Selective disruption of the E-cadherin-catenin system by an algal toxin.

Yessotoxins (YTXs) are algal toxins that can be accumulated in edible molluscs. YTX treatment of MCF-7 breast cancer cells causes the accumulation of a 100 kDa fragment of E-cadherin, which we have named ECRA(100). A relative decrease in the concentrations of intact E-cadherin did not accompany the accumulation of ECRA(100) in cytosoluble extracts of MCF-7 cells on the first day of YTX treatment, but a collapse of the E-cadherin system was detected after 2-5 days of treatment with the toxin. An analysis of the general structure of ECRA(100) revealed that it consists of an E-cadherin fragment lacking the intracellular domain of the protein. ECRA(100) was not released into culture media of YTX-treated cells. Accumulation of ECRA(100) was observed in other epithelial cells, such as human intestine Caco-2 and MDCK cells after treatment with YTX. In turn, YTX could not induce accumulation of fragments of other members of the cadherin family, such as N-cadherin in the PC12 cell line and K-cadherin in sensitive cells (MCF-7, Caco-2, MDCK). The accumulation of a 100 kDa fragment of E-cadherin devoid of its intracellular domain induced by YTX was accompanied by reduced levels of beta- and gamma-catenins bound to E-cadherin, without a concomitant decrease in the total cytosoluble pools of beta- and gamma-catenins. Taken together, the results we obtained show that YTX causes the selective disruption of the E-cadherin-catenin system in epithelial cells, and raise some concern about the potential that an algal toxin found in seafood might disrupt the tumour suppressive functions of E-cadherin.

Caspase activation and death induced by yessotoxin in HeLa cells.

We have studied the death response induced by yessotoxin (YTX) in cultured HeLa cells, and have compared it to that triggered by okadaic acid (OA) in the same experimental system. Sub-nanomolar concentrations of YTX were found to induce HeLa cell death after a 48-96-h incubation. YTX caused loss of intact poly(ADP-ribose)-polymerase (PARP) in HeLa cells, and detection of the 85kDa fragment, which is indicative of proteolytic attack by caspases. Measurements of caspase activities using extracts prepared from YTX-treated cells and substrates of the caspase-3/7 and caspase-2 isoforms, showed that the relative proteolysis of caspase-3/7 substrate was about eight-fold higher than that of caspase-2, the levels of which were about twice those measured with extracts from control cells. These findings were matched by Western blot analyses of caspase-2, -3 and -7 in HeLa cell extracts, which showed that the levels of pro-caspase-2 were not greatly affected by YTX treatment, whereas pro-caspase-3 and -7 were activated in YTX-treated cells. Taken together, these data complement others previously obtained with OA, and support the notion that caspase isoforms involved in cell death induced by OA and YTX are cell- and toxin-specific.

The toxic responses induced by okadaic acid involve processing of multiple caspase isoforms.

The recognized role of caspases as executioners of apoptosis, led us to investigate their involvement in death responses induced by okadaic acid (OA) in HeLa S(3) and MCF-7 cells. A one-day treatment with OA induced accumulation of the 85kDa poly(ADP-ribose) polymerase (PARP) fragment in cell lysates but the response was prevented if cells were treated with OA in the presence of the caspase inhibitors Z-VAD-FMK and Z-DEVD-FMK. The HeLa S(3) and MCF-7 cells were found to contain measurable levels of the intact caspase-2, -7, -8 and -9 zymogens, whereas caspase-3 was found only in HeLa cells. After one day of OA treatment, pro-caspase-2, -3, -7 and -9 isoforms were found processed in HeLa cells, whereas only pro-caspase-2 was processed in MCF-7 cells. Pro-caspase-8, in turn, was mostly unprocessed in both cell lines. The possible interference of caspase inhibitors on cell death was also evaluated, and we found that both Z-VAD-FMK and Z-DEVD-FMK could contribute different extents of protection of MCF-7 and HeLa cells from toxic effects caused by OA. We concluded that OA triggers multiple pathways of caspase processing, contributing to death responses triggered by OA in HeLa S(3) and MCF-7 cells.

Different sensitivities of p42 mitogen-activated protein kinase to phorbol ester and okadaic acid tumor promoters among cell types.

