- Email: [email protected]
Regulation of endometrial adenocarcinoma cell proliferation by Activin-A and its modulation by 17β-estradiol
Molecular and Cellular Endocrinology 192 (2002) 187 /195 www.elsevier.com/locate/mce
Regulation of endometrial adenocarcinoma cell proliferation by Activin-A and its modulation by 17b-estradiol Nicoletta Di Simone a, Alan L. Schneyer b, Dario Caliandro a, Roberta Castellani a, Alessandro Caruso a,* b
a Department of Obstetrics and Gynecology, Universita’ Cattolica del S. Cuore, Largo Gemelli 8, 00168 Rome, Italy National Center for Infertility Research and the Reproductive Endocrine Sciences Center, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA
Received 12 March 2001; accepted 27 August 2001
Abstract A role for activins in regulating cellular transformation is suggested by the a-inhibin knockout mouse in which development of gonadal tumors is associated with elevated activin levels. It was the purpose of the current study to determine whether activin had similar actions on endometrial cell lines, specifically on a well differentiated estrogen-responsive endometrial adenocarcinoma cell line (ISH) and estrogen-unresponsive cells (HEC-50) obtained from a poorly differentiated endometrial adenocarcinoma. Activin was secreted by both adenocarcinoma cell lines. Using reverse transcription-PCR, messenger RNA type I and type II activin receptor subtypes were detected in both cell lines: expression of IB and IIB was approximately three- to fourfold greater in ISH cells than in HEC-50 cells, while activin receptor IA and IIA messenger RNA levels were approximately equal in both cell lines. Activin treatment (30 /300 ng/ml) caused a dose- and time-dependent inhibition of ISH cells proliferation and resulted in a significant decrease in Bcl-2 protein and mRNA levels. No difference was observed in Bax expression. There was no significant effect of activin when the cultures of ISH cells were exposed to 17b-estradiol. In contrast, activin showed a weak, but significant, mitogenic effect on HEC-50 cells without modifications in Bax and Bcl-2 mRNA and protein levels. The results demonstrate that activin is a regulator of endometrial cancer cell growth. 17b-Estradiol may promote resistance of estrogen-responsive endometrial cancer cells to the growth-retarding effects of activin and one of the mechanisms might be a down-regulation of the activin receptors. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Activin; Cancer; Endometrium
1. Introduction Activins are members of the transforming growth factor-b (TGF-b) superfamily and are involved with regulation of proliferation, differentiation and/or disorders of growth regulation such as oncogenesis (Ying and Zhang, 1996; Dalkin et al., 1996; Di Simone et al., 1996, 1998; Welt et al., 1997; Vale et al., 1994). The potential involvement of inhibin and activin in cellular transformation is suggested by a-inhibin subunit-deficient mutant mice, in which nearly 100% of males and females develop gonadal and adrenal tumors in early adulthood (Matzuk et al., 1992, 1994). Whether * Corresponding author. Tel./fax: /39-6-3551-0031. E-mail address: [email protected] (A. Caruso).
