Suppression of ps20 Expression in the Rat Uterus by Tamoxifen and Estrogens
http://www.100md.com
内分泌学杂志 2005年第5期
Laboratory of Molecular Endocrinology, Division of Cellular and Molecular Research, National Cancer Center of Singapore, Singapore 169610
Address all correspondence and requests for reprints to: Dr. Hung Huynh, Laboratory of Molecular Endocrinology, Division of Cellular and Molecular Research, National Cancer Center of Singapore, Singapore 169610. E-mail: cmrhth@nccs.com.sg.
Abstract
Using differential display methodology, we isolated a tamoxifen-regulated cDNA. This cDNA was identical to the ps20 cDNA isolated from urogenital sinus mesenchymal cells. ps20 expression was detected in various female rat tissues, with the highest expression in lung and heart. ps20 transcripts were low during estrus and proestrus, but high during the diestrous stage of the estrous cycle coincident with estrogen-induced uterine cell proliferation. Treatment of ovary-intact or ovariectomized rats with estrogens or tamoxifen resulted in increased uterine weight and decreased ps20 expression. Uterine involution associated with ovariectomy or antiestrogen treatment led to up-regulation of ps20. Antibody against rat ps20 recognized the native rat ps20 in conditioned medium of primary rat uterine cells and stable ps20-transfected MCF-7 cells with molecular masses of approximately 24, 27, and 29 kDa. In primary rat uterine cells, ps20 secretion was enhanced by ICI 182,780, but was inhibited by estrogens and tamoxifen. Immunohistochemistry revealed that ps20 was localized to smooth muscle and luminal epithelial cells as well as the glandular population of uterine tissue. Conditioned medium derived from ps20-transfected MCF-7 cells, but not Escherichia coli recombinant ps20, exhibited mild growth suppression on PC-3 cells. The data indicate that ps20 expression is negatively regulated by estrogens and tamoxifen and suggest that ps20 may function as a mediator of local growth.
Introduction
ESTROGENS PROMOTE THE growth, differentiation, and remodeling of the uterus during the estrous cycle and pregnancy (1, 2, 3). The uterine tissue expresses and secretes a number of growth factors and other regulatory polypeptides in response to ovarian steroid hormones. These polypeptides are believed to play a part in directing or limiting the growth and development of the uterus.
Tamoxifen (TAM) has been reported to improve the survival of women with breast cancer and has proved to be clinically useful for the treatment of metastatic estrogen receptor-positive breast tumors (4, 5). However, long-term administration of TAM has been reported to be associated with endometrial thickening in some postmenopausal women (6) and increased risk of endometrial cancer in postmenopausal women (7). Over 40% of women receiving TAM had an endometrium more than 8 mm thick compared with only 5% of control women receiving placebo (6). It has been suggested that the estrogenic effect of TAM on the atrophic postmenopausal endometrium causes hyperplasia that may progress to atypia and cancer in a manner similar to that seen with estrogen replacement therapy (6). The molecular mechanisms responsible for TAM-induced endometrial hyperplasia are not well understood. We previously reported that TAM significantly increased uterine weight, whereas ICI 182,780 administration suppressed it (8). We also proposed that TAM might alter, in addition to enhancing IGF-I (8) and UO-44 (9) and inhibiting IGFBP-3 (10) gene expression in the uterus, the expression of other genes involved in the regulation of proliferation. To identify additional TAM- and estradiol-regulated genes in the uterus, differential display was used to examine the transcript expression profile of the ovariectomized (OVX) uterus under conditions of TAM supplementation. We isolated a TAM-regulated cDNA. This cDNA is identical to the cDNA encoded for a growth inhibitor isolated from urogenital sinus mesenchymal cells, ps20 (11). By virtue of its activation by antiestrogen and inhibition by estradiol and TAM, ps20 protein may play an important role in the growth regulation of normal and neoplastic uterine tissues.
Materials and Methods
Reagents
Horseradish peroxidase-conjugated donkey antimouse or antirabbit secondary antibodies were purchased from Pierce Chemical Co. (Rockford, IL). The chemiluminescent detection system was supplied by Amersham Biosciences (Arlington Heights, IL). Tissue culture dishes were purchased from Nunc, Inc. (Naperville, IL). RPMI 1640 medium, fetal bovine serum (FBS), Lipofectamine reagent, and penicillin-streptomycin were obtained from Invitrogen Life Technologies, Inc. (Grand Island, NY).
mRNA differential display
OVX rats were either untreated or treated with 2 mg TAM/kg·d (OVX-TAM) for 14 d. Total RNA was isolated from uteri using TRIzol reagent (Invitrogen Life Technologies, Inc., Carlsbad, CA) as described previously (9). Differential display was performed using RNA derived from uteri of OVX-vehicle- and OVX-TAM-treated rats according to the protocol supplied with the RNAmap kit (GeneHunter Corp., Nashville, TN) and as previously described (9).
Isolation of ps20 cDNA
The 270-bp probe of rat ps20 cDNA was used to screen rat uterine cDNA library as previously described (9). Clones identified by this probe were isolated and sequenced by the Sanger dideoxy chain termination method as previously described (9).
Animals and drug administration
Animal experiments were approved by the local animal care committee. Fifty-day-old, ovary-intact female Sprague Dawley rats were obtained from National University of Singapore, Singapore. OVX rats used in the experiments were 2 wk after ovariectomy.
To determine whether ps20 gene expression changes during the estrous cycle, uteri were collected from rats at estrous, proestrous, metestrous, and diestrous stages of cycle and frozen in liquid nitrogen for Northern blot analysis. Different stages of estrous cycle were ascertained by examining vaginal smears as described previously (9).
To study the in vivo effects of the antiestrogen ICI 182,780, TAM, and estrogens on ps20 gene expression in the rat uterine tissue, groups of ovary-intact rats were implanted with 0.5-, 1.0-, or 1.5-cm SILASTIC brand silicon tubes (inside diameter, 0.078 in.; Dow Corning, Midland, MI) containing 17?-estradiol (Sigma-Aldrich, Singapore) on the back of the neck. Control rats experienced the same surgical implantation with empty SILASTIC brand silicon tubes. Based on previous work (12), the release rate of 17?-estradiol from the implants was documented to be 2.4 μg/cm·d. TAM (Sigma-Aldrich) was dissolved in castor oil at a concentration of 10 mg/ml. Female rats daily received either 2 or 5 mg TAM/kg BW via sc injections. Preformulated ICI 182,780 (AstraZeneca Pharmaceuticals, Macclesfield, UK) was supplied at a concentration of 50 mg/ml in castor oil solution. Female rats were received weekly sc injections of either castor oil alone or three different doses (1, 1.5, or 2 mg) of ICI 182,780/kg BW for 3 wk. At the end of the experiments, animals were killed by carbon dioxide exposure. The uteri were excised, trimmed, weighed, and snap-frozen in liquid nitrogen and stored at –70 C for RNA extraction. Part of the uterine tissue was fixed in neutral buffer containing 10% formalin and embedded in paraffin for immunohistochemistry. To study the effects of ovarian hormones on ps20 gene expression, female rats were OVX, and the uteri were collected at various times after ovariectomy for ps20 mRNA determination. To study the effects of estradiol and progesterone on ps20 gene expression, OVX rats were implanted with 1.5-cm SILASTIC brand silicon tubes containing either 17?-estradiol or progesterone, and the uterine tissue was harvested at the indicated times after hormonal supplementation.
ps20 Antibody
Rabbit antirat ps20 antibodies were raised against amino acid peptides in the N terminal of rat ps20. The amino acid sequences of rat ps20-specific peptide were as follows: TWEAMLPVRLAEKSQAEEVA. Affinity-purified ps20 antibody was diluted in Tris-buffered saline (20 mM Tris and 200 mM NaCl, pH 7.6) containing 0.1% Tween 20 at a final concentration of 1 μg/ml. The specificity of ps20 antiserum was verified by Western blot analysis of conditioned medium (CM)-derived MCF-7 cells that were transfected with full-length rat ps20 cDNA.
Immunohistochemical analysis and assessment
For immunohistochemistry, sections (5 μm thick) were cut, dewaxed in xylene, and then rehydrated as previously described (13). Antigen retrieval was performed by boiling the slides in 10 mM citrate buffer, pH 6.0, for 20 min. After blocking unspecific binding with 5% skim milk, sections were incubated with rabbit antirat ps20 (1:1000 dilution) overnight at 4 C as previously described (14). Immunohistochemistry was performed using the streptavidin-biotin peroxidase complex method according to the manufacturer’s instructions (Lab Vision, Fremont, CA) using 3,3'-diaminobenzidine as the chromogen. Sections known to stain positively were incubated in each batch, and negative controls were also prepared by replacing primary antibody with preimmune serum.