The operational equivalence of different types of tumor promoters was studied by comparing immediate, early, and late effects of okadaic acid (OA) and 12-O-tetradecanoylphorbol-13-acetate (TPA) on the phosphorylation state of p42 mitogen-activated protein kinase isoform (ERK2) in eight different cell lines. In normal human and mouse fibroblasts, both agents stimulated immediate/early (15-60 min) phosphorylation of ERK2. In mouse 3T3 cells, enhanced phosphorylation of ERK2 was detected only within the first hour of treatment with TPA but not with OA. The early response to both TPA and OA, in turn, was lost in another established cell line, the PNT2 prostate epithelial cells, where we could detect increased levels of phosphorylated ERK2 only after a 24-hr treatment with OA. When the effect of OA was evaluated in different PNT cell strains, we observed that their capacity to respond to this agent, by stabilizing phosphorylated forms of ERK2, was lost in less differentiated strains. In HeLa S3 and HTC tumor cells, however, neither TPA nor OA treatment led to any detectable increase in ERK2 phosphorylation at any time point analyzed. We conclude that the effects of OA and TPA on the phosphorylation states of ERK2 could be related to the cell type, and that the operational equivalence between these two different tumor promoters is maximal in normal cells.

Inhibitors of phosphoprotein phosphatases 1 and 2A cause activation of a 53 kDa protein kinase accompanying the apoptotic response of breast cancer cells.

Treatment of MCF-7 breast cancer cells with 50 nM okadaic acid triggers an apoptotic response which is accompanied by a 7-fold increase in the activity of a protein kinase with a relative molecular mass of 53 kDa. The activity of the kinase was stimulated by cell treatment with inhibitors of phosphoprotein phosphatase 1 and 2A, but not by stressing conditions. Okadaic acid-induced stimulation of the 53 kDa protein kinase was not abolished by coincubation of cells with cycloheximide. We conclude that stimulation of the 53 kDa protein kinase by inhibitors of phosphoprotein phosphatases involves pre-existing molecular components whose activity depends on the phosphorylation state of serine/threonine residues.

The level of pancreatic PLA2 receptor is closely associated with the proliferative state of rat uterine stromal cells.

Rat uterine stromal cells (U(III)) express pancreatic type PLA2 (PLA2-I) receptor and internalize the enzyme bound to receptors. Here, we investigate the proliferating effect and alterations in binding of PLA2-I. There is a dramatic decline in PLA2-I binding in U(III) cells as they progress from a non-confluent proliferating state (40,000 sites/cell) to a confluent state (1300 sites/cell). Intracellular concentration of PLA2-I changed with the alteration in binding, suggesting that regulation in the PLA2 binding capacity may have important implications in growth control mechanisms.

Binding and internalization of extracellular type-I phospholipase A2 in uterine stromal cells.

The cellular uptake of extracellular type-I phospholipase A2 (PLA2) was investigated in rat uterine stromal cells (UIII) in culture, which were found to express the high-affinity binding site for mammalian type-I PLA2, with a measured KD of 6.4 nM, a Bmax of 0.1-1 pmol/mg of DNA at 4 degrees C, and a molecular mass of about 200 kDa. When UIII cells were treated with type-I PLA2 at 37 degrees C, the ligand specifically associated with the cells increased, reaching a plateau after 90 min of incubation, whose level was about 5-fold higher than that measured if cells were maintained at 4 degrees C. We could determine that the PLA2 was bound to plasma membrane receptors which were responsible for internalization of the ligand, and that the binding sites were still suitable for binding at the level of plasma membrane during UIII cell incubation at 37 degrees C. Proteolysis of internalized PLA2 could be clearly detected only after 90 min of UIII cell incubation with the ligand at 37 degrees C, and most of the intracellular PLA2 consisted of the apparently intact 14 kDa enzyme. By cross-linking studies, we found that most of the internalized PLA2 was not associated with the receptor, supporting the conclusion that in our experimental system a single pool of membrane receptors for mammalian type-I PLA2 undergoes cycles of ligand binding, intracellular transfer and release of PLA2, followed by restoration of binding sites on the plasma membrane. We calculated that the rate of internalization of the ligand by one receptor molecule in UIII cells at 37 degrees C is about three molecules of type-I PLA2 per h.

Interference between DNA binding activities of AP-1 and GR transcription factors in rat thymocytes undergoing dexamethasone-induced apoptosis.

