17-Estradiol Induces Apoptosis in the Developing Rodent Prostate Independently of ER or ER
http://www.100md.com
《内分泌学杂志》
Monash Institute of Reproduction and Development (R.A.T., P.C., G.P.R.), Monash University, Clayton, Victoria 3168, Australia
Department of Environmental and Molecular Toxicology (J.F.C.), North Carolina State University, Raleigh, North Carolina 27695
Receptor Biology Section (J.F.C., K.S.K.), Laboratory of Reproductive and Developmental Toxicology, National Institute of Environmental Health Sciences/National Institutes of Health, Research Triangle Park, North Carolina 27709
Abstract
Estrogens induce both proliferative and antiproliferative responses in the prostate gland. To date, antiproliferative effects of estrogens are generally considered to be due to systemic antiandrogenic actions. However, estrogen action mediated through estrogen receptor (ER) was recently suggested as another mechanism of induction of apoptosis in the prostate. This study aimed to explore the hypothesis that the antiproliferative effects of estrogen are directly mediated through ER using a prostate organ culture system. We previously reported effects of 17-estradiol (E2) using rat ventral prostate (VP) tissues, and adapted the system for culturing mouse tissues. In both rat and mouse models, estrogen-induced apoptosis was detected that was spatially and regionally localized to the epithelium of the distal tips. Using organ cultures of ER knockout (ERKO) and ERKO prostates, we failed to demonstrate that apoptosis induced by E2 was mediated through either receptor subtype. Activation of ER-selective ligands (ER, propyl pyrazole triol, ER, diaryl-proprionitrile, and 5-androstane-3,17-diol) in organ culture experiments failed to induce apoptosis, as did the membrane impermeable conjugate E2:BSA, discounting the possibility of nongenomic effects. Consequently, E2 regulation of androgen receptor (AR) expression was examined and, in the presence of nanomolar testosterone levels, E2 caused a specific reduction in AR protein expression in wild-type, ERKO, and ERKO mice, particularly in the distal region where apoptosis was detected. This down-regulation of AR protein provides a possible mechanism for the proapoptotic action of E2 that is independent of ERs or nongenomic effects.
Introduction
HIGH DOSES OF estrogens inhibit prostate growth both in vivo and in vitro. However, in vivo experiments are complicated by the fact that estrogens exert indirect effects on the prostate, by negative feedback on the hypothalamic-pituitary gonadal axis, thereby suppressing androgen production. Any reduction in androgen levels, androgen receptor (AR) expression, or androgen signaling will result in atrophy, decreased secretory activity, and apoptosis in the epithelium (1). In addition to indirect effects of estrogen in modulating androgen production, estrogens also have direct effects on the prostate itself. Evidence that estrogens act directly upon the prostate is implied because the estrogen receptors (ERs) and reside within the gland. The expression of the ERs has been described in the developing and adult prostate and they are differentially localized; ER is predominantly found in the stroma (2, 3), although it can be detected in the epithelium after estrogen exposure (4), and ER is predominantly found in the epithelium (3, 5, 6).
Estrogens induce both proliferative and antiproliferative effects in the prostate. Several studies have identified that ER is responsible for mediating proliferative actions by estrogens on the prostate during development and adulthood (4, 5). A previous study by Prins et al. (5) used ER knockout (ERKO) and ERKO mice to demonstrate that ER was the dominant ER form mediating developmental estrogenization of the prostate gland. Squamous metaplasia is another biological response induced by exogenous estrogen exposure that is mediated via ER as demonstrated by tissue recombination studies (4).
Although ER is the predominant epithelial ER residing in the prostate, a biological endpoint mediated through this receptor remains to be defined. It was postulated that direct antiproliferative effects of estrogens in the prostate are mediated through ER (7, 8, 9, 10). This hypothesis is based on several lines of correlative evidence. Firstly, mice deficient in ER (ERKO) were reported to demonstrate foci of epithelial prostate hyperplasia upon aging that was associated with higher proliferative and lower apoptotic indexes (8, 9, 10), although this phenotype was not confirmed by other laboratories (5, 11, 12). Secondly, administration of antiestrogens (tamoxifen and ICI 182,780) to the human prostate cancer cell line, DU145, which express only ER, and not ER, caused growth inhibition (13) and cotreatment of DU145 cells with ER-antisense oligonucleotide reversed the antiproliferative effects induce by ICI 182,780 (13), suggesting that ER mediated growth-inhibitory actions of antiestrogens in tumor cells. Thirdly, ER binds to electrophile/antioxidant response elements that stimulate induction of genes such as quinone reductase and gluthanthione S-transferase that are protective against carcinogenesis (14). Although these associated data infer antiproliferative actions mediated by ER, direct evidence is not available.
The problem with demonstrating that ER induces antiproliferative or proapoptotic responses relates to the exquisite sensitivity of the prostate to changes in androgens, or androgen signaling, associated with estrogen administration in vivo. Therefore, to circumvent the centrally mediated effects of estrogens in decreasing serum androgen levels, we have developed an organ culture system where prostate lobes are grown in a defined hormonal environment, so that androgen levels are maintained, and direct effects of estrogens can be examined. We previously used this system using rat ventral prostate (VP) lobes (15). Here we report for the first time that this organ culture model can be adapted to allow maintenance of mouse prostate lobes in vitro, thereby permitting the use of genetically altered mouse models in such experiments. In this study, we have used ER gene-targeted models (ERKO and ERKO mice) to study the specific role of each receptor subtype in mediating antiproliferative effects of estrogen on the developing prostate. In addition, we have adapted a complementary, alternative approach using several ER subtype-selective ligands to delineate ER-mediated events. The aim of this study was to test the hypothesis that ER activation directly mediates apoptotic effects in the developing rodent prostate, independent of regulation of androgen levels.
Materials and Methods
Animals and tissue collection
All animal handling and procedures were carried out in accordance with National Health and Medical Research Council guidelines for the Care and Use of Laboratory Animal Act and according to the Animal Experimentation and Ethics Committee at Monash Medical Centre (Clayton, Australia). Sprague Dawley outbred rats (d 0), (ERKO [C57BL/6J black mice (16)], ERKO [C57BL/6J black mice (17)] and normal C57BL/J6 mice were killed on d 1–2 after birth. VP lobes were microdissected for organ culture.
Organ culture
Organ culture was carried out as previously described in rat (15, 18), although there were some modifications to satisfy mouse organ culture conditions. Pilot studies in C57BL/6J mice demonstrated that mice needed to be at least postnatal d 1 (weight range 1.5–2.0 g) to undergo normal branching morphogenesis (data not shown). Briefly, all animals from each litter (ERKO and ERKO colonies) were cultured blind because the genotypes were unknown at the beginning of the culture. Mice were killed by decapitation and tail snips collected from each of the KO mice for genotyping.
Tissue collection for mouse and rat tissues involved microdissection of VPs (rat lobes collected independently and mouse lobes collected as two lobes together). Each explant was cultured on Millicell CM filters (Millipore Corp., Bedford, MA) floating on 500 μl of nutrient media in a four-well plate at 37 C in a humidified 5% CO2 incubator. A basal medium of DMEM/Ham’s F-12, 1:1 (vol/vol) lacking phenol red, supplemented with insulin (10 μg/ml) and transferrin (10 μg/ml) was used in all experimental groups. The medium was supplemented with 10 nM testosterone (T), and treatment groups were cultured with high doses of 17-estradiol (E2; 15 μM) together with 10 nM T or E2 alone. Other estrogenic compounds were tested including 3Adiol (5-androstane-3,17-diol; putative ER ligand), E2:BSA (Sigma-Aldrich, St. Louis, MO), an ER-selective ligand [1,35-(4-hydroxyphenyl)-4-propyl-1H pyrazole; propyl pyrazole triol; PPT]; Tocris Cookson Ltd., Avonmouth, UK), and an ER-selective ligand [2,3-bis-(4-hydroxyphenyl)-propionitrile; diaryl-propionitrile; DPN; Tocris]. Steroids were tested at a range of doses (10–5 M to 10–13 M) and vehicle controls were conducted. At least four explants of each genotype were subjected to each hormonal treatment.
The organs were harvested after 5 d of culture. Explants were photographed, fixed in Bouins fixative for 1 h at room temperature, and then processed to paraffin for histological and stereological analysis.
Histology
Fixed tissues were processed and embedded in paraffin for histological analysis. Care was taken to ensure tissues were embedded so that longitudinal sections could be obtained. Serial 5-μm sections were cut on a microtome and dried onto Superfrost Plus+ slides (Menzel-Glazer, Braunschweig, Germany) before histological and stereological examination. Hematoxylin and eosin (H&E) staining was performed on every fifth section to examine the pathology throughout the whole tissue.
Detection of apoptosis
Apoptosis was analyzed by ApopTag In Situ Apoptosis Detection Kit (Integren, Purchase, NY). Briefly, sections were dewaxed, incubated in Equilibration Buffer (Intergren), then treated with TdT enzyme in Reaction Buffer (Integren) for 1 h at 37 C. Sections were washed in Stop Wash Buffer (Integren) for 30 min at 37 C, treated with 3% (vol/vol) H2O2 in methanol for 15 min, and blocked with CAS block (Zymed Laboratories Inc., South San Francisco, CA). Apoptotic cells were detected by antidigoxigenin conjugate (Integren) for 30 min at room temperature and color reacted with 3,3'-diaminobenzidine tetrahydrochloride (liquid substrate kit; Zymed). The reactions were stopped with water and sections were counterstained with Mayer’s hematoxylin, dehydrated, cleared and mounted.
Immunohistochemistry
Immunolocalization of AR and ERs (ER) and (ER) was detected using the Dako Autostainer Universal Staining System (Dako A/S, Carpinteria, CA). Immunostaining was conducted on n = 3 animals from each group, including n = 10 sections (i.e. every fifth or sixth section, depending on the tissue size) from each tissue to sample throughout the explant. Immunoreactivity for AR, ER, and ER was detected after sections were subjected to microwave antigen retrieval in 0.01 M citrate buffer (pH 6.0), boiling for 20 min. Antigen retrieval for caspase-3 immunoreactivity was carried out in 0.01 M sodium citrate buffer (pH 6.0), boiling for 10 min. All sections were then treated with peroxidase blocking reagent (Dako Envision System kit; Dako) for 30 min, and nonspecific binding was blocked using CAS block for at least 30 min (Zymed). Monoclonal antibody AR (N-20) (Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a working concentration of 0.33 μg/ml or concentration-matched rabbit IgG (Dako) for 1 h at room temperature. Monoclonal ER (1D5) (Dako Corp., Carpinteria, CA) and monoclonal antibody ER (NCL-ER-) (Novocastra Laboratories Ltd., UK) at working concentrations of 14 μg/ml and 1 μg/ml, respectively, for 2 h at room temperature. Polyclonal antibody cleaved caspase-3 (Asp175) (Cell Signaling Technology, Inc., Beverly, MA) at a working concentration of 1 μg/ml for 2 h at room temperature. Antibodies were detected by incubation with peroxidase-labeled polymer (Dako Envision System; Dako), which is conjugated to antimouse and antirabbit Igs, for 15 min at room temperature and then color reacted with 3,3'-diaminobenzidine tetrahydrochloride (liquid substrate kit) for 5 min. The reactions were stopped in water, and sections were counterstained with Mayer’s hematoxylin, dehydrated, cleared, and mounted.
