. For tamoxifen resistant T47D cells, combined treatment with dasatinib and fulvestrant was superior to treatment with dasatinib alone. Src located at the membrane has potential as a new biomarker for reduced benefit of tamoxifen.     

Toremifene-atamestane; alone or in combination: predictions from the preclinical intratumoral aromatase model

Since most breast cancers occur in postmenopausal women and are hormone dependent, we developed a model system that mimics this situation. In this model, tumors of human estrogen receptor (ER) positive breast cancer cells stably transfected with aromatase (Ac-1) are grown in immune-compromised mice. Using this model we have explored a number of therapeutic strategies to maximize the antitumor efficacy of antiestrogens (AEs) and aromatase inhibitors (AIs). This intratumoral aromatase xenograft model has proved accurate in predicting the outcome of several clinical trials. In this current study we compared the effect of an AE toremifene and steroidal AI atamestane, alone or in combination, on growth of hormone-dependent human breast cancer. We have also compared toremifene plus atamestane combination with tamoxifen in this study. The growth of Ac-1 cells was inhibited by tamoxifen, toremifene and atamestane in vitro with IC(50) values of 1.8+/-1.3 microM, 1+/-0.3 microM and 60.4+/-17.2 microM, respectively. The combination of toremifene plus atamestane was found to be better than toremifene or atamestane alone in vitro. The effect of this combination was then studied in vivo using Ac-1 xenografts grown in ovariectomized female SCID mice. The mice were injected with toremifene (1000 microg/day), atamestane (1000 microg/day), tamoxifen (100 microg/day), or the combination of toremifene plus atamestane. In this study, our results indicate that the combination of toremifene plus atamestane was as effective as toremifene or tamoxifen alone but may not provide any additional benefit over toremifene alone or tamoxifen alone.     

Effect of antiestrogens and aromatase inhibitor on basal growth of the human breast cancer cell line MCF-7 in serum-free medium

Antiestrogens are efficient inhibitors of estrogen-mediated growth of human breast cancer. Besides inhibiting estradiol-stimulated growth, antiestrogens may have a direct growth-inhibitory effect on estrogen receptor (ER) positive cells and thus be more efficient than aromatase inhibitors, which will only abrogate estrogen-dependent tumor growth. To address this issue, we have used the human breast cancer cell line MCF-7/S9 as a model system which is maintained in a chemically defined medium without serum and estrogen. The addition of estradiol results in an increase in cell growth rate. Thus, the MCF-7/S9 cell line is estrogen-responsive but not estrogen-dependent. Three different types of antiestrogens, namely tamoxifen, ICI 182,780 and EM-652 were found to exert a significant and dose-dependent inhibition of basal growth of MCF-7/S9 cells. The growth-inhibitory effect of the three antiestrogens was prevented by simultaneous estradiol treatment. Antiestrogen treatment also reduced the basal pS2 mRNA expression level, thus indicating spontaneous estrogenic activity in the cells. However, treatment with the aromatase inhibitor had no effect on basal cell growth, excluding that endogenous estrogen synthesis is involved in basal growth. These data demonstrate that in addition to their estrogen antagonistic effect, antiestrogens have a direct growth-inhibitory effect which is ER-mediated. Consequently, in the subset of ER positive breast cancer patients with estrogen-independent tumor growth, antiestrogen therapy may be superior to treatment with aromatase inhibitors which only inhibit estrogen formation but do not affect cancer cell growth in the absence of estrogens.     

