| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
1Department of Physiology/Endocrinology, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at Göteborg University; 2Reproductive Medicine, Department of Obstetrics and Gynecology, Institute of Clinical Sciences, Sahlgrenska Academy, Sahlgrenska University Hospital; 3Swegene Centre for Cellular Imaging, Göteborg University, Gothenburg, Sweden; and 4Department of Physiology, Institute of Biomedicine, University of Helsinki, Helsinki, Finland
Submitted 17 July 2006 ; accepted in final form 11 October 2006
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
-estradiol and testosterone levels in vivo. Immunohistochemical analysis of fallopian tube epithelial cells revealed that nuclear localization of AR increased in parallel with decreased AR in the cytoplasm following eCG treatment. Moreover, we found that treatment with flutamide upregulated AR expression in immature mice in association with a decrease in serum testosterone levels, whereas the same treatment resulted in downregulation of AR expression in gonadotropin-stimulated mice with concomitant decreases in serum 17-estradiol concentrations, suggesting that androgen differs from estrogen in the regulation of AR expression. Furthermore, we demonstrated that DES increased both AR protein expression and nuclear location over a 48-h time course. DHT had rapid effects, with induction of AR expression and translocation at 6 h after injection, but unlike DES it had prolonged efficacy. In addition, we provided direct in vivo evidence that nuclear protein interaction between AR and p21Cip1, a previously reported AR-regulated gene, was enhanced by gonadotropin stimulation. To our knowledge, this study provides the first demonstration to illustrate that estrogen as a principal regulator may contribute to regulate and activate AR in the fallopian tubes in vivo. androgen receptor; gonadotropins
The mammalian fallopian tubes play a central role in gamete maturation, fertilization, and early embryo development (13). It is widely accepted that, during the reproductive cycle, tubal development is under the control of steroid hormones (estrogen and progesterone) (27), which are synthesized by the granulosa cells of the ovary (11) and delivered directly from the ovary down the lumen of the fallopian tubes (13). Since immunoreactive AR is found in human (71), mouse (42), and rat (41, 42, 44) fallopian tubes, it is reasonable to assume that this tissue may represent a target for androgen action. AR, a member of the nuclear receptor superfamily, is a ligand-dependent transcription factor that elicits highly selective and tissue-specific effects (5, 19, 26). Similar to other steroid receptors (66), the AR protein contains domains for ligand binding, nuclear location, transcriptional activation, and DNA binding that are required for interaction with specific gene sequences (19). The binding of androgen induces a conformational change within AR, promoting receptor dimerization, phosphorylation, and nuclear translocation of cytoplasmic AR (4, 20). In the nucleus, AR acts as a transcription factor, driving the expression of genes that contain androgen response elements (ARE). This leads to recruitment of AR coregulators and subsequently to modulation of the transcription of target genes (26, 67) such as p21Cip1 (24, 31), ultimately changing the phenotype of the target cell. In contrast to the physiological roles and regulation of AR in the ovary and uterus (24, 58), little is known about the hormonal regulatory profile and sublocalization of the AR protein in mouse fallopian tubes. Such knowledge is of importance for understanding the reproductive effects of androgen, and it may prove to be useful in endocrine therapy, especially in treatment with antiandrogens (generally AR antagonists) such as flutamide or cyrpoterone acetate in hyperandrogenic women (15).