Primary uterine cell isolation and treatment
Uterine tissue derived from 10-wk-old rats was finely minced and washed three times with serum-free MEM (SRF). The minced tissue was incubated with MEM containing 5% FBS and 5 mg/ml collagenase A (Roche, Indianapolis, IN) at 37 C for 12 h as described previously (10). Cells were harvested by centrifuging at 800 x g for 10 min. The cell pellets were washed three times with SRF medium and allowed to grow in MEM containing 10% FBS. To study the effects of ps20 on the proliferation of PC-3 human prostate cancer cells and primary rat uterine cells, cells were plated at a density of 1 x 104 cells/well in 96-well plates and allowed to grow in growth medium for 24 h. Cells were washed once with SRF medium, then incubated in SRF medium for another 18 h. To determine the effects of bacterial recombinant ps20, cells were then treated with indicated concentrations of bacterial recombinant ps20. To study the effects of ps20 produced by MCF-7 cells on cellular proliferation, PC-3 cells were treated with SRF medium containing either 30% CM of ps20-transfected or 30% CM of pcDNA-3-transfected MCF-7 cells for 48 h. [3H]Thymidine incorporation was determined 48 h after treatment as previously described (15). Experiments were repeated three times, and the data were expressed as the mean ± SD.
Western blot analysis
To determine whether the different forms of secreted ps20 were due to differential glycosylation, concentrated CM derived from pcDNA-3-transfected MCF-7 cells was incubated with 10 U recombinant N-glycanase enzyme (Genzyme, Cambridge, MA) as described by the manufacturer. N-Glycanase-digested proteins were subjected to Western blot analysis as previously described (13). To examine the effects of estrogens, TAM, and ICI 182,780 on ps20 secretion, primary rat uterine cells were plated at a density of 5 x 106 cells/100-mm dish in the growth medium. After 24 h, the cell monolayer was washed, then treated with SRF medium containing 0.05% ethanol, 10–9 M ICI 182,780, 10–8 M TAM, and 10–10 M 17?-estradiol for 48 h. After treatment, CM was harvested. To determine changes in the levels of ps20 in CM, 1 ml CM was concentrated to 0.1 ml and subjected to Western blot analysis as previously described (13). Blots were incubated with rabbit antirat antibodies (1:4000) and horseradish peroxidase-conjugated donkey antimouse or antirabbit secondary antibodies (1:7500). Blots were then visualized with a chemiluminescent detection system (Amersham Biosciences) as described by the manufacturer.
Northern blotting
Total RNA was isolated from indicated tissues of female rats as previously described (16). Northern blots were performed on total RNA, and blots were hybridized with rat ps20 or human GAPDH (American Type Culture Collection, Manassas, VA) cDNAs as previously described (9). mRNA levels were determined by densitometric scanning of autoradiographs.
Stable ps20-transfected MCF-7 cells
The entire coding region of rat ps20 cDNA was cloned into the mammalian expression vector pcDNA3.1 (Invitrogen Life Technologies, Inc.) to create the ps20-pcDNA3.1 expression vector. The ps20-pcDNA3.1 sequence was confirmed by sequencing. MCF-7 cells were seeded at a density of 2 x 105 in 100-mm culture dishes in 90% -MEM (Sigma-Aldrich) containing 10% FBS with 1% penicillin -streptomycin for 24 h before transfection. Cells were transfected with either 5 μg ps20-pDNA3.1 or pDNA3.1 and 28 μl Lipofectamine reagent (Invitrogen Life Technologies, Inc.) as previously described (14). Forty-eight hours after transfection, cells were subcultured at 1:10 and replaced with -MEM containing 10% FBS and 800 μg/ml G418 (Calbiochem, La Jolla, CA). After 4 wk of selection, individual clones were isolated, expanded, and assayed for ps20 expression by Western blot analysis.
Expression and purification of ps20-His
The coding region of rat ps20 cDNA without signal peptide was cloned into pQE-30 bacterial expression vector (Qiagen, Mississauga, Canada), a vector that allows incorporation of a His-Tag at the carboxyl terminus. The ps20-QE-containing clone was confirmed by sequencing and transformed into the M15 bacterial expression cell line (Qiagen). Five hundred milliliters of bacterial cultures were grown and induced with 1 mM isopropyl-1-thio-?-D-galactopyranoside for 2 h. Soluble extract was generated by suspending bacteria in buffer containing 8 M urea, 0.1 M NaH2PO4, and 0.01 M Tris-HCl, pH 8.0, and kept at room temperature for 30 min. The soluble extract was centrifuged at 13,000 rpm for 30 min. Recombinant ps20-His protein was purified on a nickel-charged resin (Qiagen) as described by the manufacturer. Fractions containing proteins were pooled and dialyzed against 1 M acetic acid (pH 2.5) overnight at 4 C. The dialyzed samples were quickly frozen, lyophilized, and stored at –80 C.
Statistical analysis
For quantitation analysis, the sum of the density of bands corresponding to mRNA hybridized to ps20 probe or protein blotting with the antibody under studied was calculated and normalized to the amount of GAPDH mRNA or -tubulin, respectively. Differences in ps20 mRNA levels, [3H]thymidine incorporation, and the levels of ps20 were analyzed by ANOVA.
Results
Using differential display methodology, several differentially displayed bands representing cDNA corresponding to genes whose expression was inhibited by TAM were isolated. One of the bands, which was highly expressed in the uterine tissue of OVX rats, but significantly reduced in uterine tissue derived from OVX and TAM-treated rats (Fig. 1A), was selected for additional characterization. This 270-bp fragment was reamplified by PCR using the same primers and conditions as differential display and was used to probe Northern blots of mRNA obtained from uterine tissue of OVX and OVX TAM-treated rats. A strong signal corresponding to approximately 1.2 kb emerged in RNA isolated from OVX vehicle-treated uterine tissue, whereas faint signal was observed in the RNA derived from uterine tissue of OVX TAM-treated rats (Fig. 1B). The data confirmed that the isolated gene was down-regulated by TAM. This 270-bp fragment was then subcloned into TA vector and subjected to nucleotide sequence analysis. Comparison of the nucleotide sequence obtained against the nonredundant nucleotide database of GenBank established that this cDNA was a 100% match with rat ps20 cDNA (11).
FIG. 1. Differentially expressed mRNA bands, ps20 expression in uterine tissue of OVX vehicle-treated and OVX TAM-treated rats, and ps20 mRNA levels in various tissues of ovary-intact female rats. A, Total RNA isolated from uterine tissue of OVX vehicle-treated and OVX TAM-treated rats was subjected to differential display. The arrowhead indicates mRNA reduced by TAM treatment. B, Northern blot analysis of total RNA from uterine tissue of OVX vehicle-treated and OVX TAM-treated rats was used to confirm the presence of a differentially expressed mRNA in OVX TAM-treated uterine tissue. Blots were hybridized with a 32P-labeled DNA fragment that was isolated from the above differential gel or glyceraldehyde-3-phosphate dehydrogenase cDNA. C, Total RNA derived from various tissues of an 80-d-old female rat was subjected to Northern blot analysis. Blots were hybridized with rat ps20 cDNA. The ethidium bromide staining of 28S and 18S ribosomal RNA was used as a control loading. Tissues are: Mg, mammary gland; Ut, uterus; Fa, abdominal fat; Ov, ovary; Lu, lung; He, heart; Ce, cerebellum; Br, brain; Int, small intestine; μ, red muscle; Li, liver; Ki, kidney; Spl, spleen; Bl, bladder; Pla, placenta.
To determine ps20 expression in rat tissues, total RNA derived from various tissues of mature female rats was analyzed by Northern blotting. Figure 1C shows that ps20 transcripts of approximately 1.2 kb were observed in all tissues examined, with the highest levels detected in lung, heart, bladder, cerebellum, ovary, uterus, and placenta. In addition, ps20 mRNA was found in rat prostate (data not shown).
To study the effects of estrogens and TAM on the uterine growth and expression of ps20 in the uterus, ovary-intact rats were treated with various doses of TAM, estradiol, and the antiestrogen ICI 182,780. Figure 2A shows that after estradiol implantation for 3 wk, a dose-dependent decrease in ps20 expression was observed (P < 0.01). Similarly, at doses of 2 and 5 mg/kg body weight, TAM significantly reduced ps20 mRNA by 45% and 50%, respectively (Fig. 2B; P < 0.01). ICI 182,780, in contrast, significantly increased ps20 mRNA levels in uterine tissue compared with the controls (Fig. 2C; P < 0.01). Maximal induction was observed at a dose of 1 mg ICI 182,780/kg body weight/wk (Fig. 2C). The results indicate that TAM acts as an estrogen agonist to inhibit ps20 expression. In all cases, an inverse relationship between ps20 gene expression and uterine weight existed.
FIG. 2. Effects of estradiol, TAM, and ICI 182,780 on ps20 gene expression. Ovary-intact rats were treated with the indicated doses of 17?-estradiol by implantation (A) or were injected with two different doses of TAM (B) or the three indicated doses of ICI 182,780 (C) for 3 wk as described in Materials and Methods. Total RNA derived from uterine tissue was subjected to Northern blot analysis. Blots were hybridized with glyceraldehyde-3-phosphate dehydrogenase and rat ps20 cDNAs. Densitometric scanning of the ps20 band is shown. Data are expressed as the mean ± SE. Bars with different letters are significant at P < 0.01.