Inhibitory effects of danazol and medroxyprogesterone acetate on [3H]thymidine incorporation in human endometrial cancer cells

Effects of medroxyprogesterone acetate (MPA) and danazol (1 nM-10 microM) on cultured cancer cells from human endometrial adenocarcinomas obtained by hysterectomy were simultaneously investigated. Of twenty-four endometrial adenocarcinomas examined, five tumors were successfully maintained in primary cell culture. The addition of MPA as well as danazol in culture of cells from the five tumors resulted in a significant inhibition of [3H]thymidine incorporation in cancer cells from three tumors having progesterone receptors (PR). The minimum effective concentrations of MPA and danazol for the inhibition of [3H]thymidine incorporation were found to be 10 and 100 nM, respectively. The difference in effective concentration could be explained by a higher affinity of MPA to PR than that of danazol in cancer cells. On the other hand, neither danazol nor MPA affected [3H]thymidine incorporation in cultured cells from the remaining two tumors, in which PR was absent in one but present in the other. These findings, together with our previous findings that danazol inhibited the growth of a human endometrial cancer cell line with PR, suggest that a growth-inhibitory effect of danazol on human endometrial cancer cells is mediated through PR in the cells.     

Role of estrogen receptors and antiestrogen binding sites in an early effect of antiestrogens, the inhibition of cholesterol biosynthesis

The effects of estradiol (E2), 4-hydroxy-tamoxifen (OH-Tam), and LY117018 on cholesterogenesis were investigated in two human breast cancer cell lines (MCF-7 and BT20), and in rat hepatoma (HTC) and fibroblastic (NRK-49F) cell lines. It was found that 10(-10) M E2 stimulated and 10(-8) M OH-Tam inhibited cholesterol synthesis in the estrogen-sensitive MCF-7 cell line. The OH-Tam effect occurred in less than 15 min whereas E2 only stimulated after 8 h. The inhibition of cholesterol synthesis was not reversed by E2. E2 was without effect in the HTC and estrogen-resistant BT20 cell lines whereas OH-Tam was as effective as in the MCF-7 cells. LY117018 had nearly as much effect on cholesterol synthesis as OH-Tam, in both MCF-7 and BT20 cells. Neither E2 nor OH-Tam had any effect on the NRK-49F cell line, even at micromolar concentrations. The three lines (MCF-7, BT20, HTC), whose cholesterol synthesis has been shown to be OH-Tam sensitive, appeared to contain high-affinity antiestrogen binding sites (AEBS); since the OH-Tam-resistant line (NRK) only contained low-affinity AEBS, there appears to be some relationship between OH-Tam sensitivity and high-affinity AEBS content. This suggests that the cholesterogenesis inhibition induced by antiestrogens is ER-independent and may involve AEBS. The cholesterogenesis stimulation induced by E2 occurred via a different pathway that appears to be related to the presence of ER in the cells.     

Can vitamin A modify the activity of docetaxel in MCF-7 breast cancer cells?

Docetaxel is one of the most effective chemotherapeutic agents in the treatment of breast cancer. On the other hand, the vitamin A family compounds play the essential roles in many biological processes in mammary gland. The aim of our study was to investigate the effect of all-trans retinol, carotenoids (beta-carotene, lycopene) and retinoids (9-cis, 13-cis and all-trans retinoic acid) on the activity of docetaxel and to compare these effects with the estradiol and tamoxifen actions on human ER(+) MCF-7 breast cancer cell line. The evaluation was based on [3H] thymidine incorporation and the proliferative activity of PCNA and Ki 67 positive cells. In our study, the incorporation of [3H] thymidine into cancer cells was inhibited to 50% by 0.2, 0.5 and 1 microM of docetaxel in the 24-hour culture and addition of estradiol (0.001 microM) didn't influence the results. However, addition of tamoxifen caused a statistically significant decrease of the percentage of the proliferating cells in the culture medium with 0.2 and 0.5 microM of docetaxel (38.99 +/- 2.84%, p<0.01 and 40.67 +/- 5.62%, p<0.01) in comparison to the docetaxel only group. The above-mentioned observations were also confirmed with the use of the immunohistochemical investigations. Among the examined vitamin A family compounds, the simultaneous application of beta-carotene (0.1 microM) and docetaxel (0.2 microM) resulted in a statistically significant reduction in the percentage of proliferating cells (40.25 +/- 14.62%, p<0.01). Lycopene (0.1 microM), which stimulates the growth of breast cancer cells in a 24-hour culture, had an inhibitory effect (42.97 +/- 9.58%, p<0.01) when combined with docetaxel (0.2 microM). Although, beta-carotene and lycopene belong to the different chemical groups, they surprisingly had a similar inhibitory influence on both growth and proliferation of MCF-7 breast cancer cells when combined with docetaxel. The application of docetaxel either with beta-carotene or lycopene had comparable inhibitory effect on breast cells growth and proliferation as tamoxifen. Therefore, it may suggest a possible important role of these carotenoids in the breast cancer therapy in women especially when docetaxel is applied.     