In the present study, we address these issues by investigating the effects of gonadotropins, AR antagonist, estrogen [diethylstilbestrol (DES)], androgen [dihydrotestosterone (DHT)], and progesterone (P4) on the regulation of AR protein expression and its localization in mouse fallopian tubes in vivo. Hormonal regulation of AR mRNA and protein has been extensively studied in rodent ovary, including the granulosa cells (61–63, 65). For this reason, we have included studies done in parallel on ovarian granulosa cells as a control for evaluating the gonadotropin effects under our experimental conditions. Furthermore, we examine whether nuclear AR physically interacts with its target gene, p21Cip1, under gonadotropin stimulation using immunoprecipitation followed by Western blot analysis. Our findings suggest that increases in p21Cip1 protein expression are at least in part the result of a direct interaction of the AR in mouse fallopian tubes. These data provide significant information regarding the regulation and cellular function of AR protein in female reproduction.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Equine chorionic gonadotropin (eCG), DES (category no. D-4628), DHT (D-5027), P4 (P-0130), the AR antagonist flutamide (F-9397) (59), monoclonal anti--actin (A-5441), and alkaline phosphatase-conjugated goat-anti-mouse immunoglobulin (A-1682) were purchased from Sigma Chemical (St. Louis, MO); human chorionic gonadotropin (hCG) was obtained from Organon (Oss, The Netherlands); and normal goat serum (x-0902) was purchased from DAKO (Carpinteria, CA). The antibodies used to detect AR in this study were raised against human ARs and recognized the ligand-binding domain (AR sc-816) or the DNA-binding domain (AR sc-815) of AR in rodent tissues by Western blotting analysis. Rabbit polyclonal anti-ARs (sc-815 and sc-816) and their respective blocking peptides (sc-815p and sc-816p) mouse monoclonal anti-p21Cip1 (sc-6246), rabbit polyclonal anti-p27Kip1 (sc-528), and rabbit polyclonal anti-lamin B1 (sc-20682) were purchased from Santa Cruz Biotechnology, (Santa Cruz, CA); rabbit polyclonal anti-paxillin (2545) was purchased from Cell Signaling Technology, (Danvers, MA); alkaline phosphatase conjugated goat-anti-rabbit immunoglobin (AC31RL) was purchased from Tropix (Bedford, MA); and Pansorbin cell (507858) was purchased from Calbiochem (San Diego, CA). Other reagents not mentioned in the text were purchased from Sigma or Merck (Darmstadt, Germany) and were of the highest purity grade available.
Animal Studies
All experimental procedures and protocols used in the present study were approved by the local ethics committee, Göteborg University, Gothenburg, Sweden. Immature female (21-day-old) and adult male (10-wk-old) C57BL/6 mice were obtained from Taconic M&B (Copenhagen, Denmark). The animals were kept under a 12:12-h light-dark schedule at 21 ± 2°C with ad libitum access to chow and water. Animals were allowed to acclimate to the animal facilities for 
5 days before initiation of treatment and experiment. Female mice at 26 days of age with a body weight ranging from 13 to 15 g were used in this study.
Hormone Treatment and Tissue Preparation
eCG is a glycoprotein with substantial follicle-stimulating hormone bioactivity (10), and hCG is closely related to luteinizing hormone with slower metabolic clearance (51). Treatment of immature female mice with eCG results in follicular steroid secretion (17-estradiol in granulosa cells and androgen in theca cells), whereas subsequent treatment with hCG mimics the stimulatory effect of luteinizing hormone to induce P4 production during luteinization of follicular cells (52, 55). To examine the effect of gonadotropins on AR protein expression, 26-day-old mice were randomly divided into two groups. The eCG group (n = 5–10/time point) received 5 IU eCG (ip) in 100 µl of saline. The eCG/hCG group (n = 5–10/time point) was given 5 IU eCG 48 h before ip injection of 5 IU hCG in 100 µl of saline. Mice were killed by decapitation at time points 6, 24, and 48 h after eCG injection (eCG6–48 h) and at 3, 6, 12, 24, and 48 h after hCG injection (hCG3–48 h). Both fallopian tubes and ovaries were dissected out and then trimmed free of the broad ligament and fat. Fallopian tubes at each time point were either instantly frozen in liquid nitrogen and stored at –135°C for AR protein analysis or fixed in 4% formaldehyde neutral buffered solution (Sigma) for 24 h, transferred to 70% ethanol at 4°C, and then paraffin embedded for immunohistochemical studies, whereas the ovaries of the same animal were used to isolate granulosa cells and were kept at –135°C until analysis. Ovarian hormone determinations were performed in grouped, nonstressed mice treated with gonadotropins in another experiment.