To investigate whether ps20 expression in the uterine tissue was estrogen dependent, ps20 mRNA levels were determined at different times after ovariectomy. As shown in Fig. 3A, 6 h after ovariectomy the levels of ps20 mRNA were increased by 4-fold and reached maximum levels by 48 h. Treatment of OVX rats with 17?-estradiol resulted in a down-regulation of ps20 expression (Fig. 3B). Progesterone, a hormone that has little or no effect on OVX-induced uterine involution, had very little or no effect on OVX-induced ps20 gene expression (data not shown).
FIG. 3. Effects of ovariectomy and estradiol on ps20 expression. A, Ovary-intact rats were OVX, and uterine tissue was collected at the indicated times after ovariectomy as described in Materials and Methods. B, OVX rats were implanted with SILASTIC brand silicon tubes containing 17?-estradiol. The rate of estradiol released was approximately 3.6 μg/d. Total RNA derived from uterine tissue was subjected to Northern blot analysis. Blots were hybridized with glyceraldehyde-3-phosphate dehydrogenase and rat ps20 cDNAs. Densitometric scanning of the ps20 band is shown. Data are expressed as the mean ± SE.
FIG. 6. Immunolocalization of ps20 protein in rat uterine tissue. Rat uterine tissue from vehicle-, estradiol-, TAM-, and ICI 182,780-treated rats was collected, and paraffin blocks were prepared. Uterine tissue sections were stained with rabbit antirat ps20 antibody as described in Materials and Methods. In vehicle-treated uterine tissue (A), ps20 protein was detected in smooth muscle cells as well as in luminal epithelial cells. The immuno-signal was significantly reduced in estradiol-treated (B) and TAM-treated (C) cells of uterine tissue. In ICI 182,780-treated uterus (D), ps20 protein was also observed on the apical surface of glandular and luminal epithelial cells. Original magnification, x400.
Because ps20 gene expression was regulated by estrogens and TAM, we wanted to determine whether ps20 gene expression fluctuated during different stages of estrous cycle. Total RNA derived from uterine tissue of rats at different stages of the cycle was analyzed by Northern blotting. As shown in Fig. 4, minimal expression of ps20 occurred at proestrous and estrous stages coincident with estrogen-induced uterine cell proliferation. High levels of ps20 mRNA were detected in uterine tissue during metestrous and diestrous.
FIG. 4. Expression of ps20 gene in rat uterus at different stages of the estrous cycle. Total RNA derived from uterine tissue of rats at different stages of the estrous cycle was analyzed by Northern blotting. Blots were hybridized with either glyceraldehyde-3-phosphate dehydrogenase or ps20 cDNAs. A, Each lane represents one phase of the estrous cycle. Densitometric scanning of the ps20 band is shown in B. Data are expressed as the mean ± SE. Bars with different letters are significant at P < 0.01.
To gain additional information about the function of ps20 in uterine tissue, we developed the rabbit polyclonal antibody against rat recombinant ps20 protein. CM from MCF-7 cells transfected with rat ps20 cDNA was used to determine the antibody specificity. Western blot analysis revealed that three specific bands of apparent molecular mass 24, 27, and 29 kDa were detected in CM derived from ps20-transfected, but not pc-DNA-transfected MCF-7 cells, confirming that the antibody was specific (Fig. 5A). A similar pattern was observed in CM derived from primary rat uterine cells (Fig. 5B). Treatment of CM with N-glycanase converted the 27- and 29-kDa species to the 24-kDa form (Fig. 5B), suggesting that ps20 is glycosylated.
FIG. 5. Determination of the anti-ps20 antibody specificity, glycosylation status of ps20, and ps20 growth inhibitory activity. Rabbit polyclonal antibody against rat ps20 was developed as described in Materials and Methods. CM collected from mock-transfected and ps20-transfected MCF-7 cells was concentrated and subjected to Western blot analysis as described in Materials and Methods. Note that the antibody recognized proteins of approximately 24, 27, and 29 kDa in CM derived from ps20-transfected, but not mock-transfected, MCF-7 cells (A). CM derived from ps20-transfected MCF-7 cells was incubated with (+) and without (–) N-glycanase (B). Note that the 24-kDa band became prominent when ps20 protein was digested with N-glycanase. C, Purified bacterial recombinant ps20. D, Primary rat uterine cells and PC-3 cells were grown and treated with SRF containing the indicated doses of bacterial recombinant ps20 for 48 h as described in Materials and Methods. E, PC-3 cells were treated with 30% CM derived from pcDNA or ps20-transfected MCF-7 cells for 48 h. [3H]Thymidine incorporation was determined as described in Materials and Methods. Experiments were performed in quadruplicate, with the results reflecting the mean ± SE. Bars with different letters are significantly different from one another at P < 0.01. The experiments were repeated three times with similar results.
It has been reported that recombinant ps20 inhibited PC-3 prostate cancer cell proliferation (11). To determine whether ps20 had any growth inhibitory activity on rat uterine cells, cells were treated with various doses of bacterial recombinant ps20. PC-3 cells were used as controls. Unlike previous studies (11), which showed that bacterial recombinant ps20 inhibited PC-3 cells in vitro, we observed that bacterial recombinant ps20 had no effect on the growth of either primary uterine cells or PC-3 cells (Fig. 5D). Our data raised the possibility that posttranslational modification of ps20 protein may play a role in growth regulation. To investigate this possibility, PC-3 cells were treated with serum-free medium containing either 30% CM obtained from stably ps20-transfected MCF-7 cells or 30% CM derived from pcDNA-transfected MCF-7 cells. Figure 5E showed that [3H]thymidine incorporation was significantly lower in cells treated with CM derived from ps20-transfected MCF-7 cells compared with cells grown in CM-derived, mock-transfected MCF-7 cells (P < 0.01). These observations raise the possibility that posttranslational modification of ps20 protein may play a role in its growth inhibitory activity.
To determine the cell type-specific expression of ps20 in uterine tissue, immunohistochemistry was performed on sections of uteri derived from vehicle-, ICI 182,780-, estradiol-, and TAM-treated rats using antirat ps20 antibody. High ps20 expression was detected in smooth muscle cells. Faint signal was also detected in luminal epithelial cells and the glandular population of the uteri (Fig. 6A). Treatment of rats with 17?-estradiol (Fig. 6B) and TAM (Fig. 6C) resulted in a significant reduction in ps20 protein in smooth muscle and luminal epithelial cells. ps20 expression in the smooth muscle cells and particularly in luminal epithelial cells was greatly increased by ICI 182,780 treatment (Fig. 6D). In ICI 182,780-treated uterine tissue, a strong immunostaining was also detected on the apical surface of glandular and luminal epithelial cells (Fig. 6D). Similar staining patterns were seen when the sections derived from uterine tissue of OVX rats were stained with ps20 antibody (data not shown). No staining was observed in smooth muscle cells when the antibody was preadsorbed with ps20 peptide (data not shown).
To determine whether estrogens also regulated ps20 secretion, primary rat uterine cells were isolated, grown, and treated with SRF medium containing vehicle, 10–9 M ICI 182,780, 10–8 M TAM, and 10–10 M 17?-estradiol for 48 h. Figure 7 shows that although ICI 182,780 caused a 2-fold increase in ps20 accumulation in CM, exposure of cells to TAM and estradiol resulted in 50% and 90% reductions in basal ps20 secretion, respectively.
FIG. 7. Effects of estrogens, TAM, and ICI 182,780 on ps20 protein secretion by primary rat uterine cells. Primary rat uterine cells were isolated, grown, and treated with SRF containing vehicle (C), 10–9 M ICI 182,780 (ICI), 10–8 M TAM, and 10–10 M 17?-estradiol (E2) for 48 h as described in Materials and Methods. CM was harvested for Western blot analysis as described in Materials and Methods. Blots were incubated with antirat ps20 antibody. Representative blots are shown. Experiments were repeated three times with similar results.
Discussion
Long-term administration of TAM has been reported to be associated with increased risk of endometrial cancer in postmenopausal women (7). However, the molecular mechanisms underlying this effect are not well understood. In the present study we isolate an estradiol- and TAM-regulated cDNA, ps20, using differential display and cDNA library screening. TAM has been known to regulate the expression of a series of estrogen-regulated genes, including the progesterone receptor, epidermal growth factor, TGF-, TGF-?, IGF-I receptor, epidermal growth factor receptor, and pS2 (reviewed in Ref. 4), IGF-I (8), UO-44 (9), and IGFBP-3 (10) in uterine tissue. In the present study we demonstrate for the first time that estrogens and TAM inhibit ps20 expression in this tissue. Levels of ps20 transcripts are high in heart, lung, and tissues containing a high proportion of smooth muscle cells, but are low in tissues with more epithelial cells. The localization of ps20 in smooth muscle as well as luminal epithelial cells in uterine tissue suggests that ps20 may play an autocrine and/or paracrine role in the regulation of cellular proliferation and the maintenance of homeostasis in this tissue. The high expression of ps20 found in uterine smooth muscle is consistent a previous study (11) showing that in the prostate, smooth muscle cells express high levels of ps20.
Low expression of ps20 occurred at the estrous and proestrous stages of the ovarian cycle coincident with estrogen-induced uterine cell proliferation, in contrast to high levels during metestrus and diestrus. In rodents, circulating levels of estrogens are high in proestrus and low in diestrus. Concomitant with this rise and fall in circulating estrogen levels, there is a decline and rise in ps20 mRNA levels in uterine tissue. This pattern of expression during the reproductive cycle suggests that ps20 plays a role in growth regulation. A rapid increase in ps20 gene expression after ovariectomy and antiestrogen ICI 182,780 treatment also supports the inhibitory effects of estrogens on ps20 expression.
