Prolonged treatment of breast cancer cells with antiestrogens increases the activating protein-1-mediated response: involvement of the estrogen receptor.

At micromolar (pharmacological) concentrations, the action of tamoxifen on the proliferation of estrogen-dependent cells can be mediated not only by the estrogen receptor (ER), but also by other target molecules, such as protein kinase-C (PKC), which are easily inhibited by antiestrogens in cell-free experiments. By developing MTLN and MDT cell lines, in which any modulation of PKC activity is reflected by a variation of the expression of an activating protein-1 (AP-1)-controlled firefly luciferase gene, we investigated whether such antiestrogen inhibitory effects on PKC occurred in intact breast cancer cells. Firstly, in short term (4-h) treatment of both cell lines, antiestrogens only inhibited the 12-O-tetradecanoyl-phorbol-13-acetate-induced luciferase activity at very high concentrations (30 microM). A cytolytic effect was also observed. Secondly, in prolonged (4-day) treatments of MTLN (ER-positive) cells, low antiestrogen concentrations (nanomolar) decreased the basal AP-1 response by about 2 and increased the 12-O-tetradecanoyl-phorbol-13-acetate-stimulated AP-1 response by about 3-4. This stimulation was mediated by ER, because 1) dose-response curves established with tamoxifen and hydroxytamoxifen were in agreement with their affinity for ER; 2) when present with antiestrogens, estradiol abolished this phenomenon; and 3) this effect was not observed in MDT (ER-negative) cells. Such a latent activation of AP-1 pathway could appear in the course of breast cancer antiestrogen treatment, in conditions where natural PKC activators are abnormally produced with unexpected consequences on the results of a long term antiestrogen treatment.

Potentiation by cholesterol and vitamin D3 oxygenated derivatives of arachidonic acid release and prostaglandin E2 synthesis induced by the epidermal growth factor in NRK 49F cells: the role of protein kinase C.

We have previously demonstrated that oxysterols and calcitriol potentiate arachidonic acid (AA) release and prostaglandin (PG) synthesis when NRK cells (fibroblastic clone 49F) are activated by foetal calf serum. As serum is essential for a full oxysterol effect, we hypothesized that these compounds could act on one or more of the events triggered by serum growth factor binding to their specific receptors and leading to PLA2 activation; we showed that the oxysterol effect on AA release is synergistic with, but not fully dependent on, protein kinase C (PKC) activity and Ca2+ ion fluxes, suggesting that oxysterols could effect early events in the cell signalling pathway. In the present paper, we investigated the effect of some oxysterols and calcitriol on epidermal growth factor (EGF)-induced AA release and PGE2 synthesis in NRK cells. The clear potentiation of EGF effect by most of the oxygenated sterols--chiefly when polyoxidized--cannot be explained by a modification of EGF high affinity binding site number which was only moderately increased after a 4 h incubation of cells with these compounds, and moreover was not related to the ability of a given oxysterol to increase PLA2 activity; whatever the compound, the dissociation constant (Kd) of either a high or low affinity binding site was unchanged (respectively, 3.5 x 10(-11) M and 4.4 x 10(-10) M). Genistein, a known inhibitor of EGF receptor tyrosine kinase, changed neither the EGF effect on AA release nor its potentiation by oxysterol, whereas it inhibited PGE2 synthesis in both situations. PKC activation by phorbol ester TPA increased the effect of EGF alone as well as the oxysterol potentiating effect, whereas PKC down-regulation strongly decreased both of these effects, showing that both are dependent on PKC activity. Nevertheless staurosporine, a PKC inhibitor, did not reproduce the effects of PKC down-regulation on EGF activation: stimulatory when AA release was induced by EGF alone, inhibitory when AA release is induced by TPA alone, this compound did not modify the oxysterol potentiating effect. In conclusion, the potentiating effect of oxysterols on AA release seems to be exerted downstream to the growth factor receptor (as demonstrated here with EGF) and probably at the PKC level, but not exclusively.

Oxysterol activation of arachidonic acid release and prostaglandin E2 biosynthesis in NRK 49F cells is partially dependent on protein kinase C activity [corrected].