Semiquantitation of apoptosis
The incidence of apoptosis was estimated based on a method that allowed an unbiased semiquantitation of the percentage of apoptotic cells in both treated and control samples (19). Tissue sections were mapped at x40 magnification to define tissue boundaries. Random fields were systematically selected by computer-assisted software: CAST version 1.10 software (Computer Assisted Stereological Toolbox) (Olympus Danmark A/S, Albertslund, Denmark), and sampling was conducted using a three-by-two unbiased counting frame. Frame counting was performed on 10 sections uniformly spaced throughout the tissue, n = 3 for each group, with an average of 1500 cells counted per section. Cells were classified as either apoptotic or nonapoptotic-based on Apoptag labeling (only fully stained apoptotic cells that exhibited histological characteristics typical of cells undergoing programmed cell death were counted as positive, thereby avoiding counts of end stage apoptotic bodies) and represented as percentage of total cells. All data were expressed as the mean ± SEM. Control and treatment groups were compared using a two-tailed paired t test, with the significance threshold employed at a level of 5% (P < 0.05). All data analyses were conducted using Prism 2.01 software (GraphPad Software, Inc., San Diego, CA).
Results
Effect of E2 on rat prostate development in vitro
Organ culture of rat VP in the presence of T (10 nM) has been described previously (15), mimicking in vivo development in vitro (Fig. 1A). In addition, we previously demonstrated that culture of neonatal rat VPs in the presence of T plus E2 (15 μM) retards growth and reduces the extent of branching morphogenesis (15) (Fig. 1B). Administration of T plus E2 to neonatal rat VPs in vitro altered stromal cell differentiation, particularly smooth muscle development, as well as perturbed epithelial differentiation (15). Figure 1, C and D, shows that, in explants of rat VP, reduced growth is associated with an increase in the incidence of apoptosis in organs cultured in the presence of T plus E2, compared with T alone, as detected by ApopTag labeling. The incidence of apoptosis was very low in media containing T alone (Fig. 1C), but in the presence of T plus E2, there was a significant increase in the incidence of apoptosis that was predominantly found in the differentiating epithelial cells of the distal region, where branching morphogenesis and epithelial cord formation occurs (Fig. 1D). The incidence of apoptosis was confirmed using a second detection technique, specifically immunolocalization of cleaved caspase-3 (one of the key executioners of apoptotic pathway). Using this approach, a similar incidence of apoptosis was observed after T alone and T plus E2 (data not shown). The incidence of apoptosis was estimated using a semiquantitation technique, which revealed an approximately 8-fold increase in the percentage of cells undergoing apoptosis after T plus E2 (5.25 ± 0.17%) treatment compared with T alone (0.65 ± 0.03%; Table 1).
Establishment of mouse VP organ culture; comparison to in vivo tissues
To perform experiments using genetically altered mice, we adapted the organ culture technique for mouse prostate and report here in vitro development of mouse VPs for the first time. Specifically, we used older mouse tissues (d 1–2 postnatal rather than day of birth), and cultured paired organs connected by shared mesenchyme, rather than isolated lobes. Using these regimes, we were able to maintain the integrity of the glands for up to 5 d of culture and the development of explants cultured in vitro was similar to that observed in age-matched in vivo tissues (Fig. 2, A–J). Tissues were isolated on d 1 after birth, and each VP lobe consisted of two main ducts that had initiated branching (Fig. 2A). A histological section of these tissues revealed undifferentiated epithelial cords invading the surrounding mesenchyme (Fig. 2B). To confirm that the development we observed in vitro mimicked that observed in vivo, we isolated mouse VP lobes from postnatal d 4 (Fig. 2C) and 8 (Fig. 2D). Over the first 6–8 d in vivo, mouse VPs undergo extensive growth and significant branching morphogenesis occurs (Fig. 2, C and D). Histologically, postnatal d 8 VPs exhibit proximal to distal maturation of the ducts with some canalization and maturation in the proximal region, whereas the distal tips remain undifferentiated and continue budding into the surrounding mesenchyme (Fig. 2E). There is little lumen formation at the stage of development. By comparison, mouse VPs that were isolated at d 1–2 (Fig. 2, A and B) were maintained in vitro in the presence of T alone for 5–6 d (equivalent of in vivo d 8) and underwent similar development involving branching morphogenesis and ductal formation (Fig. 2, F and G). The main difference between these in vitro explants and the equivalent in vivo tissues is architecture of the lobes; in vivo tissues maintain three-dimensional structure, whereas the in vitro explants tend to grow in a flattened manner, generating a more two-dimensional shape. Nonetheless, histological analysis of the in vitro explants revealed similar maturation and differentiation to comparative in vivo explants, although more advanced canalization and lumen formation (Fig. 2H). Differentiation and organization occurred in a proximal to distal manner, with the most advanced development observed in the proximal region and newly forming ductal structures in the distal region (Fig. 2, I and J). In the proximal region, the epithelium became polarized and secretory epithelial cells were observed in addition to the underlying basal cells (Fig. 2I), whereas the epithelium was actively differentiating and proliferating in the distal tips (Fig. 2J).
Effect of E2 on wild-type (WT) mouse VP growth in vitro
In the presence of 10 nM T, the growth inhibitory effects of E2 were dose responsive over a concentration range 10 nM to 20 μM (Fig. 3, A–F), consistent with previous observations (15). High E2 concentrations (micromolar) significantly reduced prostate growth, whereas lower concentrations (nanomolar) reduced the inhibitory effect such that explants were not significantly different from control explants. Similar to previous studies in rat VPs (15), a concentration of 15 μM was chosen to further investigate the mechanism of estrogen-induced apoptosis in the mouse VP.
Mouse VPs from WT mice were cultured in the presence of T alone (10 nM) and normal growth was observed in vitro, as described above (Fig. 4A). When cultured in the presence of T plus E2 (15 μM), branching morphogenesis was reduced (Fig. 4B) so that there were generally fewer secondary and terminal branches; the main branches were elongated and distended, rather than forming new branches. In some regions, these ducts were dilated and cystic (Fig. 4B). The incidence of apoptosis in mouse VPs after culture in the presence of T alone, and T plus E2 is shown in Fig. 4, C–F. An examination of proximal and distal regions revealed very few apoptotic cells after culture in the presence of T alone (Fig. 4C). In contrast, treatment with T plus E2 induced a significant amount of apoptosis that was specifically in the epithelium of the distal tips (Fig. 4D). These observations in the WT mouse were similar to the responses observed in the rat VP organ cultures (Fig. 1). Interestingly, apoptotic cells were noted adjacent to cells actively undergoing mitosis, indicating that proliferation was concurrent with cell death (Fig. 4, C and D). The incidence of apoptosis was confirmed using immunolocalization of cleaved caspase-3 and using this approach, a similar incidence of apoptosis was observed after T alone (Fig. 4E) and T plus E2 (Fig. 4F). Semiquantitation revealed a significant increase in the percentage of cells undergoing apoptosis in the mouse VPs, at similar levels to that observed in the rat tissues (T alone, 0.54 ± 0.13% vs. T plus E2, 5.10 ± 1.13%; Table 1).
Localization of ERs
After culture in the presence of T alone, regional variation in the expression of ERs was observed. ER immunostaining was restricted to individual cells within the proximal region, specifically in mesenchymal cells (Fig. 5, A and B). In contrast to ER immunoreactivity, ER immunoreactivity was observed in both the proximal and distal regions. Proximally, nuclear ER staining was localized to smooth muscle, stromal and epithelial cell types (Fig. 5C). Comparatively, in the distal region ER immunoreactivity was predominantly localized to the smooth muscle and stromal cells; however, weak basal epithelial cell staining was apparent (Fig. 5D). After 5 d of culture in the presence of T plus E2, immunolocalization of ER was down-regulated so that it was low/undetectable in both proximal and distal regions (Fig. 5E), whereas ER was up-regulated, particularly in the epithelial cells in the distal regions of the gland (Fig. 5F), where the majority of apoptotic cells were observed. There was no apparent difference in stromal cell expression of ER throughout the gland when compared with T-treated controls after E2 treatment. This apparent increase in ER, together with the concurrent expression in the region where cells were undergoing apoptosis, further supported the idea that estrogen-induced apoptosis may be mediated through this receptor. Therefore, this hypothesis was tested in ERKO mouse VPs.
Effect of E2 on ERKO and ERKO mouse VP growth in vitro
After 5 d of culture with T alone, ERKO and ERKO VPs had apparently normal branching morphogenesis and no gross abnormalities were observed (Fig. 6, A and B). In the presence of T alone, ERKO and ERKO mouse VPs showed a very low incidence of apoptosis throughout, after examination of proximal and distal regions (Fig. 6, C and D; and Table 1), similar to WT. After culture in the presence of T plus E2, ERKO and ERKO VPs responded similar to WT mice, with reduced branching and distended duct formation (Fig. 6, E and F). In ERKO mice, growth was more reminiscent of explants cultured with T alone, although some regions still had ducts that were dilated and cystic (Fig. 6E). However, in the presence of T plus E2, both ERKO and ERKO mice exhibited a significant increase in the incidence of apoptosis, particularly in the distal epithelial tips (Fig. 6, G and H), indicating that estrogen-induced apoptosis still occurred in the absence of functional ER or ER. This increase in apoptosis in ERKO and ERKO mouse VPs after T plus E2 treatment was confirmed by immunolocalization of cleaved caspase-3 (data not shown) and semiquantitation (Table 1).
Effect of ER- and ER-selective ligands on mouse prostate development in vitro
To confirm whether estrogen-induced apoptosis was mediated through either ER or ER, we investigated the biological response to ER- and ER-selective ligands on mouse prostate development. WT C57BL/J6 mice were used for this study. To validate the use of the subtype-specific ER ligands, vehicle only controls were conducted and explants were similar to T-treated explants (data not shown). In the presence of T alone, explants developed similar to that observed in vivo, and extensive branching morphogenesis occurred (Fig. 7A). Histologically, the epithelium became polarized and secretory epithelial cells were observed in addition to the underlying basal cells, concurrent with differentiation and organization of the surrounding stromal tissue (Fig. 7B). The incidence of apoptosis was low, consistent with previous observations (Fig. 7C).
When cultured in the presence of T plus an ER-specific ligand (PPT; 10 μM), explants showed a striking reduction in growth (Fig. 7D). The size of the explants was significantly smaller and there was less branching morphogenesis. Concurrent with the reduction in growth of PPT-treated explants, histological examination by H&E staining revealed significant morphological alterations with ductal branching, elongation and canalization interrupted throughout the entire gland (Fig. 7E). The spatial and regional organization was not apparent, and the ductal architecture was severely altered. Similar to T-treated controls, the incidence of apoptosis was also low because very few cells were detected by ApopTag labeling (Fig. 7F) or immunolocalization of cleaved caspase-3 (data not shown).
After 5 d of culture with T plus an ER-specific ligand (DPN; 10 μM), mouse VPs were larger than explants treated with T alone (Fig. 7G). This enlargement appeared to be the result of excessive dilation of the ducts that may be associated with increased production of secretory products. H&E examination revealed several differences in cellular morphology when compared with T-treated control explants (Fig. 7H). In T plus DPN-treated explants, branching morphogenesis appeared reduced with generally fewer secondary and tertiary branches. The main ductal branches appeared elongated and distended with large lumen spaces. Epithelial cell differentiation appeared well developed throughout the entire tissue, with pseudostratified columnar epithelium lining the majority of the ductal length (Fig. 7H). Similarly to T and T plus PPT (ER-specific ligand)-treated explants, the incidence of apoptosis remained very low in T plus DPN (ER-specific ligand)-treated explants (Fig. 7I).
