L. Castagnetta

Expression of Different 17β-Hydroxysteroid Dehydrogenase Types and Their Activities in Human Prostate Cancer Cells1

The 17beta-hydroxysteroid dehydrogenase (17betaHSD) enzyme system governs important redox reactions at the C17 position of steroid hormones. Different 17betaHSD types (no. 1-4) have been identified to date in peripheral human tissues, such as placenta, testis, and breast. However, there is little information on their expression and activity in either normal or malignant prostate. In the present work, we have inspected pathways of 17beta-oxidation of either androgen or estrogen in human prostate cancer cells (LNCaP, DU145, and PC3) in relation to the expression of messenger RNAs (mRNAs) for 17betaHSD types 1-4. These cell systems feature distinct steroid receptor status and response to hormones. We report here that high expression levels of 17betaHSD4 were consistently observed in all three cell lines, whereas even greater amounts of 17betaHSD2 mRNA were detected solely in PC3 cells. Neither 17betaHSD1 nor 17betaHSD3 mRNAs could be detected in any cell line. From a metabolic standpoint, intact cell analysis showed a much lower extent of 17beta-oxidation of both androgen [testosterone (T)] and estrogen [estradiol (E2)] in LNCaP and DU145 cells compared to PC3 cells, where a greater precursor degradation and higher formation rates of oxidized derivatives (respectively, androstenedione and estrone) were observed. Using subcellular fractionation, we have been able to differentiate among 17betaHSD types 1-4 on the basis of their distinct substrate specificities and subcellular localization. This latter approach gave rise to equivalent results. PC3 cells, in fact, displayed a high level of microsomal activity with a low E2/T activity ratio and approximately equal apparent Km values for E2 and T, suggesting the presence of 17betaHSD2. Dehydrogenase specific activity with both E2 and T was also detected, although at lower levels, in LNCaP and DU145 cells. No evidence for reductase activity could be obtained in either the soluble or microsomal fraction of any cell line. As comparable expression levels of 17betaHSD4 were seen in the three cell lines, 17betaHSD2 is a likely candidate to account for the predominant oxidative activity in PC3 cells, whereas 17betaHSD4 may account for the lower extent of E2 oxidation seen in both LNCaP and DU145 cells. This is the first report on the expression of four different 17betaHSD types in human prostate cancer cells. It ought to be emphasized that for the first time, analysis of different 17betaHSD activities in either intact or fractionated cells harmonizes with the expression of relevant mRNAs species.

Steroid-growth factor interaction in human prostate cancer. 1. Short-term effects of transforming growth factors on growth of human prostate cancer cells

In order to better define potential mechanisms of growth regulation in human prostate cancer cells, we have compared biological responses (such as short-term response to both transforming growth factor alpha and beta; TFG alpha and TFG beta) in relation to hormone sensitivity of LNCaP, DU145, and PC3 cells. Androgen receptor (AR) and epidermal growth factor receptor (EGF-R) content of each cell line was also investigated. In addition, expression of EGF, TGF alpha, and TGF beta was evaluated through immunofluorescent staining. Growth of androgen non-responsive PC3 cells was stimulated by TGF alpha (about 35%) and inhibited by TGF beta (more than 50%), with respect to controls, after 48 h exposure. Conversely, AR-positive, hormone-responsive LNCaP cells proved to be poorly sensitive, at least short-term, to either growth factor. Furthermore, high levels of both EGF-R and TGF alpha, and a fairly high amount of EGF, were found in DU145 cells and, to a lesser extent, in LNCaP cells; in contrast, PC3 cells exhibited low expression levels of both receptors (EGF-R) and ligands (EGF, TGF alpha), but displayed remarkable TGF beta binding and relatively high levels of endogenous TGF beta. Overall, these results suggest a differential sensitivity to TGF alpha and TGF beta by prostate cancer cells; TGF alpha response seems not to be proportional to the EGF-R content of individual cell lines.

Evidence for soluble and nuclear site I binding of estrogens in human aorta

The purpose of this study was to establish the estrogen receptor (ER) expression and content in human aorta fragments removed at the time of by-pass surgery. To this end, we adopted a radioligand binding assay to evaluate either soluble (S) or nuclear (N) ER using dextran-coated charcoal (DCC) and filtration methods, respectively. To better define the intratissular distribution and content of ER, we also measured the presence of a 27 kDa heat shock protein (HSP27), a well established ER-associated protein, using D5 monoclonal antibody. Finally, we analysed the different molecular isoforms of both S and N ER using size exclusion-high performance liquid chromatography (SE-HPLC). High affinity (type I) sites of estrogen binding were detected in 17 out of 19 samples in either S or N fraction, although only 9 out of 19 cases displayed site 1 ER in both cell compartments. ER levels in aortic tissues, detected by radioligand method, compare well with those we have found in other hormone-sensitive human cancer tissues and cells. SE-HPLC analysis revealed two main receptor isoforms in the soluble fraction, having 65 kDa and 18 kDa molecular mass, while a minor component of 29 kDa was also found; the nuclear fraction displayed again two major components of 38 and 23 kDa. Using the HSP27 immunohistochemistry we observed a major staining occurring in smooth muscle cells (SMC), with an increasing intensity towards the lumen. All samples, including the ER negative ones, exhibited some degree of histochemical staining. Using an arbitrary cut-off value, 7 out of 12 samples displayed a highly positive staining, 6 of which showed nuclear ER. Furthermore, SE-HPLC separation indicated the presence of a 64.9 kDa component in the soluble fraction, according to the well known relative molecular mass of ER. Following HSP27 immunohistochemistry, the overall staining intensity in aortic SMC approaches that seen in endometrial and breast epithelia, whilst the muscle ER content is generally lower. Although our data are compatible with a direct role of estrogens in arterial function, the extent of the link with arterial disease remains to be established.