The precise mechanism(s) by which TAM inhibits ps20 gene expression is not known at present. TAM is a classical partial agonist and exhibits tissues specificity for inducing either an agonist or antagonist response. In rats and humans, TAM exhibits partial agonist (reviewed in Ref. 4). It produces antagonist effects in breast, but agonist effects in vagina and endometrium (reviewed in Ref. 4). Long-term TAM use is generally associated with a reduced incidence of contralateral breast cancer (antagonist), a reduced incidence of primary breast cancer in high risk women (antagonist), maintenance of bone density (agonist), and an increased risk of endometrial carcinomas (agonist) (17). ICI 182,780 seems to be an antagonist; it is devoid of agonist activity in the uterus or vagina of rats (18). Although TAM and estradiol suppress ps20 expression in uterine tissue, TAM is approximately 800- to 1500-fold less potent than estradiol. The decrease in ps20 expression after TAM treatment suggests that TAM acts as an estrogen agonist in the uterus. The results reported here are clinically relevant, because women treated with TAM frequently exhibit endometrial hyperplasia (19) and rarely show neoplasia (7, 20, 21). TAM has been proposed as a treatment for neoplastic conditions of the uterus (22), but the results of clinical trials have not been impressive, and there are clinical and laboratory data suggesting that stimulation of endometrial neoplastic growth and leiomyoma growth by TAM are possible (23, 24). It is possible that the estrogen-like effect of TAM on ps20 expression in uterine tissue, as described here, is related to its adverse effects.
An immunohistochemical study reveals that ps20 protein expression is not only confined to smooth muscle cells, but is also present in luminal and glandular epithelial cells of the uterus. This observation is consistent with the recent report (25) that ps20 is localized in stromal cells as well as in a more aggressive epithelial phenotype. High levels of ps20 gene expression are found in smooth muscle cells and the apical surface of luminal epithelial cells of ICI 182,780-treated uterine tissue. Although ps20 expression is inversely correlated with uterine growth, the biological function of ps20 in the uterus is as yet unclear. ps20 Belongs to the whey acidic protein-type, four-disulfide core domain protein family (11). They are serine protease inhibitors that exhibit a variety of growth and differentiation functions. Recently, McAlhany et al. (26) proposed that ps20 is a TGF-?1-induced regulator of angiogenesis. They suggest that ps20 functions by either promoting endothelial cell migration or contributing to pericyte stabilization of newly formed vascular structure (26). In the present study we observed that estrogens, which are known to induce angiogenesis, potently inhibit ps20 expression. Ovariectomy and ICI 182,780, in contrast, are potent stimulators of ps20 expression and inhibitors of angiogenesis. Thus, in the uterus, ps20 may function as a local growth regulator, rather than a regulator of angiogenesis.
Previous studies have shown that recombinant ps20 inhibited the proliferation of prostate carcinoma PC-3 cells (11) and was suggested to function as a mediator of local growth and differentiation mechanisms. In the present study we did not observe any growth inhibition when bacterial recombinant ps20 was used to treat the same cell line. The discrepancy between our study and the previous one (11) is not clear. We believe that posttranslational modification of the ps20 protein plays an important role in determining its growth inhibitory activity. This hypothesis is supported by the observations that three specific bands of apparent molecular mass 24, 27, and 29 kDa are detected in CM of ps20-transfected MCF-7 cells. Similar results are observed when CM from rat primary uterine cells is analyzed by Western blot analysis. However, the 24-kDa form of ps20 becomes the major species when CM is treated with N-glycanase. This suggests that ps20 undergoes posttranslational modification before secretion.
In summary, our data demonstrate that ps20 gene expression in uterine tissue is a molecular marker that inversely correlates well with the positive or negative uterotropic effects of estrogen receptor antagonists and partial agonists. The characterization of ps20 protein provides new knowledge about the roles of ps20 protein in estradiol- and TAM-induced cellular proliferation and cancer in this tissue.
References
Jensen EV, DeSombre ER 1972 Mechanism of action of the female sex hormones. Annu Rev Biochem 41:203–230
Anderson JN, Peck Jr EJ, Clark JH 1975 Estrogen-induced uterine responses and growth: relationship to receptor estrogen binding by uterine nuclei. Endocrinology 96:160–167
Katzenellenbogen BS, Gorski J 1975 Biochemical actions of hormones. New York: Academic Press
Clarke R, Leonessa F, Welch JN, Skaar TC 2001 Cellular and molecular pharmacology of antiestrogen action and resistance. Pharmacol Rev 53:25–71
Jordan VC, Morrow M 1999 Tamoxifen, raloxifene, and the prevention of breast cancer. Endocr Rev 20:253–278
Powles TJ, Chang J 1997 Chemoprevention of breast cancer: why is tamoxifen not the answer? Endocr Relat Cancer 4:135–139
Fornander T, Cedermark B, Mattsson S, Skoog L, Theve T, Askergren J, Rutqvist LE, Glas U, Silfversward C, Somell A, Wilking N, Hjalmar ML 1989 Adjuvant tamoxifen in early breast cancer: occurrence of new primary cancers. Lancet 117–120
Huynh H, Pollak M 1993 IGF-1 gene expression in the uterus is stimulated by tamoxifen and inhibited by the pure antiestrogen ICI 182780. Cancer Res 53:5585–5588
Huynh H, Ng CY, Lim KB, Ong CK, Ong CS, Tran E, Tuyen Nguyen TT, Chan TW 2001 Induction of UO-44 gene expression by tamoxifen in the rat uterus and ovary. Endocrinology 142:2985–2995
Huynh H, Pollak M 1994 Uterotrophic actions of estradiol and tamoxifen are associated with inhibition of uterine IGF binding protein 3 gene expression. Cancer Res 54:3115–3119
Larsen M, Ressler SJ, Lu B, Gerdes MJ, McBride L, Dang TD, Rowley DR 1998 Molecular cloning and expression of ps20 growth inhibitor: a novel WAP-type "four-disulfide core" domain protein expressed in smooth muscle. J Biol Chem 273:4574–4584
Robaire B, Ewing LL, Irby DC, Desjardins C 1979 Interaction of testosterone and estradiol-17? on the reproductive tract of the male rat. Biol Reprod 21:455–463
Huynh H, Chow PK, Ooi LL, Soo KC 2002 A possible role for insulin-like growth factor-binding protein-3 autocrine/paracrine loops in controlling hepatocellular carcinoma cell proliferation. Cell Growth Differ 13:115–122
Ong CK, Ng CY, Leong C, Ng CP, Foo KT, Tan PH, Huynh H 2004 Genomic structure of human OKL38 gene and its differential expression in kidney carcinogenesis. J Biol Chem 279:743–754
Huynh H, Yang X, Pollak M 1996 Estradiol and antiestrogens regulate a growth inhibitory insulin-like growth factor binding protein 3 autocrine loop in human breast cancer cells. J Biol Chem 271:1016–1021
Huynh H, Tetenes E, Wallace L, Pollak M 1993 In vivo inhibition of insulin-like growth factor-I gene expression by tamoxifen. Cancer Res 53:1727–1730
Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, Vogel V, Robidoux A, Dimitrov N, Atkins J, Daly M, Wieand S, Tan-Chiu E, Ford L, Wolmark N 1998 Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 90:1371–1388
Wakeling AE, Bowler J 1988 Novel antioestrogens without partial agonist activity. J Steroid Biochem 31:645–653
Fornander T, Rutqvist LE, Wilking N 1991 Effects of tamoxifen on the female genital tract. Ann NY Acad Sci 622:469–476
Andersson M, Storm HH, Mouridser HT 1991 Incidence of new primary cancers after adjuvant tamoxifen therapy and radiotherapy for early breast cancer. J Natl Cancer Inst 83:1013–1017
Magriples U, Naftolin F, Schwartz PE, Carcangiu ML 1993 High-grade endometrial carcinoma in tamoxifen-treated breast cancer patients. J Clin Oncol 11:485–490
Swenerton KD 1980 Treatment of advanced endometrial adenocarcinoma with tamoxifen. Cancer Treat Rep 64:805–811
Jordan VC, Gottardis MM, Satyaswaroop PG 1991 Tamoxifen-stimulated growth of human endometrial carcinoma. Ann NY Acad Sci 622:439–446
Dilts PV, Hopkins MP, Chang AE, Cody RL 1992 Rapid growth of leiomyoma in patient receiving tamoxifen. Am J Obstet Gynecol 166:167–168
McAlhany SJ, Ayala GE, Frolov A, Ressler SJ, Wheeler TM, Watson JE, Collins C, Rowley DR 2004 Decreased stromal expression and increased epithelial expression of WFDC1/ps20 in prostate cancer is associated with reduced recurrence-free survival. Prostate 61:182–191
McAlhany SJ, Ressler SJ, Larsen M, Tuxhorn JA, Yang F, Dang TD, Rowley DR 2003 Promotion of angiogenesis by ps20 in the differential reactive stroma prostate cancer xenograft model. Cancer Res 63:5859–5865(Huynh Hung)
Address all correspondence and requests for reprints to: Dr. Hung Huynh, Laboratory of Molecular Endocrinology, Division of Cellular and Molecular Research, National Cancer Center of Singapore, Singapore 169610. E-mail: cmrhth@nccs.com.sg.