Effect of 3Adiol on rodent prostate development in vitro
The T metabolite 3Adiol is reported to be a specific ligand for ER in the prostate (9, 20). To test whether 3Adiol was able to induce apoptosis in the developing rodent prostate similar to E2, organ culture experiments were conducted on both rat and mouse VPs. To validate the use of 3Adiol, vehicle-only controls were conducted and explants were similar to T-treated explants (data not shown). Results presented represent the rat organ culture explants (Fig. 8, A–D), but these effects were also observed in organ cultures using C57BL/J6 mouse ventral prostate lobes (data not shown). In the presence of T, 3Adiol was tested at several doses (including 20 μM, 10 μM, and 1 μM), similar to the dose range required for E2 to induce apoptosis. At the highest dose (20 μM), 3Adiol caused little change in prostate size and growth (Fig. 8, A and B). The extent and pattern of branching morphogenesis was not altered and development occurred similar to explants cultured with T alone (Fig. 8A). In terms of apoptosis, the incidence of programmed cell death in control explants cultured with T alone was very low, with few cells immunolabeled with the ApopTag detection kit (Fig. 8C). Similarly, the incidence of apoptosis in explants cultured in the presence of T plus 3Adiol (20 μM) was also very low (Fig. 8D).
Effect of E2:BSA on mouse prostate development in vitro
To test whether estrogen-induced apoptosis was the result of nongenomic estrogenic actions, we determined the effect of a membrane impermeable conjugate of E2, namely E2:BSA. Results presented represent the mouse C57BL/J6 organ culture explants (Fig. 8, E–H). In the presence of T plus E2:BSA, branching morphogenesis was similar to T-treated control explants (Fig. 8, E and F). In terms of apoptosis, the incidence of programmed cell death in control explants cultured with T alone was very low, with few cells immunolabeled with the ApopTag detection kit (Fig. 8G). Similarly, the incidence of apoptosis in explants cultured in the presence of T plus E2:BSA was also very low (Fig. 8H).
AR expression
Because the estrogen-induced apoptosis observed in both mouse and rat organ culture experiments was not shown to be mediated through ER or ER by two experimental models (using ERKO mice and ER-selective ligands), we investigated the possibility that estrogen was acting indirectly via modulation of AR expression after treatment with T alone and T plus E2. After culture in the presence of T alone, AR immunolocalization was detected in both the mesenchymal and epithelial cells throughout WT VP explants (Fig. 9, A and B). There was variability in AR immunostaining along the prostatic ducts within the proximal and distal regions. In the distal region, where most proliferation and differentiation occurred, AR immunostaining was strong and nearly all cells were immunopositive for AR (Fig. 9A). In the more differentiated proximal region, where epithelial polarization and lumen formation occurred, there were some cells, particularly in the epithelium that were immunonegative for AR (Fig. 9B). In contrast, culture in the presence of T plus E2 using WT mice resulted in a significant decrease in AR immunolocalization in both the distal (Fig. 9C) and proximal (Fig. 9D) regions, particularly in the epithelial cells. AR expression was still detectable in epithelial and mesenchymal cells; however, the number of cells that were immunopositive and the staining intensity were reduced (Fig. 9C). The decrease in the proximal region was not as significant (Fig. 9D) compared with the proximal region of explants treated with T alone. ERKO and ERKO mouse VPs treated with T alone showed comparable AR expression to WT VPs (Fig. 9, E and G). Similarly, ERKO and ERKO VPs cultured in the presence of T plus E2 also showed significant down-regulation of AR in the distal epithelium as detected by immunolocalization (Fig. 9, F and H).
Discussion
In this study, we have adapted a serum-free prostate organ culture technique to maintain neonatal murine tissues in culture. This system maintains stromal-epithelial interactions within the prostate, as well as allowing prostate growth in defined hormonal environments, where androgen levels can be kept constant. The adaptation of this model to allow organ culture of mouse prostate tissues expands the utility of the system to allow the use of genetically altered mouse models in such studies, in this case, ERKO and ERKO mice.
Here we report that the addition of high-dose E2 resulted in antiproliferative events including active induction of apoptosis in the prostate epithelium, particularly in the distal tip epithelium, where ER was previously localized (21). Data presented here also indicated that ER expression was increased in the distal tip epithelium after E2 treatment. However, using ERKO mice, we failed to demonstrate a specific role for ER or ER in mediating apoptosis because apoptosis was consistently observed after E2 exposure to prostate glands from either ERKO or ERKO mice. Ideally, we would prefer to test the effects of E2 in ERKO tissues, but because each litter only provides 1:16 double KO males, this experiment was not feasible. Instead, we adapted an alternative approach using ER-specific ligands PPT (ER) or DPN (ER) and 3Adiol (ER) and showed these agonists were also unable to induce the apoptotic response. These data implied that apoptosis was caused by a non-receptor-mediated event or was mediated through another indirect estrogen signaling mechanism. We also tested whether the observed apoptosis was due to nongenomic effects of estrogens; however, organ culture experiments using the membrane impermeable conjugate E2:BSA failed to demonstrate an effect.
However, subsequent data revealed a down-regulation of AR, which may provide an alternative explanation for the observed apoptosis in the prostate. AR regulation was independent of androgen levels because T was maintained in the culture media, suggesting there was direct modulation of AR protein expression by E2. Our data suggested that AR down-regulation occurred independently of either ER subtypes because the effect is observed in ERKO and ERKO prostates. We did not quantitate the decrease in AR protein by Western analysis because examination of whole tissue homogenates would mask the changes in cell-specific expression; using immunohistochemistry we observed a marked loss of epithelial AR. The induction of epithelial cell apoptosis by castration (androgen removal) has been well documented (22, 23) and tissue recombination studies eloquently demonstrated that this epithelial response is a paracrine event mediated through the loss of stromal ARs, rather than reduced stimulation of epithelial ARs (24). However, in this study we observed a predominant loss of epithelial AR, rather than stromal AR and so it is not convincing that loss of AR protein is the only factor involved in the apoptotic response.
Nonetheless, the regulation of AR protein expression by estrogen is important to consider. Pharmacological doses of estrogen have been shown in multiple systems, including the one presented here, to decrease AR levels (5, 9, 10, 25) although the mechanism by which this occurs is not well understood. Prins et al. (5) described altered AR protein expression by estrogens that was mediated through ER and partially ER, using an in vivo treatment regime. Further studies by the same group have indicated that estrogen alters AR expression posttranscriptionally (via protein degradation), whereas AR mRNA levels may be unaltered (26). Another suggestion by Weihua et al. (9) was that the AR protein expression is regulated by ER, although a definitive mechanism has not been described. Experiments using primary cultures of lizard testis cells (27) and Harderian gland cells (28) have demonstrated that estrogen also down-regulates AR mRNA expression. A recent report revealed a putative estrogen response element in the Hamster AR promoter mRNA (5'-untranslated region), although this consensus does not appear to be conserved in human or rodent AR mRNA (29). Clearly, the link between AR expression and estrogen exposure, at either the protein or mRNA level, requires further investigation. Nevertheless, the down-regulation of AR protein, and therefore androgen function by estrogen, may in part be responsible for the apoptotic response observed here. Interestingly, previous reports using ERKO mice have described elevated expression of AR in the prostate (9, 10), but we did not observe this finding in the current study.
Estrogen acts by multiple signaling pathways in addition to the traditional steroid hormone receptor pathway including biological responses that are 1) ligand independent, where other factors such as growth factors stimulate ERE in the absence of ligand, or 2) ERE-independent actions, where the ligand/receptor complex targets genes by binding to other DNA-bound transcription factors forming a complex that alters downstream gene regulation independent of EREs (30). Nongenomic effects also occur, where estradiol activates putative membrane-associated binding sites, linked to intracellular signal transduction pathways to generate rapid tissue responses (30, 31, 32). In contrast to the genomic effects exerted by estrogens and ERs, nongenomic effects occur very rapidly, even seconds to minutes after estrogen exposure, to induce activation of intracellular second messengers such as calcium, nitric oxide formation, activation of kinases such as tyrosine kinases, protein kinase A, protein kinase C, ERK, and protein kinase B (33). Such effects have been described in neuronal, vascular, and bone systems (34). No evidence for nongenomic effects of estrogen in the induction of apoptosis in the prostate could be detected in this study using the membrane impermeable conjugate E2:BSA.
As a complementary, alternative approach to the use of ERKO mice to study ER-mediated events, we used the recently developed, novel ER subtype-selective ligands. The compound PPT is a potent ER agonist that does not activate ER. This results from the fact that it binds with high affinity and 400-fold preference to ER, and demonstrates almost no binding for ER (35, 36). In contrast, the compound DPN is a potency-selective agonist for ER with a more than 70-fold higher binding affinity for ER than ER (37). These compounds have been used in vivo to examine ER-mediated responses in the mouse uterus (38) and pituitary (39, 40). In the current in vitro study, the ER subtype-selective ligands failed to induce apoptosis in the developing mouse prostate; however, differential biological responses on branching morphogenesis and cellular differentiation were observed; ER agonist reduced prostate growth, disrupted the spatial organization of the gland, and altered cell differentiation in both epithelial and stromal compartments, whereas ER appeared to cause dilation of the ducts as a result of enhance cellular differentiation and increased secretory activity. Future studies using these selective ligands will further our understanding of ER-mediated events in the prostate other than apoptosis.
The putative ER-specific ligand is 3Adiol was hypothesized to be a preferred ligand for ER in vivo (9, 10). 3Adiol is a 5-dihydrotestosterone metabolite that competes with E2 for binding to ER and elicits estrogenic responses (10, 41). In this study, we tested the effects of 3Adiol on the developing rodent prostate in vitro and failed to induce apoptosis. This result indicated that even if 3Adiol is the preferred ligand for ER, it was unable to elicit an apoptotic response in the prostate gland.
Despite conflicting data fetal exposure to low doses of estrogens has been shown to induce permanent alterations in the prostate gland, including increased adult prostate weight (42, 43, 44). Neonatal exposure to high pharmacological doses of estrogens has also been shown to elicit effects on the gland, including permanent suppression of prostate growth and induction of epithelial hyperplasia in adulthood (25). Studies examining the effects of high doses of estrogens on the prostate have predominantly been carried out in vivo, where the direct effects of estrogens are confounded by the down-regulation of the hypothalamic-pituitary-gonadal axis after exogenous estrogen administration (1, 45). In these in vitro organ culture experiments, we used micromolar concentrations of E2, not unlike the pharmacological doses of estrogens given to rodents in developmental toxicology studies (46, 47) and men being treated for prostate cancer in the clinic (48). A previous report using MCF-7 breast cancer cells showed that, whereas lower concentrations stimulated cell proliferation, high concentrations (>10 μM) of E2 inhibited proliferation and induced apoptosis by non-ER-mediated actions (49), similar to that reported here. The mechanism by which high concentrations induce biological changes independent of ERs remains to be determined.
In summary, we demonstrated a direct effect of E2 on the prostate that includes the induction of apoptosis in rat and mouse organ cultures that was estrogen dependent, but not mediated through ER or ER, as determined using ERKO mice and ER-selective ligands. Although androgen levels were maintained in the culture media, AR protein expression was down-regulated, and may provide a possible explanation for the estrogen-induced apoptosis observed. Overall, this study failed to support the hypothesis that ER directly mediates apoptosis in the developing prostate, but raises the possibility of antiproliferative actions in the prostate induced by high concentrations of estrogens through non-receptor-mediated pathways.