We previously demonstrated that the oxysterol potentiation of arachidonic acid release and prostaglandin biosynthesis induced by foetal calf serum activation of normal rat kidney (NRK) cells (fibroblastic clone 49F) was not related to a direct effect of oxysterols on cell free Ca2+ level. Since both Ca2+ variations and protein kinase C are involved in arachidonic acid release in some models, we looked for a possible modulation by protein kinase C in the oxysterol effect on arachidonic acid release. We show that when the phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate (TPA), a protein kinase C activator, was added to the culture medium, the oxysterol effect on arachidonic acid release and prostaglandin synthesis clearly increased. Moreover, the effect of TPA was dose-dependent and TPA EC50 (4 x 10(-9) M) was unchanged in the presence of the oxysterol. Preincubation of cells with TPA for 24 h prevented the arachidonic acid release induced by TPA alone, whereas the oxysterol effect was decreased but not abolished. In the absence of serum, TPA and ionomycin added together induced the same noticeable (arachidonic acid) release and PGE2 synthesis as serum alone. Nevertheless, the potentiating effect of cholest-5-ene-3 beta, 25-diol was much higher when serum itself was used to activate NRK cells than it was in the present serum-mimicking experimental conditions. Thus, the presence of growth factors is probably required to obtain a full oxysterol effect. We conclude that the oxysterol effect was synergistic with, but not fully dependent on, protein kinase C and Ca2+ ion fluxes, therefore oxysterols could affect earlier events triggered by serum growth factor binding to their cell membrane receptors.

Arachidonic acid release and prostaglandin biosynthesis in synchronized rat embryo fibroblasts.

Arachidonic acid (AA) release and prostaglandin (PG) biosynthesis were studied in rat embryo fibroblasts (R 129) synchronized by double thymidine-excess block. Whatever the culture medium was (medium 199 supplemented with 10% fetal calf serum (FCS) or 1% FCS plus 0.1% bovine serum albumin (BSA], AA release rapidly increased until the 4th hour of the cell cycle (S phase), remained on a plateau in G2M and G1 phases and did not increase again in the S phase of the following cell cycle. Time course and amplitude of AA release in synchronized cells did not differ from what it was observed after the simple renewal of the culture medium in asynchronous cells. So AA release seemed to be independent of the cell cycle. By contrast, PGE2 and PGF2 alpha biosynthesis clearly increased in the S phase of two consecutive cell cycles, indicating that cyclooxygenase activity and not phospholipase A2 activity vary according to the cell cycle.

Mechanism of the activation of arachidonic acid release by oxysterols in NRK 49F cells: role of calcium.

We previously demonstrated that oxysterols added to the culture medium of NRK 49F cells labelled with [14C] arachidonic acid potentiated arachidonic acid (AA) release and prostaglandin (PG) E2 biosynthesis induced by the activation of these cells with fetal calf serum (FCS). In the absence of FCS, oxysterols had no effect on AA release. As phospholipase (Plase) A2 activity is Ca2(+)-dependent, we investigated whether oxysterol potentiating effect on AA release was related to an effect of these compounds on cell Ca2+ concentration. In this paper, we show that the intensity of potentiation by oxysterol varies with the external cell Ca2+ concentration; when external Ca2+ is chelated by EGTA, the oxysterol effect persists, though it is decreased. The Ca2+ channel inhibitor nifedipine does not decrease the potentiating effect of 25-OH cholesterol, indicating that, if oxysterol favours Ca2+ entry into the cell, the nifedipine inhibited channel is not involved. At the usual concentration (5 micrograms/ml), oxysterols are not able to increase, immediately or after a short time of contact (90 min) the concentration of intracellular free Ca2+ ([Ca2+])i measured by fluorescence of Quin-2; at very high concentration of oxysterol (25 micrograms/ml), [Ca2+]i only slightly increases (+30%). The liberation of AA induced by cell activation with the Ca2+ ionophore ionomycin is also potentiated by 25-OH cholesterol. All these observations are not in favour of a proper effect of oxysterols on cell Ca2+ level.

Serum-induced arachidonic acid release and prostaglandin biosynthesis are potentiated by oxygenated sterols in NRK 49F cells.