Signaling by estrogens and tamoxifen in the human endometrium

Tamoxifen is used as adjuvant treatment for postmenopausal breast cancer patients. The mechanism of action of tamoxifen in breast cancer patients is that tamoxifen inhibits growth of cancer cells by competitive antagonism for estrogens at the estrogen receptor (ER). In the endometrium, tamoxifen has an effect that varies with the ambient concentration of estrogen: in premenopausal women (high estrogen levels), tamoxifen displays an estrogen-antagonistic effect, while in postmenopausal women (low estrogen levels), tamoxifen displays an estrogen-agonistic mode of action. Here, using microarray technology we have compared estrogen signaling with tamoxifen signaling in the human endometrium. It was observed that on the one hand tamoxifen-treatment results in modulation of expression of specific genes (370 genes) and on the other hand tamoxifen-treatment results in modulation of a set of genes which are also regulated by estrogen treatment (142 genes). Upon focusing on regulation of proliferation, we found that tamoxifen-induced endometrial proliferation is largely accomplished by using the same set of genes as are regulated by estradiol. So, as far as regulation of proliferation goes, tamoxifen seems to act as estrogen agonist. Furthermore, tamoxifen-specific gene regulation may explain why tamoxifen-induced endometrial tumors behave more aggressively than sporadic endometrial tumors.     

Synthesis, binding and structure-affinity studies of new ligands for the microsomal anti-estrogen binding site (AEBS)

New compounds have been synthesized based on the structure of the anti-tumoral drug tamoxifen and its diphenylmethane derivative, N,N-diethyl-2-[(4-phenyl-methyl)-phenoxy]-ethanamine, HCl (DPPE). These new compounds have no affinity for the estrogen receptor (ER) and bind with various affinity to the anti-estrogen binding site (AEBS). Compounds 2, 10, 12, 13, 20a, 20b, 23a, 23b, 29 exhibited 1.1-69.5 higher affinity than DPPE, and compounds 23a and 23b have 1.2 and 3.5 higher affinity than tamoxifen. Three-dimensional structure analysis, performed using the intersection of the van der Waals volume occupied by tamoxifen in its crystallographic state and the van der Waals volume of these new compounds in their calculated minimal energy conformation, correlated well with their pKi for AEBS (r = 0.84, P<0.0001, n = 18). This is the first structure-affinity relationship (SAR) ever reported for AEBS ligands. Moreover in this study we have reported the synthesis of new compounds of higher affinity than the lead compounds and that are highly specific for AEBS. Since these compounds do not bind ER they will be helpful to study AEBS mediated cytotoxicity. Moreover our study shows that our strategy is a new useful guide to design high affinity and selective ligands for AEBS.     

Regulation of SBP synthesis in human cancer cell lines by steroid and thyroid hormones

The presence of human sex steroid binding protein (SBP) in liver cells, the supposed site of SBP synthesis, and in other target cells for sex steroid hormones such as breast, endometrium and prostate epithelium, have been demonstrated by indirect immunofluorescence. It is not known whether SBP enters endometrial and prostate cells by endocytosis, possibly mediated by a cell membrane receptor process, or if SBP is synthesized in these cells. SBP mRNA has been searched in human cancer cell lines originated from liver (Hep G2/H5A), breast (MCF-7), endometrium (RL95-2) and prostate (LNCaP). It was only found in hepatoma cells where it is regulated by estradiol, antiestrogen tamoxifen and triiodothyronine, in a similar way as secreted SBP. This work provides evidence that human SBP is synthesized in the liver, and it also suggests that its regulation may involve non-steroidal factors.     