Acknowledgments
We thank Hong Wang and Ann Davies (Monash Institute of Medical Research, Monash University) for technical assistance.
Footnotes
First Published Online October 13, 2005
Abbreviations: 3Adiol, -Androstane-3,17-diol; AR, androgen receptor; DPN, diaryl-proprionitrile; E2, 17-estradiol; ER, estrogen receptor; ERKO, ER knockout; ERKO, mice deficient in ER; ERKO, mice deficient in ER; H&E, hematoxylin and eosin; PPT, propyl pyrazole triol; T, testosterone; VP, ventral prostate; WT, wild type.
Accepted for publication October 3, 2005.
References
Wright AS, Thomas LN, Douglas RC, Lazier CB, Rittmaster RS 1996 Relative potency of testosterone and dihydrotestosterone in preventing atrophy and apoptosis in the prostate of the castrated rat. J Clin Invest 98:2558–2563
Prins GS, Birch L 1997 Neonatal estrogen exposure up-regulates estrogen receptor expression in the developing and adult rat prostate lobes. Endocrinology 138:1801–1809
Omoto Y, Imamov O, Warner M, Gustafsson JA 2005 Estrogen receptor and imprinting of the neonatal mouse ventral prostate by estrogen. Proc Natl Acad Sci USA 102:1484–1489
Risbridger G, Wang H, Young P, Kurita T, Wang YZ, Lubahn D, Gustafsson JA, Cunha G, Wong YZ 2001 Evidence that epithelial and mesenchymal estrogen receptor- mediates effects of estrogen on prostatic epithelium. Dev Biol 229:432–442
Prins GS, Birch L, Couse JF, Choi I, Katzenellenbogen B, Korach KS 2001 Estrogen imprinting of the developing prostate gland is mediated through stromal estrogen receptor : studies with ERKO and ERKO mice. Cancer Res 61:6089–6097
Adams JY, Leav I, Lau KM, Ho SM, Pflueger SM 2002 Expression of estrogen receptor in the fetal, neonatal, and prepubertal human prostate. Prostate 52:69–81
Chang WY, Prins GS 1999 Estrogen receptor-: implications for the prostate gland. Prostate 40:115–124
Weihua Z, Warner M, Gustafsson J 2002 Estrogen receptor in the prostate. Mol Cell Endocrinol 193:1–5
Weihua Z, Makela S, Andersson LC, Salmi S, Saji S, Webster JI, Jensen EV, Nilsson S, Warner M, Gustafsson JA 2001 A role for estrogen receptor in the regulation of growth of the ventral prostate. Proc Natl Acad Sci USA 98:6330–6335
Imamov O, Morani A, Shim GJ, Omoto Y, Thulin-Andersson C, Warner M, Gustafsson JA 2004 Estrogen receptor regulates epithelial cellular differentiation in the mouse ventral prostate. Proc Natl Acad Sci USA 101:9375–9380
Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors (ER) and (ER) on mouse reproductive phenotypes. Development 127:4277–4291
Couse JF, Curtis Hewitt S, Korach KS 2000 Receptor null mice reveal contrasting roles for estrogen receptor and in reproductive tissues. J Steroid Biochem Mol Biol 74:287–296
Lau KM, LaSpina M, Long J, Ho SM 2000 Expression of estrogen receptor (ER)- and ER- in normal and malignant prostatic epithelial cells: regulation by methylation and involvement in growth regulation. Cancer Res 60:3175–3182
Montano MM, Jaiswal AK, Katzenellenbogen BS 1998 Transcriptional regulation of the human quinone reductase gene by antiestrogen-liganded estrogen receptor- and estrogen receptor-. J Biol Chem 273:25443–25449
Jarred RA, Cancilla B, Prins GS, Thayer KA, Cunha GR, Risbridger GP 2000 Evidence that estrogens directly alter androgen-regulated prostate development. Endocrinology 141:3471–3477
Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach KS, Smithies O 1993 Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci USA 90:11162–11166
Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korach KS, Gustafsson JA, Smithies O 1998 Generation and reproductive phenotypes of mice lacking estrogen receptor . Proc Natl Acad Sci USA 95:15677–15682
Lipschutz JH, Foster BA, Cunha GR 1997 Differentiation of rat neonatal ventral prostates grown in a serum-free organ culture system. Prostate 32:35–42
Jarred RA, Keikha M, Dowling C, McPherson SJ, Clare AM, Husband AJ, Pedersen JS, Frydenberg M, Risbridger GP 2002 Induction of apoptosis in low to moderate-grade human prostate carcinoma by red clover-derived dietary isoflavones. Cancer Epidemiol Biomarkers Prev 11:1689–1696
Imamov O, Lopatkin NA, Gustafsson JA 2004 Estrogen receptor in prostate cancer. N Engl J Med 351:2773–2774
Prins GS, Marmer M, Woodham C, Chang W, Kuiper G, Gustafsson JA, Birch L 1998 Estrogen receptor- messenger ribonucleic acid ontogeny in the prostate of normal and neonatally estrogenized rats. Endocrinology 139:874–883
Kyprianou N, Isaacs JT 1988 Activation of programmed cell death in the rat ventral prostate after castration. Endocrinology 122:552–562
Staack A, Kassis AP, Olshen A, Wang Y, Wu D, Carroll PR, Grossfeld GD, Cunha GR, Hayward SW 2003 Quantitation of apoptotic activity following castration in human prostatic tissue in vivo. Prostate 54:212–219
Kurita T, Wang YZ, Donjacour AA, Zhao C, Lydon JP, O’Malley BW, Isaacs JT, Dahiya R, Cunha GR 2001 Paracrine regulation of apoptosis by steroid hormones in the male and female reproductive system. Cell Death Differ 8:192–200
Prins GS 1992 Neonatal estrogen exposure induces lobe-specific alterations in adult rat prostate androgen receptor expression. Endocrinology 130:3703–3714
Woodham C, Birch L, Prins GS 2003 Neonatal estrogen down-regulates prostatic androgen receptor through a proteosome-mediated protein degradation pathway. Endocrinology 144:4841–4850
Cardone A, Angelini F, Varriale B 1998 Autoregulation of estrogen and androgen receptor mRNAs and downregulation of androgen receptor mRNA by estrogen in primary cultures of lizard testis cells. Gen Comp Endocrinol 110:227–236
Esposito T, Astore E, Cardone A, Angelini F, Varriale B 2002 Regulation of androgen receptor mRNA expression in primary culture of Harderian gland cells: cross-talk between steroid hormones. Comp Biochem Physiol B Biochem Mol Biol 132:97–105
Varriale B, Esposito T 2005 The hamster androgen receptor promoter: a molecular analysis. J Steroid Biochem Mol Biol 94:103–110
Hall JM, Couse JF, Korach KS 2001 The multifaceted mechanisms of estradiol and estrogen receptor signaling. J Biol Chem 276:36869–36872
Levin ER 1999 Cellular functions of the plasma membrane estrogen receptor. Trends Endocrinol Metab 10:374–377
Hewitt SC, Deroo BJ, Korach KS 2005 Signal transduction. A new mediator for an old hormone Science 307:1572–1573
Luconi M, Forti G, Baldi E 2002 Genomic and nongenomic effects of estrogens: molecular mechanisms of action and clinical implications for male reproduction. J Steroid Biochem Mol Biol 80:369–381
Kelly MJ, Levin ER 2001 Rapid actions of plasma membrane estrogen receptors. Trends Endocrinol Metab 12:152–156
Stauffer SR, Coletta CJ, Tedesco R, Nishiguchi G, Carlson K, Sun J, Katzenellenbogen BS, Katzenellenbogen JA 2000 Pyrazole ligands: structure-affinity/activity relationships and estrogen receptor--selective agonists. J Med Chem 43:4934–4947
Kraichely DM, Sun J, Katzenellenbogen JA, Katzenellenbogen BS 2000 Conformational changes and coactivator recruitment by novel ligands for estrogen receptor- and estrogen receptor-: correlations with biological character and distinct differences among SRC coactivator family members. Endocrinology 141:3534–3545
Meyers MJ, Sun J, Carlson KE, Katzenellenbogen BS, Katzenellenbogen JA 1999 Estrogen receptor subtype-selective ligands: asymmetric synthesis and biological evaluation of cis- and trans-5,11-dialkyl-5,6,11, 12-tetrahydrochrysenes. J Med Chem 42:2456–2468
Frasor J, Barnett DH, Danes JM, Hess R, Parlow AF, Katzenellenbogen BS 2003 Response-specific and ligand dose-dependent modulation of estrogen receptor (ER) activity by ER in the uterus. Endocrinology 144:3159–3166
Tena-Sempere M, Navarro VM, Mayen A, Bellido C, Sanchez-Criado JE 2004 Regulation of estrogen receptor (ER) isoform messenger RNA expression by different ER ligands in female rat pituitary. Biol Reprod 70:671–678
Sanchez-Criado JE, Martin De Las Mulas J, Bellido C, Tena-Sempere M, Aguilar R, Blanco A 2004 Biological role of pituitary estrogen receptors ER and ER on progesterone receptor expression and action and on gonadotropin and prolactin secretion in the rat. Neuroendocrinology 79:247–258
Voigt KD, Bartsch W 1986 Intratissular androgens in benign prostatic hyperplasia and prostatic cancer. J Steroid Biochem 25:749–757
Nonneman DJ, Ganjam VK, Welshons WV, vom Saal FS 1992 Intrauterine position effects on steroid metabolism and steroid receptors of reproductive organs in male mice. Biol Reprod 47:723–729
vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA, Nagel SC, Dhar MD, Ganjam VK, Parmigiani S, Welshons WV 1997 Prostate enlargement in mice due to fetal exposure to low doses of estradiol or diethylstilbestrol and opposite effects at high doses. Proc Natl Acad Sci USA 94:2056–2061
Timms BG, Petersen SL, vom Saal FS 1999 Prostate gland growth during development is stimulated in both male and female rat fetuses by intrauterine proximity to female fetuses. J Urol 161:1694–1701
Cook JC, Johnson L, O’Connor JC, Biegel LB, Krams CH, Frame SR, Hurtt ME 1998 Effects of dietary 17-estradiol exposure on serum hormone concentrations and testicular parameters in male Crl:CD BR rats. Toxicol Sci 44:155–168
Timms BG, Howdeshell KL, Barton L, Bradley S, Richter CA, Vom Saal FS 2005 Estrogenic chemicals in plastic and oral contraceptives disrupt development of the fetal mouse prostate and urethra. Proc Natl Acad Sci USA 102:7014–7019
Prins GS 1997 Developmental estrogenization of the prostate gland. In: Naz R, ed. Prostate basic and clinical aspects. New York: CRC Press; 245–263
Siddiqui K, Abbas F, Biyabani SR, Ather MH, Talati J 2004 Role of estrogens in the secondary hormonal manipulation of hormone refractory prostate cancer. J Pak Med Assoc 54:445–447
LaVallee TM, Zhan XH, Herbstritt CJ, Kough EC, Green SJ, Pribluda VS 2002 2-Methoxyestradiol inhibits proliferation and induces apoptosis independently of estrogen receptors and . Cancer Res 62:3691–3697(R. A Taylor, P. Cowin, J. F. Couse, K. S)
Department of Environmental and Molecular Toxicology (J.F.C.), North Carolina State University, Raleigh, North Carolina 27695
Receptor Biology Section (J.F.C., K.S.K.), Laboratory of Reproductive and Developmental Toxicology, National Institute of Environmental Health Sciences/National Institutes of Health, Research Triangle Park, North Carolina 27709
Abstract
Estrogens induce both proliferative and antiproliferative responses in the prostate gland. To date, antiproliferative effects of estrogens are generally considered to be due to systemic antiandrogenic actions. However, estrogen action mediated through estrogen receptor (ER) was recently suggested as another mechanism of induction of apoptosis in the prostate. This study aimed to explore the hypothesis that the antiproliferative effects of estrogen are directly mediated through ER using a prostate organ culture system. We previously reported effects of 17-estradiol (E2) using rat ventral prostate (VP) tissues, and adapted the system for culturing mouse tissues. In both rat and mouse models, estrogen-induced apoptosis was detected that was spatially and regionally localized to the epithelium of the distal tips. Using organ cultures of ER knockout (ERKO) and ERKO prostates, we failed to demonstrate that apoptosis induced by E2 was mediated through either receptor subtype. Activation of ER-selective ligands (ER, propyl pyrazole triol, ER, diaryl-proprionitrile, and 5-androstane-3,17-diol) in organ culture experiments failed to induce apoptosis, as did the membrane impermeable conjugate E2:BSA, discounting the possibility of nongenomic effects. Consequently, E2 regulation of androgen receptor (AR) expression was examined and, in the presence of nanomolar testosterone levels, E2 caused a specific reduction in AR protein expression in wild-type, ERKO, and ERKO mice, particularly in the distal region where apoptosis was detected. This down-regulation of AR protein provides a possible mechanism for the proapoptotic action of E2 that is independent of ERs or nongenomic effects.