Abstract
Using differential display methodology, we isolated a tamoxifen-regulated cDNA. This cDNA was identical to the ps20 cDNA isolated from urogenital sinus mesenchymal cells. ps20 expression was detected in various female rat tissues, with the highest expression in lung and heart. ps20 transcripts were low during estrus and proestrus, but high during the diestrous stage of the estrous cycle coincident with estrogen-induced uterine cell proliferation. Treatment of ovary-intact or ovariectomized rats with estrogens or tamoxifen resulted in increased uterine weight and decreased ps20 expression. Uterine involution associated with ovariectomy or antiestrogen treatment led to up-regulation of ps20. Antibody against rat ps20 recognized the native rat ps20 in conditioned medium of primary rat uterine cells and stable ps20-transfected MCF-7 cells with molecular masses of approximately 24, 27, and 29 kDa. In primary rat uterine cells, ps20 secretion was enhanced by ICI 182,780, but was inhibited by estrogens and tamoxifen. Immunohistochemistry revealed that ps20 was localized to smooth muscle and luminal epithelial cells as well as the glandular population of uterine tissue. Conditioned medium derived from ps20-transfected MCF-7 cells, but not Escherichia coli recombinant ps20, exhibited mild growth suppression on PC-3 cells. The data indicate that ps20 expression is negatively regulated by estrogens and tamoxifen and suggest that ps20 may function as a mediator of local growth.
Introduction
ESTROGENS PROMOTE THE growth, differentiation, and remodeling of the uterus during the estrous cycle and pregnancy (1, 2, 3). The uterine tissue expresses and secretes a number of growth factors and other regulatory polypeptides in response to ovarian steroid hormones. These polypeptides are believed to play a part in directing or limiting the growth and development of the uterus.
Tamoxifen (TAM) has been reported to improve the survival of women with breast cancer and has proved to be clinically useful for the treatment of metastatic estrogen receptor-positive breast tumors (4, 5). However, long-term administration of TAM has been reported to be associated with endometrial thickening in some postmenopausal women (6) and increased risk of endometrial cancer in postmenopausal women (7). Over 40% of women receiving TAM had an endometrium more than 8 mm thick compared with only 5% of control women receiving placebo (6). It has been suggested that the estrogenic effect of TAM on the atrophic postmenopausal endometrium causes hyperplasia that may progress to atypia and cancer in a manner similar to that seen with estrogen replacement therapy (6). The molecular mechanisms responsible for TAM-induced endometrial hyperplasia are not well understood. We previously reported that TAM significantly increased uterine weight, whereas ICI 182,780 administration suppressed it (8). We also proposed that TAM might alter, in addition to enhancing IGF-I (8) and UO-44 (9) and inhibiting IGFBP-3 (10) gene expression in the uterus, the expression of other genes involved in the regulation of proliferation. To identify additional TAM- and estradiol-regulated genes in the uterus, differential display was used to examine the transcript expression profile of the ovariectomized (OVX) uterus under conditions of TAM supplementation. We isolated a TAM-regulated cDNA. This cDNA is identical to the cDNA encoded for a growth inhibitor isolated from urogenital sinus mesenchymal cells, ps20 (11). By virtue of its activation by antiestrogen and inhibition by estradiol and TAM, ps20 protein may play an important role in the growth regulation of normal and neoplastic uterine tissues.
Materials and Methods
Reagents
Horseradish peroxidase-conjugated donkey antimouse or antirabbit secondary antibodies were purchased from Pierce Chemical Co. (Rockford, IL). The chemiluminescent detection system was supplied by Amersham Biosciences (Arlington Heights, IL). Tissue culture dishes were purchased from Nunc, Inc. (Naperville, IL). RPMI 1640 medium, fetal bovine serum (FBS), Lipofectamine reagent, and penicillin-streptomycin were obtained from Invitrogen Life Technologies, Inc. (Grand Island, NY).
mRNA differential display
OVX rats were either untreated or treated with 2 mg TAM/kg·d (OVX-TAM) for 14 d. Total RNA was isolated from uteri using TRIzol reagent (Invitrogen Life Technologies, Inc., Carlsbad, CA) as described previously (9). Differential display was performed using RNA derived from uteri of OVX-vehicle- and OVX-TAM-treated rats according to the protocol supplied with the RNAmap kit (GeneHunter Corp., Nashville, TN) and as previously described (9).
Isolation of ps20 cDNA
The 270-bp probe of rat ps20 cDNA was used to screen rat uterine cDNA library as previously described (9). Clones identified by this probe were isolated and sequenced by the Sanger dideoxy chain termination method as previously described (9).
Animals and drug administration
Animal experiments were approved by the local animal care committee. Fifty-day-old, ovary-intact female Sprague Dawley rats were obtained from National University of Singapore, Singapore. OVX rats used in the experiments were 2 wk after ovariectomy.
To determine whether ps20 gene expression changes during the estrous cycle, uteri were collected from rats at estrous, proestrous, metestrous, and diestrous stages of cycle and frozen in liquid nitrogen for Northern blot analysis. Different stages of estrous cycle were ascertained by examining vaginal smears as described previously (9).
To study the in vivo effects of the antiestrogen ICI 182,780, TAM, and estrogens on ps20 gene expression in the rat uterine tissue, groups of ovary-intact rats were implanted with 0.5-, 1.0-, or 1.5-cm SILASTIC brand silicon tubes (inside diameter, 0.078 in.; Dow Corning, Midland, MI) containing 17?-estradiol (Sigma-Aldrich, Singapore) on the back of the neck. Control rats experienced the same surgical implantation with empty SILASTIC brand silicon tubes. Based on previous work (12), the release rate of 17?-estradiol from the implants was documented to be 2.4 μg/cm·d. TAM (Sigma-Aldrich) was dissolved in castor oil at a concentration of 10 mg/ml. Female rats daily received either 2 or 5 mg TAM/kg BW via sc injections. Preformulated ICI 182,780 (AstraZeneca Pharmaceuticals, Macclesfield, UK) was supplied at a concentration of 50 mg/ml in castor oil solution. Female rats were received weekly sc injections of either castor oil alone or three different doses (1, 1.5, or 2 mg) of ICI 182,780/kg BW for 3 wk. At the end of the experiments, animals were killed by carbon dioxide exposure. The uteri were excised, trimmed, weighed, and snap-frozen in liquid nitrogen and stored at –70 C for RNA extraction. Part of the uterine tissue was fixed in neutral buffer containing 10% formalin and embedded in paraffin for immunohistochemistry. To study the effects of ovarian hormones on ps20 gene expression, female rats were OVX, and the uteri were collected at various times after ovariectomy for ps20 mRNA determination. To study the effects of estradiol and progesterone on ps20 gene expression, OVX rats were implanted with 1.5-cm SILASTIC brand silicon tubes containing either 17?-estradiol or progesterone, and the uterine tissue was harvested at the indicated times after hormonal supplementation.
ps20 Antibody
Rabbit antirat ps20 antibodies were raised against amino acid peptides in the N terminal of rat ps20. The amino acid sequences of rat ps20-specific peptide were as follows: TWEAMLPVRLAEKSQAEEVA. Affinity-purified ps20 antibody was diluted in Tris-buffered saline (20 mM Tris and 200 mM NaCl, pH 7.6) containing 0.1% Tween 20 at a final concentration of 1 μg/ml. The specificity of ps20 antiserum was verified by Western blot analysis of conditioned medium (CM)-derived MCF-7 cells that were transfected with full-length rat ps20 cDNA.
Immunohistochemical analysis and assessment
For immunohistochemistry, sections (5 μm thick) were cut, dewaxed in xylene, and then rehydrated as previously described (13). Antigen retrieval was performed by boiling the slides in 10 mM citrate buffer, pH 6.0, for 20 min. After blocking unspecific binding with 5% skim milk, sections were incubated with rabbit antirat ps20 (1:1000 dilution) overnight at 4 C as previously described (14). Immunohistochemistry was performed using the streptavidin-biotin peroxidase complex method according to the manufacturer’s instructions (Lab Vision, Fremont, CA) using 3,3'-diaminobenzidine as the chromogen. Sections known to stain positively were incubated in each batch, and negative controls were also prepared by replacing primary antibody with preimmune serum.