Product of aromatase activity in intact LNCaP and MCF-7 human cancer cells☆

We investigated conversion rates of androgens to estrogens in cultured, hormone-responsive prostate (LNCaP) and breast (MCF-7) human cancer cells. For this purpose, we adopted an intact cell analysis, whereby cells were incubated for different incubation times in the presence of close-to-physiological (1 nM) or supraphysiological (1 microM) concentrations of labelled androgen precursors, i.e. testosterone (T) and androstenedione (delta4Ad). The aromatase activity, as measured by estrogen formation, was detected in LNCaP cells (0.5 pmol/ml), even though to a significantly lower extent than in MCF-7 cells (5.4 pmol/ml), using 1 microM T after 72 h incubation. Surprisingly, LNCaP cells displayed a much higher aromatase activity when T was used as a substrate with respect to delta4Ad. In either cell line, T transformation to delta4Ad was relatively low, attaining only 2.8% in LNCaP and 7.5% MCF-7 cells. However, T was mostly converted to conjugates (over 95%), glucuronides and some sulphates, in LNCaP cells, whereas it was only partly converted to sulphates (<10%) in MCF-7 cells. Aromatase activity seems to be inconsistent in LNCaP cells, being strongly affected by culture conditions, especially by fetal calf serum (FCS). Further studies should assess the regulation of aromatase expression by serum or growth factors in different human cancer cells, also using anti-aromatase and/or anti-estrogen compounds, in different culture conditions.

Intra-tumoural variation of oestrogen receptor status in endometrial cancer

Soluble and nuclear oestrogen receptor status was determined in both the central and peripheral portions of tumour for 37 cases of adenocarcinoma of the endometrium. Of these, 29 had functional receptor in the peripheral biopsy, but only 19 retained functional receptor in the centre. Six of the 10 patients whose tumours showed this difference came from the group of 12 patients who were immediately post-menopausal (4.50 +/- 1.45 y post-menopausal age). Receptor status was not related to tumour classification into histological grades I and II. However, receptor-negative central biopsies were significantly more likely (P less than 0.05) to be Grade III. Early relapse was also related to a receptor-negative central biopsy.

The prognosis of breast cancer patients in relation to the oestrogen receptor status of both primary disease and involved nodes

Nodal involvement is accepted as the best single marker of prognosis in breast cancer. However, there is little information on the sub-division of node-positive patients according to the oestrogen receptor status of the nodal tissue. We have previously reported (Eur. J. Ca. 1987, 23, 31) that, in almost all cases, involved nodes are only oestrogen receptor positive (ER+) in patients whose primary tumours are uniformly ER+. This paper presents clinical follow-up on a larger group of patients with node positive breast cancer. For each patient, both soluble and nuclear receptor concentrations were determined in three separate parts of the primary tumour and in at least one involved node (we have previously defined tumours which contained ER in all six fractions of the primary as HS++, those lacking receptor in some fractions as HS+- and wholly receptor negative tumours as HS--). Median follow-up time was 71.5 months. As expected, patients whose tumours were HS++ had a significant (P less than 0.008) survival advantage. More importantly, patients with ER in both the soluble and nuclear fractions of their involved nodes survived significantly (P less than 0.003) longer than those with ER- nodes. Thus, full oestrogen receptor status of involved nodes will give sufficient prognostic information when adequate primary tissue is not available.