Introduction
HIGH DOSES OF estrogens inhibit prostate growth both in vivo and in vitro. However, in vivo experiments are complicated by the fact that estrogens exert indirect effects on the prostate, by negative feedback on the hypothalamic-pituitary gonadal axis, thereby suppressing androgen production. Any reduction in androgen levels, androgen receptor (AR) expression, or androgen signaling will result in atrophy, decreased secretory activity, and apoptosis in the epithelium (1). In addition to indirect effects of estrogen in modulating androgen production, estrogens also have direct effects on the prostate itself. Evidence that estrogens act directly upon the prostate is implied because the estrogen receptors (ERs) and reside within the gland. The expression of the ERs has been described in the developing and adult prostate and they are differentially localized; ER is predominantly found in the stroma (2, 3), although it can be detected in the epithelium after estrogen exposure (4), and ER is predominantly found in the epithelium (3, 5, 6).
Estrogens induce both proliferative and antiproliferative effects in the prostate. Several studies have identified that ER is responsible for mediating proliferative actions by estrogens on the prostate during development and adulthood (4, 5). A previous study by Prins et al. (5) used ER knockout (ERKO) and ERKO mice to demonstrate that ER was the dominant ER form mediating developmental estrogenization of the prostate gland. Squamous metaplasia is another biological response induced by exogenous estrogen exposure that is mediated via ER as demonstrated by tissue recombination studies (4).
Although ER is the predominant epithelial ER residing in the prostate, a biological endpoint mediated through this receptor remains to be defined. It was postulated that direct antiproliferative effects of estrogens in the prostate are mediated through ER (7, 8, 9, 10). This hypothesis is based on several lines of correlative evidence. Firstly, mice deficient in ER (ERKO) were reported to demonstrate foci of epithelial prostate hyperplasia upon aging that was associated with higher proliferative and lower apoptotic indexes (8, 9, 10), although this phenotype was not confirmed by other laboratories (5, 11, 12). Secondly, administration of antiestrogens (tamoxifen and ICI 182,780) to the human prostate cancer cell line, DU145, which express only ER, and not ER, caused growth inhibition (13) and cotreatment of DU145 cells with ER-antisense oligonucleotide reversed the antiproliferative effects induce by ICI 182,780 (13), suggesting that ER mediated growth-inhibitory actions of antiestrogens in tumor cells. Thirdly, ER binds to electrophile/antioxidant response elements that stimulate induction of genes such as quinone reductase and gluthanthione S-transferase that are protective against carcinogenesis (14). Although these associated data infer antiproliferative actions mediated by ER, direct evidence is not available.
The problem with demonstrating that ER induces antiproliferative or proapoptotic responses relates to the exquisite sensitivity of the prostate to changes in androgens, or androgen signaling, associated with estrogen administration in vivo. Therefore, to circumvent the centrally mediated effects of estrogens in decreasing serum androgen levels, we have developed an organ culture system where prostate lobes are grown in a defined hormonal environment, so that androgen levels are maintained, and direct effects of estrogens can be examined. We previously used this system using rat ventral prostate (VP) lobes (15). Here we report for the first time that this organ culture model can be adapted to allow maintenance of mouse prostate lobes in vitro, thereby permitting the use of genetically altered mouse models in such experiments. In this study, we have used ER gene-targeted models (ERKO and ERKO mice) to study the specific role of each receptor subtype in mediating antiproliferative effects of estrogen on the developing prostate. In addition, we have adapted a complementary, alternative approach using several ER subtype-selective ligands to delineate ER-mediated events. The aim of this study was to test the hypothesis that ER activation directly mediates apoptotic effects in the developing rodent prostate, independent of regulation of androgen levels.
Materials and Methods
Animals and tissue collection
All animal handling and procedures were carried out in accordance with National Health and Medical Research Council guidelines for the Care and Use of Laboratory Animal Act and according to the Animal Experimentation and Ethics Committee at Monash Medical Centre (Clayton, Australia). Sprague Dawley outbred rats (d 0), (ERKO [C57BL/6J black mice (16)], ERKO [C57BL/6J black mice (17)] and normal C57BL/J6 mice were killed on d 1–2 after birth. VP lobes were microdissected for organ culture.
Organ culture
Organ culture was carried out as previously described in rat (15, 18), although there were some modifications to satisfy mouse organ culture conditions. Pilot studies in C57BL/6J mice demonstrated that mice needed to be at least postnatal d 1 (weight range 1.5–2.0 g) to undergo normal branching morphogenesis (data not shown). Briefly, all animals from each litter (ERKO and ERKO colonies) were cultured blind because the genotypes were unknown at the beginning of the culture. Mice were killed by decapitation and tail snips collected from each of the KO mice for genotyping.
Tissue collection for mouse and rat tissues involved microdissection of VPs (rat lobes collected independently and mouse lobes collected as two lobes together). Each explant was cultured on Millicell CM filters (Millipore Corp., Bedford, MA) floating on 500 μl of nutrient media in a four-well plate at 37 C in a humidified 5% CO2 incubator. A basal medium of DMEM/Ham’s F-12, 1:1 (vol/vol) lacking phenol red, supplemented with insulin (10 μg/ml) and transferrin (10 μg/ml) was used in all experimental groups. The medium was supplemented with 10 nM testosterone (T), and treatment groups were cultured with high doses of 17-estradiol (E2; 15 μM) together with 10 nM T or E2 alone. Other estrogenic compounds were tested including 3Adiol (5-androstane-3,17-diol; putative ER ligand), E2:BSA (Sigma-Aldrich, St. Louis, MO), an ER-selective ligand [1,35-(4-hydroxyphenyl)-4-propyl-1H pyrazole; propyl pyrazole triol; PPT]; Tocris Cookson Ltd., Avonmouth, UK), and an ER-selective ligand [2,3-bis-(4-hydroxyphenyl)-propionitrile; diaryl-propionitrile; DPN; Tocris]. Steroids were tested at a range of doses (10–5 M to 10–13 M) and vehicle controls were conducted. At least four explants of each genotype were subjected to each hormonal treatment.
The organs were harvested after 5 d of culture. Explants were photographed, fixed in Bouins fixative for 1 h at room temperature, and then processed to paraffin for histological and stereological analysis.
Histology
Fixed tissues were processed and embedded in paraffin for histological analysis. Care was taken to ensure tissues were embedded so that longitudinal sections could be obtained. Serial 5-μm sections were cut on a microtome and dried onto Superfrost Plus+ slides (Menzel-Glazer, Braunschweig, Germany) before histological and stereological examination. Hematoxylin and eosin (H&E) staining was performed on every fifth section to examine the pathology throughout the whole tissue.
Detection of apoptosis
Apoptosis was analyzed by ApopTag In Situ Apoptosis Detection Kit (Integren, Purchase, NY). Briefly, sections were dewaxed, incubated in Equilibration Buffer (Intergren), then treated with TdT enzyme in Reaction Buffer (Integren) for 1 h at 37 C. Sections were washed in Stop Wash Buffer (Integren) for 30 min at 37 C, treated with 3% (vol/vol) H2O2 in methanol for 15 min, and blocked with CAS block (Zymed Laboratories Inc., South San Francisco, CA). Apoptotic cells were detected by antidigoxigenin conjugate (Integren) for 30 min at room temperature and color reacted with 3,3'-diaminobenzidine tetrahydrochloride (liquid substrate kit; Zymed). The reactions were stopped with water and sections were counterstained with Mayer’s hematoxylin, dehydrated, cleared and mounted.
Immunohistochemistry
Immunolocalization of AR and ERs (ER) and (ER) was detected using the Dako Autostainer Universal Staining System (Dako A/S, Carpinteria, CA). Immunostaining was conducted on n = 3 animals from each group, including n = 10 sections (i.e. every fifth or sixth section, depending on the tissue size) from each tissue to sample throughout the explant. Immunoreactivity for AR, ER, and ER was detected after sections were subjected to microwave antigen retrieval in 0.01 M citrate buffer (pH 6.0), boiling for 20 min. Antigen retrieval for caspase-3 immunoreactivity was carried out in 0.01 M sodium citrate buffer (pH 6.0), boiling for 10 min. All sections were then treated with peroxidase blocking reagent (Dako Envision System kit; Dako) for 30 min, and nonspecific binding was blocked using CAS block for at least 30 min (Zymed). Monoclonal antibody AR (N-20) (Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a working concentration of 0.33 μg/ml or concentration-matched rabbit IgG (Dako) for 1 h at room temperature. Monoclonal ER (1D5) (Dako Corp., Carpinteria, CA) and monoclonal antibody ER (NCL-ER-) (Novocastra Laboratories Ltd., UK) at working concentrations of 14 μg/ml and 1 μg/ml, respectively, for 2 h at room temperature. Polyclonal antibody cleaved caspase-3 (Asp175) (Cell Signaling Technology, Inc., Beverly, MA) at a working concentration of 1 μg/ml for 2 h at room temperature. Antibodies were detected by incubation with peroxidase-labeled polymer (Dako Envision System; Dako), which is conjugated to antimouse and antirabbit Igs, for 15 min at room temperature and then color reacted with 3,3'-diaminobenzidine tetrahydrochloride (liquid substrate kit) for 5 min. The reactions were stopped in water, and sections were counterstained with Mayer’s hematoxylin, dehydrated, cleared, and mounted.
