Expression of androgen receptor coactivators in normal and cancer prostate tissues and cultured cell lines.

BACKGROUND: In prostate cancer cell lines, androgen receptor (AR) coactivators modulate the transcriptional activity of AR. However, very little is known about their expression in normal prostate tissue and during progression to cancer. METHODS: AR and coactivators ARA54, ARA55, ARA70, and SRC1 RNA were analyzed by RT-PCR in normal and tumoral tissues of the same prostate, in prostate cell lines, and after hormonal treatments of prostate epithelial cells. RESULTS: AR-coactivators were expressed in normal and tumoral tissues and in cultured prostate cells; only ARA55 expression was decreased in tumoral relative to normal tissue of all seven prostates analyzed. It was not expressed in LNCaP and DU145 cancer cells and low in PNT2 immortalized cells in which all coactivator's expression were down regulated by DHT and up regulated by E2. In addition, coactivator's expression was increased in hyperplastic relative to normal prostate fibroblasts. CONCLUSIONS: ARA55 is both an AR coactivator and a focal adhesion protein (Hic-5). Its role in the progression of prostate carcinoma may therefore involve these two different functions. Its decrease in cancer tissue suggests that it plays a different role than that expected, namely, facilitate cell proliferation and therefore mobility and metastasis.

Alterations of expression and regulation of transforming growth factor beta in human cancer prostate cell lines.

TGF beta can promote and/or suppress prostate tumor growth through multiple and opposing actions. Alterations of its expression, secretion, regulation or of the sensitivity of target cells can lead to a favorable environment for tumor development. To gain a better insight in TGF beta function during cancer progression, we have used different cultured human prostate cells: preneoplastic PNT2 cells, the androgen-dependent LNCaP and the androgen-independent PC3 and DU145 prostate cancer cell lines. We have studied by specific ELISA assays in conditioned media (CM), the secretion of TGF beta 1 and TGF beta 2 in basal conditions and after hormonal treatment (DHT or E2) and the expression of TGF beta 1 mRNA by Northern blot. We have also compared the effect of fibroblast CM on TGF beta secretion by the different cell types. Compared to PNT2 cells, cancer cell lines secrete lower levels of active TGF beta which are not increased in the presence of fibroblast CM. LNCaP cells respond to androgen or estrogen treatment by a 10-fold increase of active TGF beta secretion while PC3 and DU145 are unresponsive. In conclusion, prostate cancer cell lines have lost part of their ability to secrete and activate TGF beta, and to regulate this secretion through stromal-epithelial interactions. Androgen-sensitive cancer cells may compensate this loss by hormonal regulation.

Estrogen and antiestrogen actions on transforming growth factorbeta (TGFbeta) in normal human breast epithelial (HBE) cells.

We have previously shown that estradiol (E2) increases the growth of normal human breast epithelial (HBE) cells and the antiestrogen 4-hydroxytamoxifen (4-OHT) inhibits estrogen-induced proliferation. These effects of estrogens and antiestrogens on proliferation have also been well documented in breast cancer cells. One mechanism for the antiproliferative effects of antiestrogens is the stimulation of TGFbeta in hormone-dependent MCF-7 and T47D cells. The role of this inhibitory growth factor in normal human breast cells has not been well studied. Accordingly, we measured the amounts of total and active TGFbeta1 and TGFbeta2 by specific E(max) immunoassay (EIA) in culture medium from normal breast cells (epithelial and fibroblasts) and from various ER- and ER+ breast cancer cell lines. We established that HBE cells are sensitive to the antiproliferative effect of TGFbetas, and studied the effect of E2 and 4-OHT, alone or in combination, on the secretion and activation of TGFbetas by HBE cells. HBE cells secrete TGFbeta1 and even more TGFbeta2, and are sensitive to these factors. However, in contrast to MCF-7 cells, TGFbeta secretion in normal breast cells is not regulated by E2 and 4-OHT.

Transforming growth factor beta in the human prostate: its role in stromal-epithelial interactions in non-cancerous cell culture.

BACKGROUND: Stromal-epithelial interactions play a critical role in prostate development, but the precise mechanisms are still unknown. Transforming growth factor-beta (TGFbeta) could be a potential mediator of these interactions, but there is as yet no clear demonstration of its role. METHODS: Separate cultures and co-cultures of fibroblasts and epithelial human prostate cells were performed. We measured TGFbeta1 and TGFbeta2 secretion by specific ELISA assay, cell growth by DNA assay, and TGFbeta type II receptor expression by RT-PCR. RESULTS: Co-culture resulted in a 20% inhibition of epithelial cell growth, similar to that obtained after TGFbeta treatment (2 ng/ml for 48 hr), but without affecting fibroblast proliferation. This was accompanied by a five- to six-fold increase in epithelial TGFbeta2 secretion. CONCLUSIONS: These results demonstrate for the first time that TGFbeta2 secretion by prostate epithelial cells is under the direct control of a diffusible factor secreted by fibroblasts. They emphasize the role of TGFbeta in stromal-epithelial interactions.