Prostate long-term epithelial cell lines : biological and biochemical features

This review reports studies on long-term prostate cell lines using multiple experimental approaches. The main goal was to investigate the metabolism of testosterone (T) through in vitro conversion rates. Extensive studies were also carried out on growth curves, tritiated thymidine incorporation, and morphometry by either hormone-responsive or hormone-unresponsive, normal and neoplastic human (PC3 and DU-145) and canine (CAPE and CPA) cell lines. All of them were characterized for their content of both soluble and nuclear androgen receptors. Receptor studies at site I binding in both soluble and nuclear fractions were carried out to establish the hormone sensitivity status of cells. In two prostate epithelial cells, steroid metabolic conversions in vitro show predominantly an oxidative metabolism of T, forming mainly androstenedione. Conversion rates were greater than 50% in the first 24 hours and still higher after 72 hours. At the same time and under exactly the same experimental conditions, the other cells showed metabolic pathways in which reductive metabolism prevails, dihydrotestosterone (DHT) being the prevalent metabolite. Different metabolic patterns of steroids of several cell lines relate to the hormone sensitivity status of the cells; steroid receptor-endowed cells are maintaining higher levels of unconverted precursor than are receptor-empty cells. In fact, hormone-sensitive cells, such as cancer canine CPA and human DU-145, produced DHT early through slowly converting T. On the contrary, unresponsive cells such as human cancer cells PC3 and normal canine CAPE quickly metabolize T, but DHT formation was not observed. These significant differences between cells are highly reproducible provided the proportion between cell number and molar concentration of precursors is constant. Differences we observe cannot be attributed to different experimental conditions. Cell viability, extraction efficiency, and all other parameters used for monitoring cell growth kinetics do not substantiate these reported significant differences in metabolic abilities of cells. The divergent steroid metabolic pathway we observe in different prostate long-term cells appears to be an intrinsic, consistent, highly reproducible property of each cell line.

Soluble and nuclear oestrogen receptor status of advanced endometrial cancer in relation to subsequent clinical prognosis.

Both soluble and nuclear oestrogen receptors have been measured in at least two separate sections from 72 endometrial cancers and 12 normal endometria. Concentration of oestrogen receptor is shown to be, in our hands, more meaningful when expressed per unit DNA than per unit protein, whether for soluble or nuclear receptor. Endometrial cancer cells from the central part of the tumour are shown to be receptor negative more frequently than those from peripheral tumour. Thus, in large cancers, biopsies from different areas are required before a tumour can be correctly designated as receptor positive, heterogeneous or receptor negative. The intratumoral variation of receptor status may relate to poor prognosis, since patients with homogeneous receptor-positive disease survive significantly longer than those with tumours showing either heterogeneous distribution of receptor or homogeneous absence of receptor. Intratumoral variation in receptor status is found to be more common in the group of patients who are within 7 years of their menopause, than in older patients.

Steroid Profiles and Optimization of High‐Performance Liquid Chromatographic Analytic Procedurea

This paper presents a short review of the results obtained to date in our laboratory, with respect to the studies of steroid excretion profiles in both breast and endometrial cancer patients, by using gas-liquid chromatographic analysis. These data demonstrate the importance of minor estrogens, including catechol estrogens, and of their ratios with the classical ones, in studies of steroid metabolism in both breast and endometrial cancer. New data concerning postmenopausal endometrial cancer are consistent with our previous observations and demonstrate the necessity of measuring these steroids directly in tumors and examining patterns of metabolism in vitro. In order to analyze steroid metabolic patterns in vitro, however, high-performance liquid chromatography rather than gas-liquid chromatography methods are preferable on account of their selectivity, specificity, sensitivity and capacity to handle labile materials. With the aim of providing methods suitable for the complete resolution and analysis of these complex natural mixtures a method of computer-aided optimization of HPLC has been developed and its practical utility has been tested.

Soluble and nuclear type I and II androgen-binding sites in benign hyperplasia and cancer of the human prostate.

SummaryThis paper presents an approach for the assessment of the androgen receptor (AR) status in benign prostatic hyperplasia (BPH) and prostate cancer (PCa) tissues. Evaluation of AR was carried out in both soluble and nuclear fractions by a standard competition method, using tritiated mibolerone as radioligand. Based on our experience with breast and endometrial cancer, this approach focused on both type I (high affinity, low capacity) binding sites, aiming mainly at establishing a putative “functional” receptor mechanism, i.e., the presence of type I AR in both cytosol and nucleus. Ancillary studies were carried out to exclude a potential overestimation of the AR content by interference with other steroid receptors, namely, progesterone (PgR) or glucocorticoid (GcR) receptors. Results showed that the interaction by PgR or GcR upon AR measurement was not relevant. The distribution of AR, namely the percent of positivity either in a single or in both cell compartments, was not significantly different in BPH (N=32) or PCa (N=24) tissues. For type I binding, the percent of positivity in both soluble and nuclear fractions (i.e., the “functional” AR status) was very close to that observed for other endocrine-related tumors, like breast cancer. Concentrations of type I AR appeared significantly higher in PCa than in BPH tissues; this was true for both soluble and nuclear fractions. In contrast, no significant difference was found in type II AR concentrations in either cell fraction. Nuclear type I AR proved to be positive in the great majority (more than 90%) of both PCa and BPH specimens; thus, this study does not support the hypothesis that nuclear AR may have a prognostic value, as previously suggested. Until long-term follow-up data on PCa patients are available, the predictive value of AR status, as estimated by this approach, cannot be assessed; however, in parallel with the studies carried out on estrogen receptor status in breast cancer patients, we suggest that a “functional” AR status is better indicated by the presence of type I binding in both cell fractions.