Semiquantitation of apoptosis
The incidence of apoptosis was estimated based on a method that allowed an unbiased semiquantitation of the percentage of apoptotic cells in both treated and control samples (19). Tissue sections were mapped at x40 magnification to define tissue boundaries. Random fields were systematically selected by computer-assisted software: CAST version 1.10 software (Computer Assisted Stereological Toolbox) (Olympus Danmark A/S, Albertslund, Denmark), and sampling was conducted using a three-by-two unbiased counting frame. Frame counting was performed on 10 sections uniformly spaced throughout the tissue, n = 3 for each group, with an average of 1500 cells counted per section. Cells were classified as either apoptotic or nonapoptotic-based on Apoptag labeling (only fully stained apoptotic cells that exhibited histological characteristics typical of cells undergoing programmed cell death were counted as positive, thereby avoiding counts of end stage apoptotic bodies) and represented as percentage of total cells. All data were expressed as the mean ± SEM. Control and treatment groups were compared using a two-tailed paired t test, with the significance threshold employed at a level of 5% (P < 0.05). All data analyses were conducted using Prism 2.01 software (GraphPad Software, Inc., San Diego, CA).
Results
Effect of E2 on rat prostate development in vitro
Organ culture of rat VP in the presence of T (10 nM) has been described previously (15), mimicking in vivo development in vitro (Fig. 1A). In addition, we previously demonstrated that culture of neonatal rat VPs in the presence of T plus E2 (15 μM) retards growth and reduces the extent of branching morphogenesis (15) (Fig. 1B). Administration of T plus E2 to neonatal rat VPs in vitro altered stromal cell differentiation, particularly smooth muscle development, as well as perturbed epithelial differentiation (15). Figure 1, C and D, shows that, in explants of rat VP, reduced growth is associated with an increase in the incidence of apoptosis in organs cultured in the presence of T plus E2, compared with T alone, as detected by ApopTag labeling. The incidence of apoptosis was very low in media containing T alone (Fig. 1C), but in the presence of T plus E2, there was a significant increase in the incidence of apoptosis that was predominantly found in the differentiating epithelial cells of the distal region, where branching morphogenesis and epithelial cord formation occurs (Fig. 1D). The incidence of apoptosis was confirmed using a second detection technique, specifically immunolocalization of cleaved caspase-3 (one of the key executioners of apoptotic pathway). Using this approach, a similar incidence of apoptosis was observed after T alone and T plus E2 (data not shown). The incidence of apoptosis was estimated using a semiquantitation technique, which revealed an approximately 8-fold increase in the percentage of cells undergoing apoptosis after T plus E2 (5.25 ± 0.17%) treatment compared with T alone (0.65 ± 0.03%; Table 1).
Establishment of mouse VP organ culture; comparison to in vivo tissues
To perform experiments using genetically altered mice, we adapted the organ culture technique for mouse prostate and report here in vitro development of mouse VPs for the first time. Specifically, we used older mouse tissues (d 1–2 postnatal rather than day of birth), and cultured paired organs connected by shared mesenchyme, rather than isolated lobes. Using these regimes, we were able to maintain the integrity of the glands for up to 5 d of culture and the development of explants cultured in vitro was similar to that observed in age-matched in vivo tissues (Fig. 2, A–J). Tissues were isolated on d 1 after birth, and each VP lobe consisted of two main ducts that had initiated branching (Fig. 2A). A histological section of these tissues revealed undifferentiated epithelial cords invading the surrounding mesenchyme (Fig. 2B). To confirm that the development we observed in vitro mimicked that observed in vivo, we isolated mouse VP lobes from postnatal d 4 (Fig. 2C) and 8 (Fig. 2D). Over the first 6–8 d in vivo, mouse VPs undergo extensive growth and significant branching morphogenesis occurs (Fig. 2, C and D). Histologically, postnatal d 8 VPs exhibit proximal to distal maturation of the ducts with some canalization and maturation in the proximal region, whereas the distal tips remain undifferentiated and continue budding into the surrounding mesenchyme (Fig. 2E). There is little lumen formation at the stage of development. By comparison, mouse VPs that were isolated at d 1–2 (Fig. 2, A and B) were maintained in vitro in the presence of T alone for 5–6 d (equivalent of in vivo d 8) and underwent similar development involving branching morphogenesis and ductal formation (Fig. 2, F and G). The main difference between these in vitro explants and the equivalent in vivo tissues is architecture of the lobes; in vivo tissues maintain three-dimensional structure, whereas the in vitro explants tend to grow in a flattened manner, generating a more two-dimensional shape. Nonetheless, histological analysis of the in vitro explants revealed similar maturation and differentiation to comparative in vivo explants, although more advanced canalization and lumen formation (Fig. 2H). Differentiation and organization occurred in a proximal to distal manner, with the most advanced development observed in the proximal region and newly forming ductal structures in the distal region (Fig. 2, I and J). In the proximal region, the epithelium became polarized and secretory epithelial cells were observed in addition to the underlying basal cells (Fig. 2I), whereas the epithelium was actively differentiating and proliferating in the distal tips (Fig. 2J).
Effect of E2 on wild-type (WT) mouse VP growth in vitro
In the presence of 10 nM T, the growth inhibitory effects of E2 were dose responsive over a concentration range 10 nM to 20 μM (Fig. 3, A–F), consistent with previous observations (15). High E2 concentrations (micromolar) significantly reduced prostate growth, whereas lower concentrations (nanomolar) reduced the inhibitory effect such that explants were not significantly different from control explants. Similar to previous studies in rat VPs (15), a concentration of 15 μM was chosen to further investigate the mechanism of estrogen-induced apoptosis in the mouse VP.
Mouse VPs from WT mice were cultured in the presence of T alone (10 nM) and normal growth was observed in vitro, as described above (Fig. 4A). When cultured in the presence of T plus E2 (15 μM), branching morphogenesis was reduced (Fig. 4B) so that there were generally fewer secondary and terminal branches; the main branches were elongated and distended, rather than forming new branches. In some regions, these ducts were dilated and cystic (Fig. 4B). The incidence of apoptosis in mouse VPs after culture in the presence of T alone, and T plus E2 is shown in Fig. 4, C–F. An examination of proximal and distal regions revealed very few apoptotic cells after culture in the presence of T alone (Fig. 4C). In contrast, treatment with T plus E2 induced a significant amount of apoptosis that was specifically in the epithelium of the distal tips (Fig. 4D). These observations in the WT mouse were similar to the responses observed in the rat VP organ cultures (Fig. 1). Interestingly, apoptotic cells were noted adjacent to cells actively undergoing mitosis, indicating that proliferation was concurrent with cell death (Fig. 4, C and D). The incidence of apoptosis was confirmed using immunolocalization of cleaved caspase-3 and using this approach, a similar incidence of apoptosis was observed after T alone (Fig. 4E) and T plus E2 (Fig. 4F). Semiquantitation revealed a significant increase in the percentage of cells undergoing apoptosis in the mouse VPs, at similar levels to that observed in the rat tissues (T alone, 0.54 ± 0.13% vs. T plus E2, 5.10 ± 1.13%; Table 1).
Localization of ERs
After culture in the presence of T alone, regional variation in the expression of ERs was observed. ER immunostaining was restricted to individual cells within the proximal region, specifically in mesenchymal cells (Fig. 5, A and B). In contrast to ER immunoreactivity, ER immunoreactivity was observed in both the proximal and distal regions. Proximally, nuclear ER staining was localized to smooth muscle, stromal and epithelial cell types (Fig. 5C). Comparatively, in the distal region ER immunoreactivity was predominantly localized to the smooth muscle and stromal cells; however, weak basal epithelial cell staining was apparent (Fig. 5D). After 5 d of culture in the presence of T plus E2, immunolocalization of ER was down-regulated so that it was low/undetectable in both proximal and distal regions (Fig. 5E), whereas ER was up-regulated, particularly in the epithelial cells in the distal regions of the gland (Fig. 5F), where the majority of apoptotic cells were observed. There was no apparent difference in stromal cell expression of ER throughout the gland when compared with T-treated controls after E2 treatment. This apparent increase in ER, together with the concurrent expression in the region where cells were undergoing apoptosis, further supported the idea that estrogen-induced apoptosis may be mediated through this receptor. Therefore, this hypothesis was tested in ERKO mouse VPs.
Effect of E2 on ERKO and ERKO mouse VP growth in vitro
After 5 d of culture with T alone, ERKO and ERKO VPs had apparently normal branching morphogenesis and no gross abnormalities were observed (Fig. 6, A and B). In the presence of T alone, ERKO and ERKO mouse VPs showed a very low incidence of apoptosis throughout, after examination of proximal and distal regions (Fig. 6, C and D; and Table 1), similar to WT. After culture in the presence of T plus E2, ERKO and ERKO VPs responded similar to WT mice, with reduced branching and distended duct formation (Fig. 6, E and F). In ERKO mice, growth was more reminiscent of explants cultured with T alone, although some regions still had ducts that were dilated and cystic (Fig. 6E). However, in the presence of T plus E2, both ERKO and ERKO mice exhibited a significant increase in the incidence of apoptosis, particularly in the distal epithelial tips (Fig. 6, G and H), indicating that estrogen-induced apoptosis still occurred in the absence of functional ER or ER. This increase in apoptosis in ERKO and ERKO mouse VPs after T plus E2 treatment was confirmed by immunolocalization of cleaved caspase-3 (data not shown) and semiquantitation (Table 1).
Effect of ER- and ER-selective ligands on mouse prostate development in vitro
To confirm whether estrogen-induced apoptosis was mediated through either ER or ER, we investigated the biological response to ER- and ER-selective ligands on mouse prostate development. WT C57BL/J6 mice were used for this study. To validate the use of the subtype-specific ER ligands, vehicle only controls were conducted and explants were similar to T-treated explants (data not shown). In the presence of T alone, explants developed similar to that observed in vivo, and extensive branching morphogenesis occurred (Fig. 7A). Histologically, the epithelium became polarized and secretory epithelial cells were observed in addition to the underlying basal cells, concurrent with differentiation and organization of the surrounding stromal tissue (Fig. 7B). The incidence of apoptosis was low, consistent with previous observations (Fig. 7C).
When cultured in the presence of T plus an ER-specific ligand (PPT; 10 μM), explants showed a striking reduction in growth (Fig. 7D). The size of the explants was significantly smaller and there was less branching morphogenesis. Concurrent with the reduction in growth of PPT-treated explants, histological examination by H&E staining revealed significant morphological alterations with ductal branching, elongation and canalization interrupted throughout the entire gland (Fig. 7E). The spatial and regional organization was not apparent, and the ductal architecture was severely altered. Similar to T-treated controls, the incidence of apoptosis was also low because very few cells were detected by ApopTag labeling (Fig. 7F) or immunolocalization of cleaved caspase-3 (data not shown).
After 5 d of culture with T plus an ER-specific ligand (DPN; 10 μM), mouse VPs were larger than explants treated with T alone (Fig. 7G). This enlargement appeared to be the result of excessive dilation of the ducts that may be associated with increased production of secretory products. H&E examination revealed several differences in cellular morphology when compared with T-treated control explants (Fig. 7H). In T plus DPN-treated explants, branching morphogenesis appeared reduced with generally fewer secondary and tertiary branches. The main ductal branches appeared elongated and distended with large lumen spaces. Epithelial cell differentiation appeared well developed throughout the entire tissue, with pseudostratified columnar epithelium lining the majority of the ductal length (Fig. 7H). Similarly to T and T plus PPT (ER-specific ligand)-treated explants, the incidence of apoptosis remained very low in T plus DPN (ER-specific ligand)-treated explants (Fig. 7I).