Fetal calf serum is able to activate arachidonic acid release from phospholipids in NRK 49F cells. We showed that this phenomenon can be potentiated by adding oxysterols to the culture medium. The oxysterol effect was dose-dependent and was not observed in the absence of fetal calf serum. Greater amounts of prostaglandin E2 and prostaglandin F2 alpha were released into the medium in the presence of oxysterols without apparent modification of the cyclooxygenase activity. The most effective oxysterols, in descending order, were the following: calcitriol greater than 7 alpha-hydroxycholesterol greater than 7 beta-hydroxycholesterol greater than 25-hydroxycholesterol. Cholesterol and 7-ketocholesterol were unable to activate phospholipase activity. The mechanism of this activation by oxysterols is still unknown.

Rat liver cytosol oxysterol-binding protein. Characterization and comparison with the HTC cell protein.

A cytosol protein that specifically binds cholesterol derivatives oxygenated on the side chain has been demonstrated in rat liver and transformed HTC cells. This protein, of which the sedimentation coefficient is about 8 S, was saturable and showed a high binding affinity (Kd about 5 X 10(-9) M) for 25-hydroxycholesterol. Its molecular mass is about 160 kDa. The physicochemical characteristics of this protein were identical whether the model was normal or transformed. This oxysterol-binding protein differs from the well-known sterol carrier proteins.

Characterization of oxysterol-binding protein in rat embryo fibroblasts and variations as a function of the cell cycle.

A cell-free system assay involving cell freeze-thawing and protein fractionation by ammonium sulfate precipitation was developed to characterize a cytosol binding protein specific for oxysterols in rat embryo fibroblasts. This protein shared common characteristics with the oxysterol-binding protein described in L cells and in normal human lymphocytes: 8 S sedimentation coefficient, sterol-protein complex of Mr 160 600, saturability, high affinity (Kd in the range of 10(-9) M) and specificity for cholesterol derivatives oxidized on the side chain. These compounds were better inhibitors of DNA synthesis than the compounds oxidized on the nucleus, whereas no difference was found between sterols oxygenated either on the side chain or on the nucleus, as far as inhibition of hydroxymethylglutaryl-CoA reductase (HMG-CoA reductase) was concerned. Macromolecular components capable of specifically binding 25-hydroxycholesterol were also detected in the fibroblast nucleus. The cytosol oxysterol-binding protein showed a sharp increase (5-fold) in the G2M phase of the cell cycle and in exponentially growing cells (maximal binding site number/cell: 43 500, versus 8850 in confluent cells). Neither the affinity nor the sedimentation coefficient of the protein changed in either situation. The quantitative (but not qualitative) variations of oxysterol-binding protein could be related to the inhibitory effect of 25-hydroxycholesterol on DNA synthesis, which becomes critical when this sterol is added in the G2M phase of the cell cycle.

A specific hydroxysterol binding protein in human lymphocyte cytosol.

A discriminating system capable of recognizing the oxygenated sterols was investigated in human lymphocytes. After labelling entire cells with 25-hydroxy [3H] cholesterol (10 nM) the cytosol was ultracentrifuged on a linear sucrose density gradient. Bound 25-hydroxy [3H] cholesterol was located in a single peak with a sedimentation coefficient of 8.3 S. Pronase treatment abolished the radioactive peak. This 8.3 S protein had a low binding capacity for 25-hydroxy [3H] cholesterol and probably a high affinity. This last parameter was not determined on account of some difficulties encountered in a cell-free system relating to the physico-chemical properties of 25-hydroxycholesterol. Only the hydroxylated sterols closely related to 25-hydroxycholesterol were capable of specifically binding to the 8.3 S protein, in contrast with cholesterol. This protein differed from the binding proteins of oxygenated derivatives of vitamin D3 and glucocorticoids. With the human lymphocyte as a model and under our experimental conditions, this hydroxylated sterol-binding protein seems to be involved rather in the cell division control than in the regulation of HMG-CoA reductase activity: indeed, the hydroxysterols able to inhibit thymidine [3H] incorporation into DNA are recognized by this protein whereas the hydroxysterols active on HMG-CoA reductase activity without affecting thymidine [3H] incorporation into DNA are not.