The early molecular events of glucocorticoid-induced apoptosis have been investigated by studying glucocorticoid receptor levels, as well as binding activities to GRE and AP-1 sequences, using nuclear extracts from dexamethasone (Dex)-treated rat thymocytes. When the time-course of glucocorticoid-receptor complexes in nuclei of thymocytes was evaluated by binding studies using the tritiated ligand, we found that nuclear accumulation of radioactive complexes occurred in the first hour of incubation, and was followed by a progressive decline. This trend was confirmed by immunoblotting of nuclear proteins using a monoclonal anti-glucocorticoid receptor antibody. When the kinetics of binding activity to AP-1 and GRE sequences were studied, using nuclear extracts prepared from Dex-treated thymocytes in gel shift assays, we found peaks at 1 and 2 h after Dex treatment, and a return to basal levels in the following hours. Binding specificity was proved by competition studies using non-radioactive sequences, including mutated AP-1. Unexpectedly, however, protein binding to GRE was better competed for by AP-1 sequence than by GRE itself. Data obtained using the super gel shift assay suggested that AP-1/Jun can be responsible for the high affinity for the GRE sequence. Thus, we report here for the first time that an interference between AP-1 and GR in the binding to DNA consensus sequences-previously described in other biological systems-also occurs during apoptosis induced by glucocorticoids in lymphoid cells.

Steroid hormones and temperature induce changes of binding parameters of their receptors in intact cells.

When MCF-7 cells were treated with 17 beta-estradiol, dexamethasone, or promegestone at 37 degrees C, the KD of receptors for their cognate ligands was found to decrease as compared to that measured at 2 degrees C. Cell incubation with hormone at 37 degrees C did not affect the Bmax of glucocorticoid and progesterone receptors, but caused a 40% increase of that of estrogen receptor. This increase required the presence of ligand, was insensitive to cycloheximide, and was completed within 10 min of cell incubation at physiological temperature. We conclude that an early step of estrogen action is the recruitment of pre-existing receptor molecules through activation of their binding capacity.

Nanomolar concentrations of untransformed glucocorticoid receptor in nuclei of intact cells.

The subcellular distribution of untransformed glucocorticoid-receptor complex in vivo has been studied by chemical crosslinking of intact cells, and using a procedure adequate for correction of experimental errors due to redistribution of components between cytosolic and nuclear fractions. We found that in HeLa S3 cells 85.4% of total glucocorticoid-receptor complexes are located in nuclei, and 14.6% are cytosolic. If measurements were performed with MCF-7 cells, we determined that the nuclear pool of glucocorticoid-receptor complexes accounts for 75.2% of the total cellular content, whereas the remaining 24.8% are cytosolic. When the subcellular distribution of estrogen-receptor complexes was determined, instead, we found that they are almost exclusively located in nuclei of MCF-7 cells, which contain 88.9% of the total. In order to estimate the molar concentration of receptors in cytosol and nuclei of intact cells, we determined the free water content of the two compartments. The volume of solvent was found to vary in the three cell lines we have studied, and our data showed that these variations are due to the cytosolic fractions, as the free water content of nuclei is essentially the same in those cells. When the free water content and the levels of glucocorticoid-receptor complexes we have measured were used to estimate the molar concentrations of receptors, we found that these range between 0.4 and 18.9 nM in cytosols, and between 3.9 and 6.3 nM in nuclei of the three cell lines we have studied. We then concluded that the relative distribution of untransformed glucocorticoid-receptor complexes between cytosol and nuclei is cell-specific but their molar concentration in the nuclear compartment does not greatly vary among different cells.

Oligomeric structures of cytosoluble estrogen-receptor complexes as studied by anti-estrogen receptor antibodies and chemical crosslinking of intact cells.

The structure of estrogen-receptor complexes recovered in cytosolic extracts of MCF-7 cells treated with hormone at 2 degrees was probed by chemical crosslinking of intact cells and sample analysis with four monoclonal anti-estrogen receptor antibodies. When MCF-7 cells were treated with either glutaraldehyde or dithiobis(succinimidyl propionate), cytosoluble estrogen-receptor complexes consisted of two major forms sedimenting as 4 S monomers and 8-9 S salt-resistant oligomers. By high salt sucrose density gradient centrifugation, we could observe that the four monoclonal anti-estrogen receptor antibodies bound different forms of receptor complexes from crosslinked cells. While H222 and H226 antibodies could interact with any form we detected, the D75 and D547 monoclonals could only recognize those showing sedimentation coefficients lower than 7 S. When cytosolic extracts from [35S]-methionine-labeled cells were subjected to immunoprecipitation with H222 and D75 anti-estrogen receptor antibodies, electrophoretic analysis of material extracted from immunoprecipitates revealed the presence of 65 kDa estrogen receptors. If extracts were prepared from crosslinked cells, instead, two more components with estimated molecular masses of 220 and 100 kDa were specifically immunoprecipitated by the H222 antibody, whereas only the 100 kDa component and the estrogen receptor were found in immunoprecipitates obtained with the D75 monoclonal. When estrogen-receptor complexes were immunopurified from extracts prepared after cells had been crosslinked with dithiobis(succinimidyl propionate), and the oligomers were dissociated by treatment with beta-mercaptoethanol, electrophoretic analysis of our samples showed that only the 65 kDa estrogen receptor and a 50 kDa protein were selectively immunoprecipitated by anti-estrogen receptor antibodies. We concluded that the structures of cytosoluble estrogen-receptor complexes in MCF-7 cells treated with hormone at 2 degrees C, include oligomeric forms which contain a 50 kDa non-steroid binding protein.