Effect of 3Adiol on rodent prostate development in vitro
The T metabolite 3Adiol is reported to be a specific ligand for ER in the prostate (9, 20). To test whether 3Adiol was able to induce apoptosis in the developing rodent prostate similar to E2, organ culture experiments were conducted on both rat and mouse VPs. To validate the use of 3Adiol, vehicle-only controls were conducted and explants were similar to T-treated explants (data not shown). Results presented represent the rat organ culture explants (Fig. 8, A–D), but these effects were also observed in organ cultures using C57BL/J6 mouse ventral prostate lobes (data not shown). In the presence of T, 3Adiol was tested at several doses (including 20 μM, 10 μM, and 1 μM), similar to the dose range required for E2 to induce apoptosis. At the highest dose (20 μM), 3Adiol caused little change in prostate size and growth (Fig. 8, A and B). The extent and pattern of branching morphogenesis was not altered and development occurred similar to explants cultured with T alone (Fig. 8A). In terms of apoptosis, the incidence of programmed cell death in control explants cultured with T alone was very low, with few cells immunolabeled with the ApopTag detection kit (Fig. 8C). Similarly, the incidence of apoptosis in explants cultured in the presence of T plus 3Adiol (20 μM) was also very low (Fig. 8D).
Effect of E2:BSA on mouse prostate development in vitro
To test whether estrogen-induced apoptosis was the result of nongenomic estrogenic actions, we determined the effect of a membrane impermeable conjugate of E2, namely E2:BSA. Results presented represent the mouse C57BL/J6 organ culture explants (Fig. 8, E–H). In the presence of T plus E2:BSA, branching morphogenesis was similar to T-treated control explants (Fig. 8, E and F). In terms of apoptosis, the incidence of programmed cell death in control explants cultured with T alone was very low, with few cells immunolabeled with the ApopTag detection kit (Fig. 8G). Similarly, the incidence of apoptosis in explants cultured in the presence of T plus E2:BSA was also very low (Fig. 8H).
AR expression
Because the estrogen-induced apoptosis observed in both mouse and rat organ culture experiments was not shown to be mediated through ER or ER by two experimental models (using ERKO mice and ER-selective ligands), we investigated the possibility that estrogen was acting indirectly via modulation of AR expression after treatment with T alone and T plus E2. After culture in the presence of T alone, AR immunolocalization was detected in both the mesenchymal and epithelial cells throughout WT VP explants (Fig. 9, A and B). There was variability in AR immunostaining along the prostatic ducts within the proximal and distal regions. In the distal region, where most proliferation and differentiation occurred, AR immunostaining was strong and nearly all cells were immunopositive for AR (Fig. 9A). In the more differentiated proximal region, where epithelial polarization and lumen formation occurred, there were some cells, particularly in the epithelium that were immunonegative for AR (Fig. 9B). In contrast, culture in the presence of T plus E2 using WT mice resulted in a significant decrease in AR immunolocalization in both the distal (Fig. 9C) and proximal (Fig. 9D) regions, particularly in the epithelial cells. AR expression was still detectable in epithelial and mesenchymal cells; however, the number of cells that were immunopositive and the staining intensity were reduced (Fig. 9C). The decrease in the proximal region was not as significant (Fig. 9D) compared with the proximal region of explants treated with T alone. ERKO and ERKO mouse VPs treated with T alone showed comparable AR expression to WT VPs (Fig. 9, E and G). Similarly, ERKO and ERKO VPs cultured in the presence of T plus E2 also showed significant down-regulation of AR in the distal epithelium as detected by immunolocalization (Fig. 9, F and H).
Discussion
In this study, we have adapted a serum-free prostate organ culture technique to maintain neonatal murine tissues in culture. This system maintains stromal-epithelial interactions within the prostate, as well as allowing prostate growth in defined hormonal environments, where androgen levels can be kept constant. The adaptation of this model to allow organ culture of mouse prostate tissues expands the utility of the system to allow the use of genetically altered mouse models in such studies, in this case, ERKO and ERKO mice.
Here we report that the addition of high-dose E2 resulted in antiproliferative events including active induction of apoptosis in the prostate epithelium, particularly in the distal tip epithelium, where ER was previously localized (21). Data presented here also indicated that ER expression was increased in the distal tip epithelium after E2 treatment. However, using ERKO mice, we failed to demonstrate a specific role for ER or ER in mediating apoptosis because apoptosis was consistently observed after E2 exposure to prostate glands from either ERKO or ERKO mice. Ideally, we would prefer to test the effects of E2 in ERKO tissues, but because each litter only provides 1:16 double KO males, this experiment was not feasible. Instead, we adapted an alternative approach using ER-specific ligands PPT (ER) or DPN (ER) and 3Adiol (ER) and showed these agonists were also unable to induce the apoptotic response. These data implied that apoptosis was caused by a non-receptor-mediated event or was mediated through another indirect estrogen signaling mechanism. We also tested whether the observed apoptosis was due to nongenomic effects of estrogens; however, organ culture experiments using the membrane impermeable conjugate E2:BSA failed to demonstrate an effect.
However, subsequent data revealed a down-regulation of AR, which may provide an alternative explanation for the observed apoptosis in the prostate. AR regulation was independent of androgen levels because T was maintained in the culture media, suggesting there was direct modulation of AR protein expression by E2. Our data suggested that AR down-regulation occurred independently of either ER subtypes because the effect is observed in ERKO and ERKO prostates. We did not quantitate the decrease in AR protein by Western analysis because examination of whole tissue homogenates would mask the changes in cell-specific expression; using immunohistochemistry we observed a marked loss of epithelial AR. The induction of epithelial cell apoptosis by castration (androgen removal) has been well documented (22, 23) and tissue recombination studies eloquently demonstrated that this epithelial response is a paracrine event mediated through the loss of stromal ARs, rather than reduced stimulation of epithelial ARs (24). However, in this study we observed a predominant loss of epithelial AR, rather than stromal AR and so it is not convincing that loss of AR protein is the only factor involved in the apoptotic response.
Nonetheless, the regulation of AR protein expression by estrogen is important to consider. Pharmacological doses of estrogen have been shown in multiple systems, including the one presented here, to decrease AR levels (5, 9, 10, 25) although the mechanism by which this occurs is not well understood. Prins et al. (5) described altered AR protein expression by estrogens that was mediated through ER and partially ER, using an in vivo treatment regime. Further studies by the same group have indicated that estrogen alters AR expression posttranscriptionally (via protein degradation), whereas AR mRNA levels may be unaltered (26). Another suggestion by Weihua et al. (9) was that the AR protein expression is regulated by ER, although a definitive mechanism has not been described. Experiments using primary cultures of lizard testis cells (27) and Harderian gland cells (28) have demonstrated that estrogen also down-regulates AR mRNA expression. A recent report revealed a putative estrogen response element in the Hamster AR promoter mRNA (5'-untranslated region), although this consensus does not appear to be conserved in human or rodent AR mRNA (29). Clearly, the link between AR expression and estrogen exposure, at either the protein or mRNA level, requires further investigation. Nevertheless, the down-regulation of AR protein, and therefore androgen function by estrogen, may in part be responsible for the apoptotic response observed here. Interestingly, previous reports using ERKO mice have described elevated expression of AR in the prostate (9, 10), but we did not observe this finding in the current study.
Estrogen acts by multiple signaling pathways in addition to the traditional steroid hormone receptor pathway including biological responses that are 1) ligand independent, where other factors such as growth factors stimulate ERE in the absence of ligand, or 2) ERE-independent actions, where the ligand/receptor complex targets genes by binding to other DNA-bound transcription factors forming a complex that alters downstream gene regulation independent of EREs (30). Nongenomic effects also occur, where estradiol activates putative membrane-associated binding sites, linked to intracellular signal transduction pathways to generate rapid tissue responses (30, 31, 32). In contrast to the genomic effects exerted by estrogens and ERs, nongenomic effects occur very rapidly, even seconds to minutes after estrogen exposure, to induce activation of intracellular second messengers such as calcium, nitric oxide formation, activation of kinases such as tyrosine kinases, protein kinase A, protein kinase C, ERK, and protein kinase B (33). Such effects have been described in neuronal, vascular, and bone systems (34). No evidence for nongenomic effects of estrogen in the induction of apoptosis in the prostate could be detected in this study using the membrane impermeable conjugate E2:BSA.
As a complementary, alternative approach to the use of ERKO mice to study ER-mediated events, we used the recently developed, novel ER subtype-selective ligands. The compound PPT is a potent ER agonist that does not activate ER. This results from the fact that it binds with high affinity and 400-fold preference to ER, and demonstrates almost no binding for ER (35, 36). In contrast, the compound DPN is a potency-selective agonist for ER with a more than 70-fold higher binding affinity for ER than ER (37). These compounds have been used in vivo to examine ER-mediated responses in the mouse uterus (38) and pituitary (39, 40). In the current in vitro study, the ER subtype-selective ligands failed to induce apoptosis in the developing mouse prostate; however, differential biological responses on branching morphogenesis and cellular differentiation were observed; ER agonist reduced prostate growth, disrupted the spatial organization of the gland, and altered cell differentiation in both epithelial and stromal compartments, whereas ER appeared to cause dilation of the ducts as a result of enhance cellular differentiation and increased secretory activity. Future studies using these selective ligands will further our understanding of ER-mediated events in the prostate other than apoptosis.
The putative ER-specific ligand is 3Adiol was hypothesized to be a preferred ligand for ER in vivo (9, 10). 3Adiol is a 5-dihydrotestosterone metabolite that competes with E2 for binding to ER and elicits estrogenic responses (10, 41). In this study, we tested the effects of 3Adiol on the developing rodent prostate in vitro and failed to induce apoptosis. This result indicated that even if 3Adiol is the preferred ligand for ER, it was unable to elicit an apoptotic response in the prostate gland.
Despite conflicting data fetal exposure to low doses of estrogens has been shown to induce permanent alterations in the prostate gland, including increased adult prostate weight (42, 43, 44). Neonatal exposure to high pharmacological doses of estrogens has also been shown to elicit effects on the gland, including permanent suppression of prostate growth and induction of epithelial hyperplasia in adulthood (25). Studies examining the effects of high doses of estrogens on the prostate have predominantly been carried out in vivo, where the direct effects of estrogens are confounded by the down-regulation of the hypothalamic-pituitary-gonadal axis after exogenous estrogen administration (1, 45). In these in vitro organ culture experiments, we used micromolar concentrations of E2, not unlike the pharmacological doses of estrogens given to rodents in developmental toxicology studies (46, 47) and men being treated for prostate cancer in the clinic (48). A previous report using MCF-7 breast cancer cells showed that, whereas lower concentrations stimulated cell proliferation, high concentrations (>10 μM) of E2 inhibited proliferation and induced apoptosis by non-ER-mediated actions (49), similar to that reported here. The mechanism by which high concentrations induce biological changes independent of ERs remains to be determined.
In summary, we demonstrated a direct effect of E2 on the prostate that includes the induction of apoptosis in rat and mouse organ cultures that was estrogen dependent, but not mediated through ER or ER, as determined using ERKO mice and ER-selective ligands. Although androgen levels were maintained in the culture media, AR protein expression was down-regulated, and may provide a possible explanation for the estrogen-induced apoptosis observed. Overall, this study failed to support the hypothesis that ER directly mediates apoptosis in the developing prostate, but raises the possibility of antiproliferative actions in the prostate induced by high concentrations of estrogens through non-receptor-mediated pathways.