Primary uterine cell isolation and treatment
Uterine tissue derived from 10-wk-old rats was finely minced and washed three times with serum-free MEM (SRF). The minced tissue was incubated with MEM containing 5% FBS and 5 mg/ml collagenase A (Roche, Indianapolis, IN) at 37 C for 12 h as described previously (10). Cells were harvested by centrifuging at 800 x g for 10 min. The cell pellets were washed three times with SRF medium and allowed to grow in MEM containing 10% FBS. To study the effects of ps20 on the proliferation of PC-3 human prostate cancer cells and primary rat uterine cells, cells were plated at a density of 1 x 104 cells/well in 96-well plates and allowed to grow in growth medium for 24 h. Cells were washed once with SRF medium, then incubated in SRF medium for another 18 h. To determine the effects of bacterial recombinant ps20, cells were then treated with indicated concentrations of bacterial recombinant ps20. To study the effects of ps20 produced by MCF-7 cells on cellular proliferation, PC-3 cells were treated with SRF medium containing either 30% CM of ps20-transfected or 30% CM of pcDNA-3-transfected MCF-7 cells for 48 h. [3H]Thymidine incorporation was determined 48 h after treatment as previously described (15). Experiments were repeated three times, and the data were expressed as the mean ± SD.
Western blot analysis
To determine whether the different forms of secreted ps20 were due to differential glycosylation, concentrated CM derived from pcDNA-3-transfected MCF-7 cells was incubated with 10 U recombinant N-glycanase enzyme (Genzyme, Cambridge, MA) as described by the manufacturer. N-Glycanase-digested proteins were subjected to Western blot analysis as previously described (13). To examine the effects of estrogens, TAM, and ICI 182,780 on ps20 secretion, primary rat uterine cells were plated at a density of 5 x 106 cells/100-mm dish in the growth medium. After 24 h, the cell monolayer was washed, then treated with SRF medium containing 0.05% ethanol, 10–9 M ICI 182,780, 10–8 M TAM, and 10–10 M 17?-estradiol for 48 h. After treatment, CM was harvested. To determine changes in the levels of ps20 in CM, 1 ml CM was concentrated to 0.1 ml and subjected to Western blot analysis as previously described (13). Blots were incubated with rabbit antirat antibodies (1:4000) and horseradish peroxidase-conjugated donkey antimouse or antirabbit secondary antibodies (1:7500). Blots were then visualized with a chemiluminescent detection system (Amersham Biosciences) as described by the manufacturer.
Northern blotting
Total RNA was isolated from indicated tissues of female rats as previously described (16). Northern blots were performed on total RNA, and blots were hybridized with rat ps20 or human GAPDH (American Type Culture Collection, Manassas, VA) cDNAs as previously described (9). mRNA levels were determined by densitometric scanning of autoradiographs.
Stable ps20-transfected MCF-7 cells
The entire coding region of rat ps20 cDNA was cloned into the mammalian expression vector pcDNA3.1 (Invitrogen Life Technologies, Inc.) to create the ps20-pcDNA3.1 expression vector. The ps20-pcDNA3.1 sequence was confirmed by sequencing. MCF-7 cells were seeded at a density of 2 x 105 in 100-mm culture dishes in 90% -MEM (Sigma-Aldrich) containing 10% FBS with 1% penicillin -streptomycin for 24 h before transfection. Cells were transfected with either 5 μg ps20-pDNA3.1 or pDNA3.1 and 28 μl Lipofectamine reagent (Invitrogen Life Technologies, Inc.) as previously described (14). Forty-eight hours after transfection, cells were subcultured at 1:10 and replaced with -MEM containing 10% FBS and 800 μg/ml G418 (Calbiochem, La Jolla, CA). After 4 wk of selection, individual clones were isolated, expanded, and assayed for ps20 expression by Western blot analysis.
Expression and purification of ps20-His
The coding region of rat ps20 cDNA without signal peptide was cloned into pQE-30 bacterial expression vector (Qiagen, Mississauga, Canada), a vector that allows incorporation of a His-Tag at the carboxyl terminus. The ps20-QE-containing clone was confirmed by sequencing and transformed into the M15 bacterial expression cell line (Qiagen). Five hundred milliliters of bacterial cultures were grown and induced with 1 mM isopropyl-1-thio-?-D-galactopyranoside for 2 h. Soluble extract was generated by suspending bacteria in buffer containing 8 M urea, 0.1 M NaH2PO4, and 0.01 M Tris-HCl, pH 8.0, and kept at room temperature for 30 min. The soluble extract was centrifuged at 13,000 rpm for 30 min. Recombinant ps20-His protein was purified on a nickel-charged resin (Qiagen) as described by the manufacturer. Fractions containing proteins were pooled and dialyzed against 1 M acetic acid (pH 2.5) overnight at 4 C. The dialyzed samples were quickly frozen, lyophilized, and stored at –80 C.
Statistical analysis
For quantitation analysis, the sum of the density of bands corresponding to mRNA hybridized to ps20 probe or protein blotting with the antibody under studied was calculated and normalized to the amount of GAPDH mRNA or -tubulin, respectively. Differences in ps20 mRNA levels, [3H]thymidine incorporation, and the levels of ps20 were analyzed by ANOVA.
Results
Using differential display methodology, several differentially displayed bands representing cDNA corresponding to genes whose expression was inhibited by TAM were isolated. One of the bands, which was highly expressed in the uterine tissue of OVX rats, but significantly reduced in uterine tissue derived from OVX and TAM-treated rats (Fig. 1A), was selected for additional characterization. This 270-bp fragment was reamplified by PCR using the same primers and conditions as differential display and was used to probe Northern blots of mRNA obtained from uterine tissue of OVX and OVX TAM-treated rats. A strong signal corresponding to approximately 1.2 kb emerged in RNA isolated from OVX vehicle-treated uterine tissue, whereas faint signal was observed in the RNA derived from uterine tissue of OVX TAM-treated rats (Fig. 1B). The data confirmed that the isolated gene was down-regulated by TAM. This 270-bp fragment was then subcloned into TA vector and subjected to nucleotide sequence analysis. Comparison of the nucleotide sequence obtained against the nonredundant nucleotide database of GenBank established that this cDNA was a 100% match with rat ps20 cDNA (11).
FIG. 1. Differentially expressed mRNA bands, ps20 expression in uterine tissue of OVX vehicle-treated and OVX TAM-treated rats, and ps20 mRNA levels in various tissues of ovary-intact female rats. A, Total RNA isolated from uterine tissue of OVX vehicle-treated and OVX TAM-treated rats was subjected to differential display. The arrowhead indicates mRNA reduced by TAM treatment. B, Northern blot analysis of total RNA from uterine tissue of OVX vehicle-treated and OVX TAM-treated rats was used to confirm the presence of a differentially expressed mRNA in OVX TAM-treated uterine tissue. Blots were hybridized with a 32P-labeled DNA fragment that was isolated from the above differential gel or glyceraldehyde-3-phosphate dehydrogenase cDNA. C, Total RNA derived from various tissues of an 80-d-old female rat was subjected to Northern blot analysis. Blots were hybridized with rat ps20 cDNA. The ethidium bromide staining of 28S and 18S ribosomal RNA was used as a control loading. Tissues are: Mg, mammary gland; Ut, uterus; Fa, abdominal fat; Ov, ovary; Lu, lung; He, heart; Ce, cerebellum; Br, brain; Int, small intestine; μ, red muscle; Li, liver; Ki, kidney; Spl, spleen; Bl, bladder; Pla, placenta.
To determine ps20 expression in rat tissues, total RNA derived from various tissues of mature female rats was analyzed by Northern blotting. Figure 1C shows that ps20 transcripts of approximately 1.2 kb were observed in all tissues examined, with the highest levels detected in lung, heart, bladder, cerebellum, ovary, uterus, and placenta. In addition, ps20 mRNA was found in rat prostate (data not shown).
To study the effects of estrogens and TAM on the uterine growth and expression of ps20 in the uterus, ovary-intact rats were treated with various doses of TAM, estradiol, and the antiestrogen ICI 182,780. Figure 2A shows that after estradiol implantation for 3 wk, a dose-dependent decrease in ps20 expression was observed (P < 0.01). Similarly, at doses of 2 and 5 mg/kg body weight, TAM significantly reduced ps20 mRNA by 45% and 50%, respectively (Fig. 2B; P < 0.01). ICI 182,780, in contrast, significantly increased ps20 mRNA levels in uterine tissue compared with the controls (Fig. 2C; P < 0.01). Maximal induction was observed at a dose of 1 mg ICI 182,780/kg body weight/wk (Fig. 2C). The results indicate that TAM acts as an estrogen agonist to inhibit ps20 expression. In all cases, an inverse relationship between ps20 gene expression and uterine weight existed.
FIG. 2. Effects of estradiol, TAM, and ICI 182,780 on ps20 gene expression. Ovary-intact rats were treated with the indicated doses of 17?-estradiol by implantation (A) or were injected with two different doses of TAM (B) or the three indicated doses of ICI 182,780 (C) for 3 wk as described in Materials and Methods. Total RNA derived from uterine tissue was subjected to Northern blot analysis. Blots were hybridized with glyceraldehyde-3-phosphate dehydrogenase and rat ps20 cDNAs. Densitometric scanning of the ps20 band is shown. Data are expressed as the mean ± SE. Bars with different letters are significant at P < 0.01.
To investigate whether ps20 expression in the uterine tissue was estrogen dependent, ps20 mRNA levels were determined at different times after ovariectomy. As shown in Fig. 3A, 6 h after ovariectomy the levels of ps20 mRNA were increased by 4-fold and reached maximum levels by 48 h. Treatment of OVX rats with 17?-estradiol resulted in a down-regulation of ps20 expression (Fig. 3B). Progesterone, a hormone that has little or no effect on OVX-induced uterine involution, had very little or no effect on OVX-induced ps20 gene expression (data not shown).