Hormonal regulation of the androgen receptor expression in human prostatic cells in culture.

The regulation of the androgen receptor (AR) expression was studied using immunocytochemical and Western blot techniques on separate cultures of epithelial cells (PNT2) and fibroblasts of human prostate. In both cell types, immunocytochemistry revealed both nuclear and cytoplasmic staining. Treatment with DHT (5 x 10(-9) M) increased both the intensity of nuclear staining and the number of cells stained. The increase, observed after DHT treatment was markedly decreased by cyproterone acetate (5 x 10(-7) M), confirming a direct action of DHT via the AR. This autoregulation of AR was confirmed by Western blot, and seems to involve transcription and protein synthesis, since it was suppressed by actinomycin D and cycloheximide. In fibroblasts, known to contain an estrogen receptor, estradiol treatment (5 x 10(-7) M) also increases the AR immunostaining. In addition, coculture studies show that epithelial cells require the presence of fibroblasts for optimal expression of the AR. These results demonstrate that prostate epithelial cells and fibroblasts have retained in culture, an hormonal sensitivity correlated with the presence of specific receptors and can serve as a model for the study of hormone action in this tissue in normal or pathological conditions.

Pharmacological and molecular evidence for the expression of the two steroid 5 alpha-reductase isozymes in normal and hyperplastic human prostatic cells in culture.

BACKGROUND: Whereas the embryological development of the human prostate is clearly dependent on steroid 5 alpha-reductase (5 alpha-R) type 2 expression, the respective expression of the two known isoforms (types 1 and 2) of 5 alpha-R in the adult human prostate remains unclear. METHODS: 5 alpha-R isoform mRNA expression (Northern blots and reverse transcriptase-polymerase chain reaction [RT-PCR]) and enzyme activity were studied in immortalized epithelial cells (NE) and in fibroblasts from normal (NF) or hyperplastic (BPHF) human prostates. RESULTS: 5 alpha-R activity (fmol/microgram DNA/hr) was 1.43 +/- 0.5 in NE, 10.7 +/- 4.7 in NF, and 79 +/- 37 in BPHF. mRNAs for both 5 alpha-R isoforms were expressed in the three cell types, as shown by Northern blot and RT-PCR analysis. LY306089, a selective 5 alpha-R type 1 inhibitor, strongly inhibited 5 alpha-R activity in all cell types (IC50: 10 nM), confirming the predominant expression of 5 alpha-R type 1 in these cells. Finasteride, a 5 alpha-R type 2 inhibitor, was less efficient (IC50: 45, 35, and 65 nM in NE, NF, and BPHF, respectively). In addition, the inhibition by finasteride decreased with serial subculture in NF only, suggesting an effect of age in culture on the expression of 5 alpha-R type 2 in these cells. SKF105657, also a 5 alpha-R type 2 inhibitor, was a poor inhibitor in this system. CONCLUSIONS: These studies demonstrate that human prostate cells in culture express both isoforms of 5 alpha-R and suggest a balance in the expression of the two isoforms as a function of various regulatory factors.

[Mutations of the androgen receptor gene a possible role in prostate cancer]. / Mutations du gène du récepteur des androgènes: rôle possible dans le cancer de la prostate.

Prostate cancer is the second most frequent cancer in men, and the first after 75 years of age. It is androgen dependent and modifications of the androgen receptor (AR) could therefore be involved in its apparition, progression and evolution towards a stage of androgen independence which always occurs. Mutations of the AR have indeed been described. Some result in a more active receptor (constitutive mutations, amplification of the AR gene expression, prevalence of shorter alleles in exon 1) and could therefore be responsible for the progression of the disease in the absence of androgens or the increased cancer risk in some populations. Other mutations, in the ligand binding domain, modify the AR specificity resulting in its activation by weak androgens or even antiandrogens, progestagens or estradiol. These mutations could also explain why prostate cancer progresses in the absence of androgens. On the other hand, mutations resulting in androgen insensitivity could explain why a significant number of prostate cancer remain latent and never develop into an aggressive tumor. Finally, exon 1 polymorphism could be used as a prognostic marker, as it has been shown that the shorter alleles result in a more active AR and are predominant in populations at higher risk for prostate cancer. At any rate, even if AR mutations may be more frequent than initially thought, they remain rare and their significance remains to be determined.

Predominant expression of 5 alpha-reductase type 1 in pubic skin from normal subjects and hirsute patients.