To study the effect of DES, DHT, or P4 on AR protein expression, 26-day-old mice were randomly treated with a single ip (100 µl) injection of DES (20 µg/kg body wt in oil), DHT (1 mg/kg body wt in oil), or P4 (2 mg/kg body wt in oil). This dose was chosen on the basis of earlier studies (21, 54) and our initial results. Animals were killed by decapitation at time points 6, 24, and 48 h after injection. In the experiments, the nonsteroid synthetic estrogen DES was specifically selected because it has a longer half-life and is not bound by 
-fetoprotein (50). DHT is regarded as a more potent androgen than testosterone because it has a higher affinity for AR compared with testosterone in vivo (34) and DHT-activated AR has a relatively long half life (47). In addition, DHT, unlike testosterone, is not aromatized to estrogen (14) and has been shown to inhibit aromatase activity in vitro (22), suggesting that it may not act through the estrogen receptor (ER). Finally, one group of mice received a single intraperitoneal injection of sesame oil and served as a control group. Fallopian tubes at each time point were stored at –135°C for AR protein analysis or paraffin embedded for immunohistochemical studies, as described in Hormone Treatment and Tissue Preparation. To evaluate the effects of DES, DHT, and P4 under our experimental conditions, it was noted that treatment with DES or DHT increased the wet uterine mass, although by different magnitudes, whereas P4 produced no change in uterine weight in our experiments (data not shown).
Treatment With AR Antagonist
In 
Fig. 3A, a schematic time course of the in vivo experiments in mice treated with or without flutamide is shown. Flutamide displays pure anti-androgenic activity without exerting agonistic or any other hormonal activity and exerts its effect through inhibition of the transactivation function of AR (28, 59). We evaluated data from previously published articles (8, 73) to establish an effective concentration for flutamide in vivo (74). Treatment with flutamide had no effect on body weight. Fallopian tubes were collected for analysis of AR as described in Hormone Treatment and Tissue Preparation. To investigate flutamide effects on ovulation in mice, the number of released oocytes was counted 24 h following hCG treatment.
|
|
|
Experiment 2: effect of flutamide on AR protein expression in eCG/hCG-stimulated mice. After 48-h eCG treatment, mice received either a single ip injection of 2 mg flutamide in 100 µl of sesame oil, 5 IU hCG, 1 mg flutamide-5 IU hCG, or vehicle (100 µl sesame oil) only. The animals were killed 24 h after treatment. In addition, 20 mice treated with flutamide/hCG were used for ovulation analysis.
Preparation of Whole Cell, Cytoplasmic, and Nuclear Protein Extracts
The preparation of tissues and whole cell protein lysates was performed as described previously (54). Subcellular fractionation was in principle performed as in Refs. 53 and 56. Protein concentrations were determined by BCA protein assay (Pierce, Rockford, IL) with BSA as standard.
Western Blot Analysis
Expression of AR, p21Cip1, and p27Kip1 proteins was detected by Western blot analysis using standard procedure (54). Protein aliquots were pretreated with 4x SDS (1x = 50 mM Tris·HCl, 2% SDS, 10% glycerol, 10% -mercaptoethanol, and 0.001% bromophenol blue) before loading and separated on 10 or 4–12% SDS-polyacrylamide gels (Novex) with a Bis-Tris-MOPS buffer system under reducing conditions. The separated samples were electrophoretically transferred to polyvinyldifluoride membranes (Amersham International, Buckinghamshire, UK). The membranes were incubated with primary antibody (AR, 1:250; p21Cip1, 1:250; p27Kip1, 1:500) in blocking buffer overnight at 4°C. The following day, the membranes were incubated with either alkaline phosphatase-conjugated goat-anti-rabbit antibody (at 1:40,000) or alkaline phosphatase conjugated goat-anti-mouse antibody (at 1:80,000) and detected using CDP-Star substrate for alkaline phosphatase (Tropix). Immunoblotted signals were exposed and developed using ECL film (Amersham International) and subsequently scanned into a computer. Individual bands were quantified directly from membranes by densitometry using the ImageQuant (version 5.0) software program (Molecular Dynamics, Sunnyvale, CA). Signal intensities of the mouse AR protein were normalized to those of either mouse -actin protein or the gels stained with Coomassie blue as ratios to produce arbitrary densitometric units (ADU) of relative abundance. Care was taken to ensure that the ADU of all the bands considered was in the range of linearity previously assessed. To determine specificity of the AR antibody, Western blot analysis of whole cell extracts prepared from ovary, granulosa cells, fallopian tube, and testis showed a single, 
110-kDa protein band with AR antibody alone (Fig. 1A, left) or with AR antibody neutralized by AR synthetic peptide (Fig. 1A, right), demonstrating specifics of the antibody for the AR of appropriate molecular weight.