FIG. 3. Effects of ovariectomy and estradiol on ps20 expression. A, Ovary-intact rats were OVX, and uterine tissue was collected at the indicated times after ovariectomy as described in Materials and Methods. B, OVX rats were implanted with SILASTIC brand silicon tubes containing 17?-estradiol. The rate of estradiol released was approximately 3.6 μg/d. Total RNA derived from uterine tissue was subjected to Northern blot analysis. Blots were hybridized with glyceraldehyde-3-phosphate dehydrogenase and rat ps20 cDNAs. Densitometric scanning of the ps20 band is shown. Data are expressed as the mean ± SE.
FIG. 6. Immunolocalization of ps20 protein in rat uterine tissue. Rat uterine tissue from vehicle-, estradiol-, TAM-, and ICI 182,780-treated rats was collected, and paraffin blocks were prepared. Uterine tissue sections were stained with rabbit antirat ps20 antibody as described in Materials and Methods. In vehicle-treated uterine tissue (A), ps20 protein was detected in smooth muscle cells as well as in luminal epithelial cells. The immuno-signal was significantly reduced in estradiol-treated (B) and TAM-treated (C) cells of uterine tissue. In ICI 182,780-treated uterus (D), ps20 protein was also observed on the apical surface of glandular and luminal epithelial cells. Original magnification, x400.
Because ps20 gene expression was regulated by estrogens and TAM, we wanted to determine whether ps20 gene expression fluctuated during different stages of estrous cycle. Total RNA derived from uterine tissue of rats at different stages of the cycle was analyzed by Northern blotting. As shown in Fig. 4, minimal expression of ps20 occurred at proestrous and estrous stages coincident with estrogen-induced uterine cell proliferation. High levels of ps20 mRNA were detected in uterine tissue during metestrous and diestrous.
FIG. 4. Expression of ps20 gene in rat uterus at different stages of the estrous cycle. Total RNA derived from uterine tissue of rats at different stages of the estrous cycle was analyzed by Northern blotting. Blots were hybridized with either glyceraldehyde-3-phosphate dehydrogenase or ps20 cDNAs. A, Each lane represents one phase of the estrous cycle. Densitometric scanning of the ps20 band is shown in B. Data are expressed as the mean ± SE. Bars with different letters are significant at P < 0.01.
To gain additional information about the function of ps20 in uterine tissue, we developed the rabbit polyclonal antibody against rat recombinant ps20 protein. CM from MCF-7 cells transfected with rat ps20 cDNA was used to determine the antibody specificity. Western blot analysis revealed that three specific bands of apparent molecular mass 24, 27, and 29 kDa were detected in CM derived from ps20-transfected, but not pc-DNA-transfected MCF-7 cells, confirming that the antibody was specific (Fig. 5A). A similar pattern was observed in CM derived from primary rat uterine cells (Fig. 5B). Treatment of CM with N-glycanase converted the 27- and 29-kDa species to the 24-kDa form (Fig. 5B), suggesting that ps20 is glycosylated.
FIG. 5. Determination of the anti-ps20 antibody specificity, glycosylation status of ps20, and ps20 growth inhibitory activity. Rabbit polyclonal antibody against rat ps20 was developed as described in Materials and Methods. CM collected from mock-transfected and ps20-transfected MCF-7 cells was concentrated and subjected to Western blot analysis as described in Materials and Methods. Note that the antibody recognized proteins of approximately 24, 27, and 29 kDa in CM derived from ps20-transfected, but not mock-transfected, MCF-7 cells (A). CM derived from ps20-transfected MCF-7 cells was incubated with (+) and without (–) N-glycanase (B). Note that the 24-kDa band became prominent when ps20 protein was digested with N-glycanase. C, Purified bacterial recombinant ps20. D, Primary rat uterine cells and PC-3 cells were grown and treated with SRF containing the indicated doses of bacterial recombinant ps20 for 48 h as described in Materials and Methods. E, PC-3 cells were treated with 30% CM derived from pcDNA or ps20-transfected MCF-7 cells for 48 h. [3H]Thymidine incorporation was determined as described in Materials and Methods. Experiments were performed in quadruplicate, with the results reflecting the mean ± SE. Bars with different letters are significantly different from one another at P < 0.01. The experiments were repeated three times with similar results.
It has been reported that recombinant ps20 inhibited PC-3 prostate cancer cell proliferation (11). To determine whether ps20 had any growth inhibitory activity on rat uterine cells, cells were treated with various doses of bacterial recombinant ps20. PC-3 cells were used as controls. Unlike previous studies (11), which showed that bacterial recombinant ps20 inhibited PC-3 cells in vitro, we observed that bacterial recombinant ps20 had no effect on the growth of either primary uterine cells or PC-3 cells (Fig. 5D). Our data raised the possibility that posttranslational modification of ps20 protein may play a role in growth regulation. To investigate this possibility, PC-3 cells were treated with serum-free medium containing either 30% CM obtained from stably ps20-transfected MCF-7 cells or 30% CM derived from pcDNA-transfected MCF-7 cells. Figure 5E showed that [3H]thymidine incorporation was significantly lower in cells treated with CM derived from ps20-transfected MCF-7 cells compared with cells grown in CM-derived, mock-transfected MCF-7 cells (P < 0.01). These observations raise the possibility that posttranslational modification of ps20 protein may play a role in its growth inhibitory activity.
To determine the cell type-specific expression of ps20 in uterine tissue, immunohistochemistry was performed on sections of uteri derived from vehicle-, ICI 182,780-, estradiol-, and TAM-treated rats using antirat ps20 antibody. High ps20 expression was detected in smooth muscle cells. Faint signal was also detected in luminal epithelial cells and the glandular population of the uteri (Fig. 6A). Treatment of rats with 17?-estradiol (Fig. 6B) and TAM (Fig. 6C) resulted in a significant reduction in ps20 protein in smooth muscle and luminal epithelial cells. ps20 expression in the smooth muscle cells and particularly in luminal epithelial cells was greatly increased by ICI 182,780 treatment (Fig. 6D). In ICI 182,780-treated uterine tissue, a strong immunostaining was also detected on the apical surface of glandular and luminal epithelial cells (Fig. 6D). Similar staining patterns were seen when the sections derived from uterine tissue of OVX rats were stained with ps20 antibody (data not shown). No staining was observed in smooth muscle cells when the antibody was preadsorbed with ps20 peptide (data not shown).
To determine whether estrogens also regulated ps20 secretion, primary rat uterine cells were isolated, grown, and treated with SRF medium containing vehicle, 10–9 M ICI 182,780, 10–8 M TAM, and 10–10 M 17?-estradiol for 48 h. Figure 7 shows that although ICI 182,780 caused a 2-fold increase in ps20 accumulation in CM, exposure of cells to TAM and estradiol resulted in 50% and 90% reductions in basal ps20 secretion, respectively.
FIG. 7. Effects of estrogens, TAM, and ICI 182,780 on ps20 protein secretion by primary rat uterine cells. Primary rat uterine cells were isolated, grown, and treated with SRF containing vehicle (C), 10–9 M ICI 182,780 (ICI), 10–8 M TAM, and 10–10 M 17?-estradiol (E2) for 48 h as described in Materials and Methods. CM was harvested for Western blot analysis as described in Materials and Methods. Blots were incubated with antirat ps20 antibody. Representative blots are shown. Experiments were repeated three times with similar results.
Discussion
Long-term administration of TAM has been reported to be associated with increased risk of endometrial cancer in postmenopausal women (7). However, the molecular mechanisms underlying this effect are not well understood. In the present study we isolate an estradiol- and TAM-regulated cDNA, ps20, using differential display and cDNA library screening. TAM has been known to regulate the expression of a series of estrogen-regulated genes, including the progesterone receptor, epidermal growth factor, TGF-, TGF-?, IGF-I receptor, epidermal growth factor receptor, and pS2 (reviewed in Ref. 4), IGF-I (8), UO-44 (9), and IGFBP-3 (10) in uterine tissue. In the present study we demonstrate for the first time that estrogens and TAM inhibit ps20 expression in this tissue. Levels of ps20 transcripts are high in heart, lung, and tissues containing a high proportion of smooth muscle cells, but are low in tissues with more epithelial cells. The localization of ps20 in smooth muscle as well as luminal epithelial cells in uterine tissue suggests that ps20 may play an autocrine and/or paracrine role in the regulation of cellular proliferation and the maintenance of homeostasis in this tissue. The high expression of ps20 found in uterine smooth muscle is consistent a previous study (11) showing that in the prostate, smooth muscle cells express high levels of ps20.
Low expression of ps20 occurred at the estrous and proestrous stages of the ovarian cycle coincident with estrogen-induced uterine cell proliferation, in contrast to high levels during metestrus and diestrus. In rodents, circulating levels of estrogens are high in proestrus and low in diestrus. Concomitant with this rise and fall in circulating estrogen levels, there is a decline and rise in ps20 mRNA levels in uterine tissue. This pattern of expression during the reproductive cycle suggests that ps20 plays a role in growth regulation. A rapid increase in ps20 gene expression after ovariectomy and antiestrogen ICI 182,780 treatment also supports the inhibitory effects of estrogens on ps20 expression.