The subcellular distribution of glucocorticoid-receptor complexes as studied by chemical crosslinking of intact HTC cells.

Treatment of intact HTC cells with glutaraldehyde results in redistribution of glucocorticoid binding sites between cytosolic and nuclear fractions. The decrease in cytosolic receptors and their accumulation at the nuclear level were found to be directly related to the glutaraldehyde concentrations employed in our procedure and inversely related to the cell density of samples. When the data from eleven separate experiments were combined, and analyzed by linear regression of cytosolic and nuclear levels of receptor complexes vs the ratios between the DNA and glutaraldehyde concentration of our samples, two lines were obtained whose intercepts on the ordinate yielded values of cytosolic and nuclear receptors corresponding to 37.5 and 62.5% of the total cellular pool, respectively. When we compared the subcellular redistribution of glucocorticoid receptor to that of the cytosolic enzyme lactate dehydrogenase upon HTC cell crosslinking with glutaraldehyde, we found that the cytosolic and nuclear levels of the enzyme were 53.2 and 46.8% of the total content, respectively. If the subcellular distribution of glucocorticoid receptor is corrected for the artefactual redistribution induced by crosslinking, using the values obtained for lactate dehydrogenase, it can be concluded that glucocorticoid receptors in HTC cells are distributed between cytosol and nuclei in a ratio which is about 2:1. Our findings lend further support to the conclusion that only a portion of glucocorticoid receptor is cytosolic in intact cells.

The quaternary structures of untransformed steroid hormone receptors: an open issue.

Oligomeric forms of steroid hormone receptors have been observed since the original detection of these proteins in cytosols from target tissues, and the capacity of receptors to arrange in supramolecular entities was found to be related to their functional untransformed states of low affinity for nuclear acceptor sites. In this paper we discuss the models of quaternary structures of untransformed receptors and steroid-receptor complexes proposed in the last two decades, including homo-oligomers as well as hetero-oligomers containing non-steroid binding proteins and/or RNA. The multiplicity of forms detected in cell-free systems, their stabilization by unphysiological experimental conditions, and their sensitivity to the ionic species employed to prepare and analyze steroid receptors, are critically evaluated, and it is concluded that the extrapolation of models to account for quaternary structure of receptors in intact cells is unwarranted. We have also discussed the experimental strategies which have been developed to circumvent the possible artefactual changes in the arrangements of oligomeric receptors following cell rupture, in order to probe the existence of these forms in vivo, and to characterize their composition and structure. The data obtained by these studies support the concept that oligomeric untransformed steroid receptors exist in intact cells, where they can be present in multiple supramolecular arrangements whose quaternary structures remain to be established.

Detection of glucocorticoid-receptor complex oligomers in nuclear extracts from cells exposed to hormone at physiological temperature.

When HeLa cells were incubated with tritiated dexamethasone mesylate at 2 degrees C, sodium dodecyl sulfate-polyacrylamide gel electrophoresis of cytosolic and nuclear extracts revealed the presence of two monomeric receptor complex forms with estimated molecular masses of about 98 and 87 kDa. If cells were subjected to crosslinking with glutaraldehyde, a third form consisting of a 250 kDa oligomer was also detected. When HeLa cells were treated with dexamethasone mesylate at 37 degrees C, and were subjected to crosslinking, electrophoresis of cytosolic glucocorticoid-receptor complexes was drastically reduced, whereas their levels in nuclear extracts were not appreciably altered.

Rapid homologous up-regulation of binding capacity of androgen receptors in intact cells.

Incubation of MCF 7 cells with 5 alpha-dihydrotestosterone (DHT) at 37 degrees C led to a 70% increase in the Bmax of androgen receptor, as compared to the values measured at 2 degrees C, without detectable changes in equilibrium dissociation constants. When MCF 7 cells were incubated with hormone at 2 degrees C, to reach steady-state levels of androgen-receptor complex, a subsequent temperature shift to 37 degrees C induced a rapid (t 1/2 = 3 min) cycloheximide-insensitive increase in DHT binding to androgen receptor. MCF 7 cell treatments at 37 degrees C either before or after incubation with DHT at 2 degrees C showed that up-regulation of binding capacity of androgen receptor could be observed only if hormone is present during incubation at physiological temperature.