Coimmunoprecipitation Studies
Immunoprecipitation was performed as described before (56). Nuclear extracts (250 µg) were immunoprecipitated with 5 µg of AR antibody at 4°C overnight. The immunocomplexes were precipitated with 50 µl of Pansorbin cells for 4 h at room temperature. The bound proteins were then given three sequential washes with 1 ml of RIPA buffer including 10 mM iodoacetamide to prevent nonspecific disulfide linkages, twice with 1 ml of PBS, and processed for Western blot for detection of p21Cip1 protein expression as described in Western Blot Analysis.
Immunohistochemistry
Immunohistochemical analysis was performed as described previously (54, 55), with minor modification. Paraffin sections were deparaffinized and rehydrated through a graded alcohol series followed by antigen retrieval in 10 mM sodium citrate buffer (pH 6.0, microwave 10 min at full power). Slides were blocked with 5% normal goat serum in Tris-buffered saline (TBS; 50 mM Tris, 0.9% NaCl, pH 7.5) for 1 h at room temperature. Furthermore, all sections were treated with 3% H2O2 to remove endogenous peroxidase activity. The primary antibody (AR, 1:50; p21Cip1, 1:100) was diluted in TBS containing 5% goat serum and incubated overnight at 4°C in a humidified chamber. After being washed with TBS, sections were stained using the avidin-biotinylated-peroxidase complex detection system (ABC kit; Vector Laboratories, Burlingame, CA) and then visualized with 3,3-diaminobenzidine tetrahydrochloride (0.5 mg/ml in TBS and 0.01% H2O2, pH 7.6). Slides were viewed on a Zeiss Axioskop II microscope and photomicrographed using Easy Image 1 (Bergström Instrument).
Hormone Measurements
Intraovarian and serum steroid hormone measurements were performed within 2 wk of sample collection. Its concentration was measured by radioimmunoassay according to a protocol provided by the manufacturer (PerkinElmer Life and Analytical Sciences, Wallac Oy, Turku, Finland) as described previously (54, 75). The sensitivity of the assay was typically better than 0.3 nmol/l for testosterone, 50 pmol/l for estradiol, and 0.8 nmol/l for P4, and the intra-assay coefficient of variation was 7.9–14.2% for testosterone, 3.8–10% for estradiol, and 3.3–7.3% for P4.
Data Analysis and Statistics
Data are means ± SE of the number of independent experiments indicated in the figure legends. Multiple comparisons between data were performed using one-way ANOVA with correction of P values with the Bonferroni's multiple range test under the Analyse-It program (Analyse-It Software). A value of P < 0.05 was set as the limit of statistical significance.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
Western blot analysis showed that dynamic changes in AR protein expression occurred during gonadotropin treatment in mouse fallopian tubes (Fig. 2A) and ovarian granulose cells (Fig. 2B). eCG treatment of immature mice resulted in a gradual increase in AR protein expression, reaching a maximum level at 24 h, and remained high at 48 h. However, sequential treatment with hCG caused a rapid decrease in AR protein expression in mouse fallopian tubes (Fig. 2A). Similarly, the levels of AR protein in mouse ovarian granulosa cells were significantly increased at 24- and 48-h eCG, remained high until additional 3-h hCG, and returned to basal levels (0 h) by 6-h hCG (Fig. 2B). The profiles of ovarian testosterone, estradiol, and P4 concentrations in mice from the different time points following gonadotropin treatment are presented in Fig. 2C. Testosterone and estradiol values were significantly higher at 24- and 48-h eCG as well as at 6-h hCG. In contrast, P4 concentrations showed the expected periovulatory rise after hCG treatment. Therefore, the fallopian tubes are under dramatic hormonal influence.