The precise mechanism(s) by which TAM inhibits ps20 gene expression is not known at present. TAM is a classical partial agonist and exhibits tissues specificity for inducing either an agonist or antagonist response. In rats and humans, TAM exhibits partial agonist (reviewed in Ref. 4). It produces antagonist effects in breast, but agonist effects in vagina and endometrium (reviewed in Ref. 4). Long-term TAM use is generally associated with a reduced incidence of contralateral breast cancer (antagonist), a reduced incidence of primary breast cancer in high risk women (antagonist), maintenance of bone density (agonist), and an increased risk of endometrial carcinomas (agonist) (17). ICI 182,780 seems to be an antagonist; it is devoid of agonist activity in the uterus or vagina of rats (18). Although TAM and estradiol suppress ps20 expression in uterine tissue, TAM is approximately 800- to 1500-fold less potent than estradiol. The decrease in ps20 expression after TAM treatment suggests that TAM acts as an estrogen agonist in the uterus. The results reported here are clinically relevant, because women treated with TAM frequently exhibit endometrial hyperplasia (19) and rarely show neoplasia (7, 20, 21). TAM has been proposed as a treatment for neoplastic conditions of the uterus (22), but the results of clinical trials have not been impressive, and there are clinical and laboratory data suggesting that stimulation of endometrial neoplastic growth and leiomyoma growth by TAM are possible (23, 24). It is possible that the estrogen-like effect of TAM on ps20 expression in uterine tissue, as described here, is related to its adverse effects.
An immunohistochemical study reveals that ps20 protein expression is not only confined to smooth muscle cells, but is also present in luminal and glandular epithelial cells of the uterus. This observation is consistent with the recent report (25) that ps20 is localized in stromal cells as well as in a more aggressive epithelial phenotype. High levels of ps20 gene expression are found in smooth muscle cells and the apical surface of luminal epithelial cells of ICI 182,780-treated uterine tissue. Although ps20 expression is inversely correlated with uterine growth, the biological function of ps20 in the uterus is as yet unclear. ps20 Belongs to the whey acidic protein-type, four-disulfide core domain protein family (11). They are serine protease inhibitors that exhibit a variety of growth and differentiation functions. Recently, McAlhany et al. (26) proposed that ps20 is a TGF-?1-induced regulator of angiogenesis. They suggest that ps20 functions by either promoting endothelial cell migration or contributing to pericyte stabilization of newly formed vascular structure (26). In the present study we observed that estrogens, which are known to induce angiogenesis, potently inhibit ps20 expression. Ovariectomy and ICI 182,780, in contrast, are potent stimulators of ps20 expression and inhibitors of angiogenesis. Thus, in the uterus, ps20 may function as a local growth regulator, rather than a regulator of angiogenesis.
Previous studies have shown that recombinant ps20 inhibited the proliferation of prostate carcinoma PC-3 cells (11) and was suggested to function as a mediator of local growth and differentiation mechanisms. In the present study we did not observe any growth inhibition when bacterial recombinant ps20 was used to treat the same cell line. The discrepancy between our study and the previous one (11) is not clear. We believe that posttranslational modification of the ps20 protein plays an important role in determining its growth inhibitory activity. This hypothesis is supported by the observations that three specific bands of apparent molecular mass 24, 27, and 29 kDa are detected in CM of ps20-transfected MCF-7 cells. Similar results are observed when CM from rat primary uterine cells is analyzed by Western blot analysis. However, the 24-kDa form of ps20 becomes the major species when CM is treated with N-glycanase. This suggests that ps20 undergoes posttranslational modification before secretion.
In summary, our data demonstrate that ps20 gene expression in uterine tissue is a molecular marker that inversely correlates well with the positive or negative uterotropic effects of estrogen receptor antagonists and partial agonists. The characterization of ps20 protein provides new knowledge about the roles of ps20 protein in estradiol- and TAM-induced cellular proliferation and cancer in this tissue.
References
Jensen EV, DeSombre ER 1972 Mechanism of action of the female sex hormones. Annu Rev Biochem 41:203–230
Anderson JN, Peck Jr EJ, Clark JH 1975 Estrogen-induced uterine responses and growth: relationship to receptor estrogen binding by uterine nuclei. Endocrinology 96:160–167
Katzenellenbogen BS, Gorski J 1975 Biochemical actions of hormones. New York: Academic Press
Clarke R, Leonessa F, Welch JN, Skaar TC 2001 Cellular and molecular pharmacology of antiestrogen action and resistance. Pharmacol Rev 53:25–71
Jordan VC, Morrow M 1999 Tamoxifen, raloxifene, and the prevention of breast cancer. Endocr Rev 20:253–278
Powles TJ, Chang J 1997 Chemoprevention of breast cancer: why is tamoxifen not the answer? Endocr Relat Cancer 4:135–139
Fornander T, Cedermark B, Mattsson S, Skoog L, Theve T, Askergren J, Rutqvist LE, Glas U, Silfversward C, Somell A, Wilking N, Hjalmar ML 1989 Adjuvant tamoxifen in early breast cancer: occurrence of new primary cancers. Lancet 117–120
Huynh H, Pollak M 1993 IGF-1 gene expression in the uterus is stimulated by tamoxifen and inhibited by the pure antiestrogen ICI 182780. Cancer Res 53:5585–5588
Huynh H, Ng CY, Lim KB, Ong CK, Ong CS, Tran E, Tuyen Nguyen TT, Chan TW 2001 Induction of UO-44 gene expression by tamoxifen in the rat uterus and ovary. Endocrinology 142:2985–2995
Huynh H, Pollak M 1994 Uterotrophic actions of estradiol and tamoxifen are associated with inhibition of uterine IGF binding protein 3 gene expression. Cancer Res 54:3115–3119
Larsen M, Ressler SJ, Lu B, Gerdes MJ, McBride L, Dang TD, Rowley DR 1998 Molecular cloning and expression of ps20 growth inhibitor: a novel WAP-type "four-disulfide core" domain protein expressed in smooth muscle. J Biol Chem 273:4574–4584
Robaire B, Ewing LL, Irby DC, Desjardins C 1979 Interaction of testosterone and estradiol-17? on the reproductive tract of the male rat. Biol Reprod 21:455–463
Huynh H, Chow PK, Ooi LL, Soo KC 2002 A possible role for insulin-like growth factor-binding protein-3 autocrine/paracrine loops in controlling hepatocellular carcinoma cell proliferation. Cell Growth Differ 13:115–122
Ong CK, Ng CY, Leong C, Ng CP, Foo KT, Tan PH, Huynh H 2004 Genomic structure of human OKL38 gene and its differential expression in kidney carcinogenesis. J Biol Chem 279:743–754
Huynh H, Yang X, Pollak M 1996 Estradiol and antiestrogens regulate a growth inhibitory insulin-like growth factor binding protein 3 autocrine loop in human breast cancer cells. J Biol Chem 271:1016–1021
Huynh H, Tetenes E, Wallace L, Pollak M 1993 In vivo inhibition of insulin-like growth factor-I gene expression by tamoxifen. Cancer Res 53:1727–1730
Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, Vogel V, Robidoux A, Dimitrov N, Atkins J, Daly M, Wieand S, Tan-Chiu E, Ford L, Wolmark N 1998 Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst 90:1371–1388
Wakeling AE, Bowler J 1988 Novel antioestrogens without partial agonist activity. J Steroid Biochem 31:645–653
Fornander T, Rutqvist LE, Wilking N 1991 Effects of tamoxifen on the female genital tract. Ann NY Acad Sci 622:469–476
Andersson M, Storm HH, Mouridser HT 1991 Incidence of new primary cancers after adjuvant tamoxifen therapy and radiotherapy for early breast cancer. J Natl Cancer Inst 83:1013–1017
Magriples U, Naftolin F, Schwartz PE, Carcangiu ML 1993 High-grade endometrial carcinoma in tamoxifen-treated breast cancer patients. J Clin Oncol 11:485–490
Swenerton KD 1980 Treatment of advanced endometrial adenocarcinoma with tamoxifen. Cancer Treat Rep 64:805–811
Jordan VC, Gottardis MM, Satyaswaroop PG 1991 Tamoxifen-stimulated growth of human endometrial carcinoma. Ann NY Acad Sci 622:439–446
Dilts PV, Hopkins MP, Chang AE, Cody RL 1992 Rapid growth of leiomyoma in patient receiving tamoxifen. Am J Obstet Gynecol 166:167–168
McAlhany SJ, Ayala GE, Frolov A, Ressler SJ, Wheeler TM, Watson JE, Collins C, Rowley DR 2004 Decreased stromal expression and increased epithelial expression of WFDC1/ps20 in prostate cancer is associated with reduced recurrence-free survival. Prostate 61:182–191
McAlhany SJ, Ressler SJ, Larsen M, Tuxhorn JA, Yang F, Dang TD, Rowley DR 2003 Promotion of angiogenesis by ps20 in the differential reactive stroma prostate cancer xenograft model. Cancer Res 63:5859–5865(Huynh Hung)