Transformation of glucocorticoid-receptor complexes is accompanied by dissociation of oligomers in intact cells.

The association of glucocorticoid-receptor complexes with other components in vivo has been evaluated by chemical crosslinking of hormone-treated cells. When cells were incubated with hormone at 2 degrees C, before being subjected to crosslinking, most glucocorticoid-receptor complexes were found untransformed, as judged by DEAE-cellulose chromatography, and sedimented as 11-6 S oligomers in sucrose gradients containing 0.3 M NaCl. If crosslinking was performed after cells were treated with hormone at 37 degrees C, about 60% of cytosolic glucocorticoid-receptor complexes were found transformed, and sedimented as 4 S monomers.

Stabilization of glucocorticoid-receptor interactions in vitro by removal of RNA bound to receptor complexes in vivo.

The dissociation of the steroid from glucocorticoid-receptor-RNA complexes at 5 degrees C was evaluated in cytosolic and nuclear extracts prepared from Hela cells crosslinked in vivo with glutaraldehyde. Sample treatment with catalytically active RNase A prevented the dissociation of the steroid which was induced by sample dilution with buffer. Dilution of the extracts with boiled cytosol, instead, stabilized steroid-receptor interactions. We conclude that some heat-stable factor should be also associated with glucocorticoid-receptor-RNA complexes from crosslinked cells, stabilizing steroid-receptor interactions, and we propose that it could counteract the labilizing effect of RNA.

Glucocorticoid-receptor complexes are associated with small RNA in vitro.

Identification of RNA associated with soluble glucocorticoid-receptor complexes of HeLa cells was performed by immunoprecipitation of receptor complexes with a monoclonal antibody raised against rat liver glucocorticoid receptor. Polyacrylamide gel electrophoresis of RNA extracted from immunoprecipitates of cytosolic complexes revealed the presence of eight RNA bands, consisting of 28S, 18S, and small RNAs, including 5.8S, 5S and tRNA. A comparison of RNA species immunoprecipitated by monoclonal anti-glucocorticoid receptor antibody and IgG purified from normal mouse serum showed that four small RNAs were preferentially recovered after immunoprecipitation with anti-glucocorticoid receptor antibody. When these species were analyzed on sequencing gels, their nucleotide lengths coincided with those of 7-3, 7S, U2, and U1 RNA. Immunoprecipitation of nuclear extracts containing glucocorticoid-receptor-RNA complexes showed that the same set of small RNAs was preferentially immunoprecipitated by anti-glucocorticoid receptor antibody. The four small RNAs we detected represented minor species in whole extracts, and their preferential immunoprecipitation by anti-glucocorticoid receptor antibody was prevented by removal of glucocorticoid-receptor complexes from HeLa cell extracts. We conclude that 7-3, 7S, U2, and U1 RNA are associated with glucocorticoid-receptor complexes in vitro, and hypothesize that post-transcriptional effects of glucocorticoids may in part be mediated through interaction of receptor complexes with these small RNAs.

Particulate untransformed glucocorticoid-receptor complexes from HeLa cells crosslinked in vivo.

When control HeLa cells were incubated at 2 degrees C in the presence of tritiated dexamethasone, most glucocorticoid-receptor complexes were found in cytosolic extracts as untransformed forms. Chemical crosslinking of intact HeLa cells resulted in the immobilization of 50% of the total cellular glucocorticoid-receptor complexes in the nuclear fraction. Under these conditions the redistribution of total protein, RNA and lactate dehydrogenase activity between cytosol and nuclei was negligible, indicating that glucocorticoid binding in the nuclear fraction was not due to a methodological artifact. High levels of glucocorticoid receptor were also found in the nuclear fraction of crosslinked cells which were not exposed to glucocorticoids. Nuclear receptor complexes could be released in soluble forms by DNase I and sonication. Evaluations of DNA binding and ionic properties of glucocorticoid receptors prepared from control and crosslinked cells maintained at 2 degrees C revealed that most of the hormone-receptor complex in cytosols and nuclear extracts behaved as untransformed forms. As opposed to glucocorticoid receptors prepared from control cells, heat treatment of extracts obtained from crosslinked cells did not result in increased DNA binding and changes in ionic properties of receptor complexes. I conclude that untransformed glucocorticoid receptors are present in both cytosol and nuclei of intact cells.