Effect of Flutamide on AR Protein Expression in Mouse Fallopian Tubes
To examine the effect of blocking potential AR-mediated regulation of AR protein expression, we assessed the effects of flutamide, a AR antagonist, on the fallopian tubes from mice treated with eCG and/or hCG (Fig. 3A). Western blot analysis showed that flutamide treatment to immature mice induced an increase in AR protein expression (Fig. 3B, left). However, cotreatment with flutamide and eCG decreased relative amounts of AR protein expression in mouse fallopian tubes compared with mice receiving eCG treatment (Fig. 3B, left). Relative amounts of AR protein were not significantly different between mice treated with flutamide alone and flutamide following 24-h hCG in eCG-stimulated mice (Fig. 3B, left). However, although the reduction of AR protein expression in the fallopian tubes of mice treated with hCG was greater than in mice treated with flutamide, amounts of AR protein in mice treated with hCG and flutamide/hCG were not different (Fig. 3B, right). We observed that treatment with flutamide decreased serum testosterone concentrations without affecting serum 17-estradiol levels in immature mice. However, treatment with flutamide in eCG-stimulated mice did not affect serum testosterone concentrations but significantly decreased serum 17-estradiol levels, in agreement with a similar observation showing that hydroxyflutamide treatment results in decreases in serum estradiol levels in eCG-stimulated rats (8, 73). Furthermore, an increase in the serum testosterone concentration was seen when expression of AR protein was decreased in eCG/hCG-stimulated mice (Fig. 3C). Quantitative assessment of ovulation showed that flutamide suppressed ovulation, and a significant reduction in the number of ovulated oocytes was detected (Table 1), in agreement with a previous study in rats (8), suggesting that dose of flutamide was biologically effective.
|
To more precisely define the subcellular localization of AR protein in the fallopian tubes, immunohistochemical studies were performed to monitor AR nuclear translocation. AR immunoactivity was prominently detected in the cytoplasm of epithelial cells in immature mouse fallopian tubes. Furthermore, high levels of AR protein were detected in the nucleus of epithelial and smooth muscle cells when mice were treated with 24-h eCG (Fig. 4A). In contrast, sequential treatment with hCG did not affect AR relocation in the epithelial cells of fallopian tubes (data not shown). In addition, no difference was seen in AR immunoactivity between the different regions of the mouse fallopian tubes (Fig. 4A).
|
Effects of DES, DHT, or P4 on AR Protein Expression and Translocation on Mouse Fallopian Tubes
To determine whether expression of AR is steroid hormone dependent, fallopian tubes were collected from mice treated with DES, DHT, or P4 at different times. There was a clear difference between DES and DHT in the stimulatory effect on AR expression in mouse fallopian tubes in vivo. Western blot analysis showed that AR protein gradually increased after DES treatment, reaching a maximum level at 24 h, and remained high until 48 h, whereas increases in AR protein expression were seen only following 6-h DHT treatment. The levels of AR protein gradually declined to control levels (0 h) between 24 and 48 h after DHT treatment, indicating a trend in the opposite direction for regulation of AR protein expression between estrogens and androgens. However, P4 did not regulate AR protein expression (Fig. 5A).
|
Effects of DES, DHT, or P4 on p21Cip1 and p27Kip1 Protein Expression in Mouse Fallopian Tubes
The transduction of steroid hormone signals into the nucleus that leads to changes in gene expression has generally been shown to involve the steroid hormone receptor protein. AR predominantly functions as a transcription factor regulating the expression of target genes in hormone-sensitive cells (19, 26). To determine whether AR was functionally active, we performed parallel experiments to compare the DES-, DHT-, or P4-mediated effects of AR activation on p21Cip1 and p27Kip1 protein expression from the same mouse fallopian tube samples. Treatment with DES significantly increased p21Cip1 protein expression, with a peak at 6 h, whereas treatment with DHT gradually increased p21Cip1 protein expression, with a peak at 48 h in mouse fallopian tubes. However, p21Cip1 protein levels remained approximately constant following P4 treatment in mouse fallopian tubes (Fig. 6). In contrast, we could not detect any changes in p27Kip1 protein expression in total cell lysates of mouse fallopian tubes over a 48-h course after DES, DHT, or P4 treatment (Fig. 6), agreeing well with a previous observation that treatment with P4 did not regulate p27Kip1 protein expression in mouse fallopian tubes (54). These results indicate that the stimulation of p21Cip1 protein expression is relatively specific because the level of the related p27Kip1 protein did not change after hormone treatment.