this is the result of reduced inhibin or increased activin remains unknown. The ability of activin to enhance proliferation in certain cell lines has been reported, suggesting that activin can act as a growth promoter and perhaps play a role in tumor formation and proliferation (Di Simone et al., 1996; Shikone et al., 1994). In contrast, activin can also inhibit proliferation and induce apoptosis in LnCAP prostatic cancer cells (Dalkin et al., 1996; Wang et al., 1996b; McPherson et al., 1997), rat pituitary (Billestrup et al., 1990) and corticotroph cells (Bilezikjian et al., 1991). The molecular basis for whether activin is stimulatory or inhibitory is incompletely understood. However, it is likely to be multifaceted because the end effect of activin would be influenced by factors such as the local levels of
0303-7207/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 3 - 7 2 0 7 ( 0 1 ) 0 0 6 4 8 - 7
188
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
follistatin (FS; Sugino et al., 1993; Nakamura et al., 1991) and the level and type of activin receptor expression (DeJong et al., 1993). FS is a monomeric glycoprotein whose only known actions are mediated through nearly irreversible binding and neutralization of activin (Nakamura et al., 1990; Schneyer et al., 1994). Members of the TGF-b superfamily act at the cellular level through a complex of two related serine/threonine kinase receptor subfamilies known as type I and type II receptors (Mathews, 1995). Although type I/II receptor complexes may exist in the unligated state, complex formation is necessary for signal transduction, so that the relative number of each type of activin receptor on the cell surface at any moment may be a critical factor in determining the cell’s responsiveness to activin. Activin subunits and activin receptors were observed in human endometrium (Gu et al., 1995), suggesting that this tissue might represent a source as well as a target of activin, with actions similar to those summarized above for prostatic or gonadal tumors. Endometrial carcinoma is thought to exist in two principal forms. The first is responsive to estrogens, occurs in relatively young perimenopausal women and is characterized by predominantly low grade, minimally invasive tumors and usually a positive prognosis (Rose, 1996). On the other hand, estrogen-independent endometrial tumors occur in relatively older postmenopausal women and are typically high grade, invasive tumors having a progressive clinical course and a poor prognosis. The availability of estrogen responsive and unresponsive endometrial adenocarcinoma cell lines allows investigation of factors that regulate growth in endometrial tumors. Such cell lines include the Ishikawa (ISH) cell line (Nishida et al., 1985), which originated from a well differentiated endometrial adenocarcinoma and contains estrogen and progesterone receptors. This cell line is responsive to estrogens (Holinda et al., 1986). Estrogen-unresponsive HEC-50 cells (Kuramoto et al., 1976; Kassan et al., 1989) were established from the ascites of a patient with poorly differentiated endometrial cancer. Using these two distinct cell lines, the biosynthesis and actions of inhibin/activin subunits and activin receptors were compared. In addition, potential regulation of activin action by 17b-estradiol (E2) was investigated in the estrogen-responsive ISH cells.
poorly differentiated endometrial adenocarcinoma (Kuramoto et al., 1976). The two cell lines were generously provided by Dr De Cicco, Department of Obstetrics and Gynecology, Catholic University, Rome, Italy. Cells used for these studies were cultured for 7/10 days, with several changes of medium, in phenol red free DMEM containing 2% FBS (Life Technologies, Grand Island, NY) stripped of steroids by absorption to dextran-coated charcoal suspension (0.025% Dextran/ 0.25% charcoal wt/vol, DCC), 2 mM L-glutamine and antibiotics (100 IU/ml penicillin and 100 mg/ml streptomycin sulfate). Cultures were maintained at 37 8C in a humidified atmosphere of 95% air /5% carbon dioxide. Media were changed every 2 days and the cells were subcultured once a week by replating trypsinized cell suspensions. The cells were grown to near confluence. After 48 h, the conditioned medium was collected and stored at /20 8C until assayed for protein hormones. 2.2. Immunoassays
2. Material and methods
Total activin-A was measured using a solid phase sandwich enzyme-linked immunoadsorbent assay (Elisa; Serotec, Turin, Italy). Wells of the microplate were dry coated with monoclonal antibody specific for the b-Asubunit of activin. Samples were incubated for 3 min at 100 8C with an equal volume of 20% SDS. Hydrogen peroxide solution (30% v/v) was then added and the samples allowed to stand at room temperature for 30 min. Then the samples were added along with assay diluent to the coated microplate followed by biotinylated monoclonal detection antibody. The plate was covered and incubated overnight at room temperature on a plate shaker. The following day the plate was washed and streptavidin /alkaline phosphatase added for 1 h. After additional washes, p -nitrophenylphosphate in 0.05 M ethanolamine buffer (pH 9.8) was added and incubated at room temperature until color developed in control wells, at which time the assay was read in a microtiter plate reader (Biorad Platerreader Model 405). Total immunoreactive inhibin was assessed as previously described (Bernstein et al., 1990) using the Monash Inhibin RIA (Schneyer et al., 1990), radioiodinated inhibin-A (Genentech), and a human follicular fluid assay standard that was calibrated at 22 pg/ml using recombinant inhibin-A. FS was quantitated by two-site monoclonal antibody for free FS as previously described (Wang et al., 1996a).