Antiestrogen binding within different pituitary cell populations. Comparison with androgen and estrogen receptors

Gonadotrope-enriched populations were prepared from 42-day old male rats by centrifugal elutriation. They contained 4.8 +/- 0.7% of the cells, 51 +/- 10% of the LH and less than 3% of the PRL (n = 4). Gonadotrope-depleted fractions were also obtained that contained most of PRL cells. Specific antiestrogen binding sites (AEBS) were quantitated in these populations after destruction of estrogen receptor. Results showed the presence of a distinct, specific high affinity binding site for antiestrogen in dispersed pituitary cells and in enriched fractions. However, AEBS are not specific of a pituitary cell type. Thus, AEBS appear different from estrogen receptors in pituitary gland: by the thermal stability of AEBS, by the localization of AEBS in particulate material, by the uniform distribution of AEBS in different populations which differ markedly for E2 binding sites. Whereas the ratio of binding AE/E2 averaged 11.4 in the initial cell suspension it reached only 2.9 in the gonadotropes. The dissociation constants for AEBS were in the same range (1.16 - 2.27 X 10(-9) M) for the different populations.     

Cytotoxicity of tamoxifen in normal and tumoral cell lines and its ability to induce cellular transformation in vitro

Tamoxifen (TAM) is a non-steroidal anti-estrogen used to treat patients with estrogen receptor-positive breast cancer and as a chemopreventive agent against breast cancer in high risk pre- and post-menopausal women. However, recent studies have shown that tamoxifen causes endometrial and hepatic cancer. In this study, we examined the effects of tamoxifen (5, 10, 25 and 50 microM) on the growth and proliferation of nine tumoral cell lines (UACC62, MCF-7, NCI-460, K-562, OVCAR-03, PC-03, HT-29, 786-0, NCI-ADR) and non-tumoral cell lines (3T3, V79, MDCK, VERO). Chinese hamster lung fibroblasts (V79) were the most sensitive lineage to tamoxifen, with 21.6% of the cells showing apoptosis at 50 microM TAM. Microscopic analysis showed that, the cellular transformation caused by TAM in V79 cells was similar to that seen with 7,12-dimethylbenz(a)anthracene, thus indicating the carcinogenicity of TAM.     

Cell-associated insulin-like growth factor-binding proteins inhibit insulin-like growth factor-I-induced endometrial cancer cell proliferation.

Insulin-like growth factor I (IGF-I) is a peptidic growth factor implicated in the proliferation of a wide variety of cell types, and especially endometrial epithelial cells. Its action is modulated by the presence of IGF-binding proteins (IGFBPs) which are secreted by IGF-I target cells. The partition of IGFBPs between cell-associated and soluble form determines the potentiation or the inhibition of IGF-I action. It is commonly accepted that cell-associated IGFBPs potentiate the IGF-I action while the soluble form of IGFBPs has an inhibitory effect. In endometrial adenocarcinoma, IGF-I is involved in tumoral progression and IGFBPs may be key modulators of the IGF-I-induced cell proliferation. Here we showed that the responsiveness of human endometrial adenocarcinoma cells (HEC-IA cell line) to the mitogenic activity of IGF-I was dependent on the pre-incubation conditions. This responsiveness to IGF-I was conditioned by a differential expression of the IGF system components (IGFBPs and IGF-I receptor) and particularly of the IGFBPs. Indeed, the IGF-I-induced proliferation of the HEC-1A cells was attenuated by the presence of cell-associated IGFBPs. Moreover, the IGF-I incubation induced a release of IGFBP-3 in the culture media as the consequence of an interaction between IGF-I and the cell-associated IGFBP-3. This effect was dose-dependent and was associated with the attenuation of the IGF-I action on cellular proliferation. Thus, IGFBP-3 might be initially expressed as a cell-associated form and then released in the interstitial fluid after a direct interaction with IGF-I. Therefore, in HEC-IA endometrial adenocarcinoma cells responsive to IGF-I, the IGFBP-3 is the main binding protein expressed and both soluble and cell-associated forms act as inhibitors of IGF-I-induced cellular proliferation.