|
To determine the effect of eCG on p21Cip1 protein expression in mouse fallopian tubes, we first demonstrated the localization of p21Cip1 in the fallopian tubes from mice treated with or without eCG. The immunohistochemical study showed weak p21Cip1 immunoreactivity in the nucleus of tubal epithelial cells in immature mice, whereas expression of p21Cip1 protein was particularly abundant in the nucleus of tubal epithelial cells in mice after 24-h eCG treatment (Fig. 7A). Furthermore, Western blot analysis confirmed that the effects of DES (Fig. 6) and eCG (Fig. 7B, left) in increasing p21Cip1 protein expression in mouse fallopian tubes are the same.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
interacts directly with AR (43) and that ERs have been identified in the fallopian tubes of human (1, 12, 29, 37) and rat (36, 44, 68). Indeed, the effects of fulvestrant (ICI 182,780), an ER antagonist, on suppression of AR expression and AR gene transcription have recently been reported (3), suggesting cross-regulation of ER and AR. These data imply both endogenous (induced by eCG) and exogenous (provided by DES) estrogens, acting directly via ERs in the fallopian tubes, as potent AR upregulators. In vivo studies are currently in progress to decipher the specific mechanisms by which ER subtypes ( and/or ) regulate AR expression in mouse fallopian tubes.
Apart from estrogen, we also show androgenic regulation of AR expression in mouse fallopian tubes. The data presented here suggest that androgen differs from estrogen to a large extent in the regulation of AR expression. First, blockade of androgen action by flutamide increased AR protein expression in the fallopian tubes in parallel with deceased serum testosterone levels in immature mice, suggesting that the eCG-induced increase in endogenous testosterone could be the cause of downregulated AR protein expression in mouse fallopian tubes in vivo. However, concomitant treatment with flutamide and eCG decreased AR protein expression in the fallopian tubes without changing serum testosterone levels. This could be explained in terms of the dose of flutamide used in the combined flutamide/eCG treatment being insufficient to suppress testosterone production, making it more difficult to block estrogen-mediated regulation of AR expression. Although the effect of flutamide treatment on the production of ovarian steroid hormones remains to be determined, it is likely that steroid hormone levels in the circulation are relative to those in ovarian production, which results in the modulation of AR protein expression. Second, significant increases in AR expression are observed at the earlier time (6 h) by treatment with DHT, a potent AR agonist, in mouse fallopian tubes. This rapid response may reflect the fact that nongenomic androgen-induced signaling appears to be mediated primarily by classical ARs that are associated with the plasma membranes of target cells (32). Previous studies have demonstrated that DHT induces AR mRNA expression in ER knockout mouse uteri (39). Therefore, it is unlikely that the action of androgen on AR regulation results from ER-dependent activation after conversion to 17-estradiol in the fallopian tubes in vivo. Last, although AR declines after long-term DHT treatment, levels of AR proteins at 48 h in DHT-treated mice are not significantly changed compared with those in oil-treated mice. Furthermore, AR immunoreactivity is predominantly detected in the cytoplasm of tubal epithelial cells, whereas AR nuclear localization is absent in the presence of DHT between 24 and 48 h. Therefore, it is considered likely that androgen-dependent decrease in AR protein expression is due to receptor recycling rather than degradation. Ovarian androgen synthesis occurs in the theca cells and can be metabolized to estrogen via P450 aromatase (P450arom) (14) in the granulosa cells of preovulatory follicles (16). Although P450arom is also expressed in human (30) and rat (64) fallopian tube epithelium, little is known concerning the regulation of this protein and its enzymatic activity in the mammal fallopian tubes. It remains unclear whether P450arom, which may influence the actions of either androgen and estrogen in vivo (22), is involved in the regulation of AR in the fallopian tubes.
Fallopian tubes are lined by a single-layered epithelium, which undergo cyclic alteration during the reproductive cycle as a result of changes in ovarian steroid hormone levels (11, 27). Since the initial subcellular localization of AR plays an important role in its cellular biological activity, we have made a detailed analysis of subcellular distribution of AR protein in the fallopian tubes. Results of the present study demonstrate that, without eCG and DES stimulation, ARs distribute in the cytoplasm of tubal epithelial cells. However, we observed a consistent and exclusive nuclear accumulation of AR (a 48-h time course) in the epithelium of mice in response to either endogenous or exogenous estrogen in vivo. Similarly, eCG-modulated AR translocation is also found in rat fallopian tube epithelium (data not shown). Action of androgen plays an essential role in the regulation of AR nuclear import and export in the target and nontarget tissues (5, 19, 26). However, our findings suggest that the accumulation of nuclear AR in rodent fallopian tubes is probably due to estrogen stimulation rather than androgen action, although short-time treatment with DHT (6 h) is capable of translocating AR into nuclei. These findings add further support that estrogens play a major role in upregulating total AR protein synthesis in the fallopian tubes. On the other hand, it is also possible that estradiol blocks the nuclear export process, disrupting the equilibrium between the cytoplasm and nucleus that results in AR being in the nucleus after exposure to DES (4). Taken together, there are major differences of the actions between estrogens and androgens on AR protein expression observed in the fallopian tubes, indicating that different or additional mechanism(s) may be operative in the regulation of AR in this tissue in vivo.