Dihydrotestosterone (DHT), the 5 alpha-reduced metabolite of testosterone, is the active molecule triggering androgen action, and 5 alpha-reductase (5 alpha-R), the enzyme converting testosterone to DHT, is a key step in this mechanism. Skin, like prostate, is a DHT- dependent tissue. Our laboratory demonstrated, many years ago, that 5 alpha-R in external genitalia was not regulated by androgens, whereas it was androgen dependent in public skin. As two genes, 5 alpha-R types 1 and 2, encoding for 5 alpha-R enzymes have been recently cloned, we undertook the present study to determine whether the two enzymes we had postulated on the basis of regulation studies were coincident with the cloned isoforms. The expression of the two isoforms was studied in genital and pubic skin fibroblasts from normal men, normal women, and hirsute patients. Messenger ribonucleic acid analysis, using Northern blot and RT-PCR techniques, indicated that both 5 alpha-R1 and -2 messenger ribonucleic acids are expressed in genital skin as well as in public skin fibroblasts. In contrast, studies using specific inhibitors of 5 alpha-R1 (LY306089) and 5 alpha-R2 (finasteride) showed that 5 alpha-R2 is predominant in pubic skin of normal men, normal women, and hirsute patients. These data raise the question of the possible use of specific 5 alpha-R1 inhibitors in the treatment of idiopathic hirsutism.

[5-alpha-reductases: physiology and pathology]. / Les 5 alpha-réductases: physiologie et pathologie.

In most androgen target tissues, the first step of androgen action is the 5 alpha-reduction of testosterone to DHT which binds to the androgen receptor with an affinity 3 to 4 fold higher than testosterone. Two genes, encoding two isozymes of 5 alpha-reductase (5 alpha-R) have been cloned. The two isoforms, 5 alpha-R1 and 5 alpha-R 2 are located on chromosomes 5 and 2 respectively and differ in optimal pH, substrate and inhibitor affinities and tissue expression. 5 alpha-R 2 is responsible for sexual differentiation. It is the major form expressed in the prostate where it seems necessary for embryonic growth and development. 5 alpha-reductase deficiency results in androgen insensitivity due to abnormal 5 alpha-R 2. Affected patients are XY individuals with a very peculiar form of male pseudohermaphroditism: they have feminine genitalia at birth and masculinize at puberty. 29 mutations, spanning the whole coding portion of the gene, have been described; correlation between mutations and enzyme activity have led to the suggestion that both the N- and the C-terminal end of the gene are involved in substrate binding, whereas the cofactor binding-site is located in the C-terminus. In contrast to androgen insensitivity due to 5 alpha-reductase deficiency, increased 5 alpha-reductase activity can result in androgen hypersensitivity as described in idiopathic hirsutism or benign prostatic hyperplasia. In these case 5 alpha-R 1 could possibly be involved.

A point mutation in the second zinc finger of the DNA-binding domain of the androgen receptor gene causes complete androgen insensitivity in two siblings with receptor-positive androgen resistance.

We have analyzed the nucleotide sequence of complementary and genomic DNAs of the human androgen receptor (AR) gene in two siblings (patients 9006 and 9030) with receptor-positive complete androgen insensitivity (Rec(+)-CAI). Northern analysis indicated that mRNA of the AR was normal in size. However, its expression was relatively reduced in both patients. Consistent with the normal androgen-binding capacity (496 and 552 fmol/mg DNA for patients 9006 and 9030, respectively) but decreased DNA-binding ability (168 fmol/mg DNA) measured in genital skin fibroblasts, no mutation was found in both N-terminal and ligand-binding domains of the AR. However, a single base substitution (G-->A) was found in the second zinc finger of the DNA-binding domain at nucleotide 2372 of the AR cDNA in both cases. This resulted in the replacement of a highly conserved arginine residue (amino acid 614) by a histidine. When the mutated receptor plasmid was cotransfected into PC-3 cells together with the reporter chloramphenicol acetyltransferase gene, chloramphenicol acetyltransferase activity was not induced by 5 alpha-dihydrotestosterone treatment, confirming that the mutation renders the AR nonfunctional and can, therefore, be held responsible for the clinical features in these patients. These results highlight the importance of Arginine-614 in the second zinc finger of the DNA-binding domain of the AR in the protein-DNA interaction.

[Expression of androgen receptor gene in normal subjects and patients with complete androgen insensitivity]. / Expression du gène du récepteur des androgènes chez des sujets normaux et des patients présentant une insensibilité complète aux androgènes.

The Androgen Receptor (AR) mRNA was studied in cultured cells from normal subjects and 10 patients with Complete (CAI, 9 patients, 5: R-, 4: R+) or Partial (PAI, 1 patient) Androgen Insensitivity. The probe was a 3,4 cDNA coding for the entire AR. AR mRNA appears as a 10 kb band. It is strongly expressed in genital skin fibroblasts, 50 to 100 times less in non genital skin. In genital skin fibroblasts, a 3 to 5 fold increase is observed after 1 h of treatment of the cultures with dihydrotestosterone (DHT, 5 nM) whereas a 22 fold decrease is observed after 24 h. A 3 fold increase is also observed after 1 h of treatment with progesterone (50 nM) or cyproterone acetate (500 nM) which does not seem to act as an antiandrogen in this model. The 10 kb band was present in all 10 A1 patients studied, though expressed at a much lower level. It is therefore possible that an abnormal regulation of the AR gene expression is involved in the mechanism of Androgen Insensitivity.