Acknowledgments
We thank Hong Wang and Ann Davies (Monash Institute of Medical Research, Monash University) for technical assistance.
Footnotes
First Published Online October 13, 2005
Abbreviations: 3Adiol, -Androstane-3,17-diol; AR, androgen receptor; DPN, diaryl-proprionitrile; E2, 17-estradiol; ER, estrogen receptor; ERKO, ER knockout; ERKO, mice deficient in ER; ERKO, mice deficient in ER; H&E, hematoxylin and eosin; PPT, propyl pyrazole triol; T, testosterone; VP, ventral prostate; WT, wild type.
Accepted for publication October 3, 2005.
References
Wright AS, Thomas LN, Douglas RC, Lazier CB, Rittmaster RS 1996 Relative potency of testosterone and dihydrotestosterone in preventing atrophy and apoptosis in the prostate of the castrated rat. J Clin Invest 98:2558–2563
Prins GS, Birch L 1997 Neonatal estrogen exposure up-regulates estrogen receptor expression in the developing and adult rat prostate lobes. Endocrinology 138:1801–1809
Omoto Y, Imamov O, Warner M, Gustafsson JA 2005 Estrogen receptor and imprinting of the neonatal mouse ventral prostate by estrogen. Proc Natl Acad Sci USA 102:1484–1489
Risbridger G, Wang H, Young P, Kurita T, Wang YZ, Lubahn D, Gustafsson JA, Cunha G, Wong YZ 2001 Evidence that epithelial and mesenchymal estrogen receptor- mediates effects of estrogen on prostatic epithelium. Dev Biol 229:432–442
Prins GS, Birch L, Couse JF, Choi I, Katzenellenbogen B, Korach KS 2001 Estrogen imprinting of the developing prostate gland is mediated through stromal estrogen receptor : studies with ERKO and ERKO mice. Cancer Res 61:6089–6097
Adams JY, Leav I, Lau KM, Ho SM, Pflueger SM 2002 Expression of estrogen receptor in the fetal, neonatal, and prepubertal human prostate. Prostate 52:69–81
Chang WY, Prins GS 1999 Estrogen receptor-: implications for the prostate gland. Prostate 40:115–124
Weihua Z, Warner M, Gustafsson J 2002 Estrogen receptor in the prostate. Mol Cell Endocrinol 193:1–5
Weihua Z, Makela S, Andersson LC, Salmi S, Saji S, Webster JI, Jensen EV, Nilsson S, Warner M, Gustafsson JA 2001 A role for estrogen receptor in the regulation of growth of the ventral prostate. Proc Natl Acad Sci USA 98:6330–6335
Imamov O, Morani A, Shim GJ, Omoto Y, Thulin-Andersson C, Warner M, Gustafsson JA 2004 Estrogen receptor regulates epithelial cellular differentiation in the mouse ventral prostate. Proc Natl Acad Sci USA 101:9375–9380
Dupont S, Krust A, Gansmuller A, Dierich A, Chambon P, Mark M 2000 Effect of single and compound knockouts of estrogen receptors (ER) and (ER) on mouse reproductive phenotypes. Development 127:4277–4291
Couse JF, Curtis Hewitt S, Korach KS 2000 Receptor null mice reveal contrasting roles for estrogen receptor and in reproductive tissues. J Steroid Biochem Mol Biol 74:287–296
Lau KM, LaSpina M, Long J, Ho SM 2000 Expression of estrogen receptor (ER)- and ER- in normal and malignant prostatic epithelial cells: regulation by methylation and involvement in growth regulation. Cancer Res 60:3175–3182
Montano MM, Jaiswal AK, Katzenellenbogen BS 1998 Transcriptional regulation of the human quinone reductase gene by antiestrogen-liganded estrogen receptor- and estrogen receptor-. J Biol Chem 273:25443–25449
Jarred RA, Cancilla B, Prins GS, Thayer KA, Cunha GR, Risbridger GP 2000 Evidence that estrogens directly alter androgen-regulated prostate development. Endocrinology 141:3471–3477
Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach KS, Smithies O 1993 Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci USA 90:11162–11166
Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korach KS, Gustafsson JA, Smithies O 1998 Generation and reproductive phenotypes of mice lacking estrogen receptor . Proc Natl Acad Sci USA 95:15677–15682
Lipschutz JH, Foster BA, Cunha GR 1997 Differentiation of rat neonatal ventral prostates grown in a serum-free organ culture system. Prostate 32:35–42
Jarred RA, Keikha M, Dowling C, McPherson SJ, Clare AM, Husband AJ, Pedersen JS, Frydenberg M, Risbridger GP 2002 Induction of apoptosis in low to moderate-grade human prostate carcinoma by red clover-derived dietary isoflavones. Cancer Epidemiol Biomarkers Prev 11:1689–1696
Imamov O, Lopatkin NA, Gustafsson JA 2004 Estrogen receptor in prostate cancer. N Engl J Med 351:2773–2774
Prins GS, Marmer M, Woodham C, Chang W, Kuiper G, Gustafsson JA, Birch L 1998 Estrogen receptor- messenger ribonucleic acid ontogeny in the prostate of normal and neonatally estrogenized rats. Endocrinology 139:874–883
Kyprianou N, Isaacs JT 1988 Activation of programmed cell death in the rat ventral prostate after castration. Endocrinology 122:552–562
Staack A, Kassis AP, Olshen A, Wang Y, Wu D, Carroll PR, Grossfeld GD, Cunha GR, Hayward SW 2003 Quantitation of apoptotic activity following castration in human prostatic tissue in vivo. Prostate 54:212–219
Kurita T, Wang YZ, Donjacour AA, Zhao C, Lydon JP, O’Malley BW, Isaacs JT, Dahiya R, Cunha GR 2001 Paracrine regulation of apoptosis by steroid hormones in the male and female reproductive system. Cell Death Differ 8:192–200
Prins GS 1992 Neonatal estrogen exposure induces lobe-specific alterations in adult rat prostate androgen receptor expression. Endocrinology 130:3703–3714
Woodham C, Birch L, Prins GS 2003 Neonatal estrogen down-regulates prostatic androgen receptor through a proteosome-mediated protein degradation pathway. Endocrinology 144:4841–4850
Cardone A, Angelini F, Varriale B 1998 Autoregulation of estrogen and androgen receptor mRNAs and downregulation of androgen receptor mRNA by estrogen in primary cultures of lizard testis cells. Gen Comp Endocrinol 110:227–236
Esposito T, Astore E, Cardone A, Angelini F, Varriale B 2002 Regulation of androgen receptor mRNA expression in primary culture of Harderian gland cells: cross-talk between steroid hormones. Comp Biochem Physiol B Biochem Mol Biol 132:97–105
Varriale B, Esposito T 2005 The hamster androgen receptor promoter: a molecular analysis. J Steroid Biochem Mol Biol 94:103–110
Hall JM, Couse JF, Korach KS 2001 The multifaceted mechanisms of estradiol and estrogen receptor signaling. J Biol Chem 276:36869–36872
Levin ER 1999 Cellular functions of the plasma membrane estrogen receptor. Trends Endocrinol Metab 10:374–377
Hewitt SC, Deroo BJ, Korach KS 2005 Signal transduction. A new mediator for an old hormone Science 307:1572–1573
Luconi M, Forti G, Baldi E 2002 Genomic and nongenomic effects of estrogens: molecular mechanisms of action and clinical implications for male reproduction. J Steroid Biochem Mol Biol 80:369–381
Kelly MJ, Levin ER 2001 Rapid actions of plasma membrane estrogen receptors. Trends Endocrinol Metab 12:152–156
Stauffer SR, Coletta CJ, Tedesco R, Nishiguchi G, Carlson K, Sun J, Katzenellenbogen BS, Katzenellenbogen JA 2000 Pyrazole ligands: structure-affinity/activity relationships and estrogen receptor--selective agonists. J Med Chem 43:4934–4947
Kraichely DM, Sun J, Katzenellenbogen JA, Katzenellenbogen BS 2000 Conformational changes and coactivator recruitment by novel ligands for estrogen receptor- and estrogen receptor-: correlations with biological character and distinct differences among SRC coactivator family members. Endocrinology 141:3534–3545
Meyers MJ, Sun J, Carlson KE, Katzenellenbogen BS, Katzenellenbogen JA 1999 Estrogen receptor subtype-selective ligands: asymmetric synthesis and biological evaluation of cis- and trans-5,11-dialkyl-5,6,11, 12-tetrahydrochrysenes. J Med Chem 42:2456–2468
Frasor J, Barnett DH, Danes JM, Hess R, Parlow AF, Katzenellenbogen BS 2003 Response-specific and ligand dose-dependent modulation of estrogen receptor (ER) activity by ER in the uterus. Endocrinology 144:3159–3166
Tena-Sempere M, Navarro VM, Mayen A, Bellido C, Sanchez-Criado JE 2004 Regulation of estrogen receptor (ER) isoform messenger RNA expression by different ER ligands in female rat pituitary. Biol Reprod 70:671–678
Sanchez-Criado JE, Martin De Las Mulas J, Bellido C, Tena-Sempere M, Aguilar R, Blanco A 2004 Biological role of pituitary estrogen receptors ER and ER on progesterone receptor expression and action and on gonadotropin and prolactin secretion in the rat. Neuroendocrinology 79:247–258
Voigt KD, Bartsch W 1986 Intratissular androgens in benign prostatic hyperplasia and prostatic cancer. J Steroid Biochem 25:749–757
Nonneman DJ, Ganjam VK, Welshons WV, vom Saal FS 1992 Intrauterine position effects on steroid metabolism and steroid receptors of reproductive organs in male mice. Biol Reprod 47:723–729
vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA, Nagel SC, Dhar MD, Ganjam VK, Parmigiani S, Welshons WV 1997 Prostate enlargement in mice due to fetal exposure to low doses of estradiol or diethylstilbestrol and opposite effects at high doses. Proc Natl Acad Sci USA 94:2056–2061
Timms BG, Petersen SL, vom Saal FS 1999 Prostate gland growth during development is stimulated in both male and female rat fetuses by intrauterine proximity to female fetuses. J Urol 161:1694–1701
Cook JC, Johnson L, O’Connor JC, Biegel LB, Krams CH, Frame SR, Hurtt ME 1998 Effects of dietary 17-estradiol exposure on serum hormone concentrations and testicular parameters in male Crl:CD BR rats. Toxicol Sci 44:155–168
Timms BG, Howdeshell KL, Barton L, Bradley S, Richter CA, Vom Saal FS 2005 Estrogenic chemicals in plastic and oral contraceptives disrupt development of the fetal mouse prostate and urethra. Proc Natl Acad Sci USA 102:7014–7019
Prins GS 1997 Developmental estrogenization of the prostate gland. In: Naz R, ed. Prostate basic and clinical aspects. New York: CRC Press; 245–263
Siddiqui K, Abbas F, Biyabani SR, Ather MH, Talati J 2004 Role of estrogens in the secondary hormonal manipulation of hormone refractory prostate cancer. J Pak Med Assoc 54:445–447
LaVallee TM, Zhan XH, Herbstritt CJ, Kough EC, Green SJ, Pribluda VS 2002 2-Methoxyestradiol inhibits proliferation and induces apoptosis independently of estrogen receptors and . Cancer Res 62:3691–3697(R. A Taylor, P. Cowin, J. F. Couse, K. S)