Previous studies (5, 19, 20) have suggested that regulation of AR activity can be achieved by modulation of AR gene expression, androgen binding to AR, and AR nuclear translocation and transactivation as well as AR protein stability. Our study does not exclude the possibility of a posttranslational control of AR protein. Several previous studies (6, 49) demonstrated that AR can be downregulated by ubiquitylation, generating proteins to promote proteasome-dependent degradation. On the other hand, it has been reported that, in an androgen-enhanced fashion, AR is modulated by the related, but functionally distinct, sumo (small ubiquitin-like modifiers) modification (sumoylation) (40, 46). Our preliminary work has also demonstrated an increase in sumoylation of AR in the fallopian tubes from mice treated with eCG (data not shown), consistent with an increase in AR protein expression by gonadotropin stimulation, suggesting that sumoylation may play a role in regulating the stabilization and function of AR in the fallopian tubes in vivo.
Although much less is known about the physiological function of AR in mammal fallopian tubes, it is well accepted (4, 19) that ligand-dependent translocation of AR into cell nuclei is of importance in stimulating cellular function. The mechanisms and factors involved in the proliferation and differentiation of fallopian tube cells are of wide interest. The cell cycle regulators p21Cip1 and p27Kip1 are members of Cip/Kip family that functions as a cyclin-dependent kinase inhibitor and appear to promote cell differentiation (57). In addition to participation in cell cycle arrest, however, p21Cip1 has also been implicated in inducing cell proliferation by interaction with proliferating cell nuclear antigen (9) and anti-apoptosis (31). Specifically, our data show that estrogen-induced AR protein expression is often associated with induction of p21Cip1, but not p27Kip1, in mouse fallopian tubes. Previously, p21Cip1 was characterized as an androgen-responsive gene since functional AREs have been reported in the promoter sequences of p21Cip1 (31). Furthermore, decreases in p21Cip1 protein expression are observed in ovarian granulosa cells in AR knockout mice (24), suggesting that regulation of p21Cip1 is AR dependent. Because nuclear localization is necessary for ARs to transactivate their target genes, our present study brings the first evidence that not only does nuclear AR physically interact with p21Cip1, but this interaction also responds to gonadotropin stimulation in mouse fallopian tubes in vivo. Since there is a high homology between the DNA-binding domains of the steroid hormone receptors (5, 19, 26), it may be that an ARE in the p21Cip1 gene (31) is not relevant only to AR-dependent transcriptional regulation in mouse fallopian tubes. At this stage, the existence of AR-independent regulation of p21Cip1 and/or additional coactivator proteins required for efficient transactivation cannot be excluded. Given the potential involvement of p21Cip1 in the receptor-mediated steroid hormone control of cell growth in vitro (38, 70), the consequences of an alteration in AR-p21Cip1 interaction during fallopian tube development in vivo need further investigation.
To the best of our knowledge, these studies are the first to demonstrate the differential expression and hormonal regulation of AR expression and localization in mouse fallopian tubes. Careful consideration of characterization studies clearly illustrates that estrogen induced upregulation of AR protein expression in the fallopian tubes in vivo. This study provides new insights on the differential regulation of AR expression and translocation by estrogen and androgen. Furthermore, in mouse fallopian tube epithelial cells, AR is located mainly in the cytoplasm and transmigrates to the nucleus following stimulation with endogenous and exogenous estrogens, which may allow AR to regulate target gene transcription. Indeed, positive regulation of interactions between AR and p21Cip1 in response to gonadotropin stimulation is observed. The biological significance of AR regulation in ovary and uterus has been described previously (24, 58), but the importance of such regulation for AR in the fallopian tubes has yet to be documented. Further studies will be needed to understand the specific roles and regulation of AR in normal tubal physiology.
|
GRANTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
* These authors contributed equally to this study. 
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONGRANTSREFERENCES |
|---|