2.1. Cells and culture conditions
2.3. RNA extraction
The experiments reported here were carried out in a well differentiated estrogen-responsive endometrial adenocarcinoma cell line (ISH; Nishida et al., 1985) and in estrogen-unresponsive cells, HEC-50, obtained from a
Confluent cells remaining after medium collection were treated with Tri Reagent (Molecular Research Center, Cincinnati, OH) according to the manufacturer’s instructions. RNA integrity was confirmed by
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
agarose gel electrophoresis and ethidium bromide staining as well as by monitoring absorbance at A260/280. The RNA concentration was determined in all samples in two independent determinations and at two dilutions before RT-PCR experiments. All samples were stored at /80 8C. 2.4. RT-PCR procedures Complementary DNA (cDNA) was obtained by reverse transcription at 42 8C for 45 min in a 20-ml reaction mixture containing 200 ng of RNA, first treated with Dnase (Life Technologies), 0.5 mM of each deoxyNTP (Promega, Life Science, Florence, Italy), 2.5 mM oligodT (Life Technologies), and 2.5 U SuperScript (Life Technologies). The PCR reaction was carried out in a volume of 25 ml containing 2.5 ml 10 / PCR buffer, 5 mM MgCl2, 0.2 mM of each deoxy-NTP, 0.5 mM of each specific primer, 0.25 ml Taq polymerase (Promega) and 0.05 mCi nucleotide triphosphate ([32P]dCTP; NEN, Dreieich, Germany) to which 1 ml of the RT reaction product was added. The sequences of the specific primers were described previously (Di Simone et al., 1996, 1998; Marone et al., 1998). Amplification was achieved using a thermal cycler (Perkin Elmer GeneAmp, PCR System 9600). The amplification profile involved preincubation at 95 8C for 3 min, denaturation at 94 8C for 0.5 min, primer annealing at 57 8C for 1.3 min and extension at 72 8C for 1.3 min. In initial experiments, cycle number and RNA concentrations were optimized for each target, so that signals were always in the exponential portion of the amplification curve (35 cycles for Bcl-2 and Bax; 32 cycles for a, b subunits and FS; 28 cycles for activin receptors and 22 cycles for b-actin). Three microliters of the PCR reaction were electrophoresed in 5% polyacrylamide gels in Tris /borate / EDTA buffer. The concentration of each messenger RNA was determined by counting the radioactivity of each target by Instant/Imager (Camberra/Packard S.r.l., Milan, Italy) and was expressed as counts per min. To monitor DNA contamination in the reagents or samples, RT was omitted from each reaction to exclude genomic or cDNA contamination. Reactions without RNA was used to monitor contamination of RT or the primers. All of these controls gave negative results. 2.5. Southern blot analyses Previously, we verified the b-actin, a, b subunits and FS PCR products identities by Southern blot (Di Simone et al., 1996), using synthetic oligos located internal to the two PCR primers. In the present study, activin receptors, Bcl-2 and Bax identities were demonstrated with the same method (data not shown). PCR-amplified products without
189
radionuclide were fractionated on a 1.5% agarose gel, visualized by UV transillumination and the product sizes were estimated by comparison to molecular weight markers (PGEM, Promega). The gels were then denatured in 50 mM NaOH /1.5 M NaCl for 30 min and neutralized in 1 M Tris /1.5 M NaCl. The DNA was transferred by capillary action to a nylon membrane (Biorad S.r.l., Milan, Italy) in 20 / SSC (3 M NaCl and 0.3 M sodium citrate, pH 7.0) and subjected to Southern blot analyses using [32P]dATP (NEN)-labeled oligonucleotide probes. 2.6. Semiquantitative analyses of activin receptors messenger RNA expression Semiquantitative analysis of messenger RNA (mRNA) expression was carried out as previously described for ovarian cancer cell lines (Di Simone et al., 1996). RNAs prepared from endometrial cells were diluted in water to concentrations ranging from 200 to 3 ng per sample. The cDNA was amplified for activin receptors type I (Ia and Ib) and type II (IIa and IIb). Three microliters of the PCR reactions were electrophoresed in 5% polyacrylamide gels in Tris /borate / EDTA buffer. To allow direct comparison of RNA concentrations between cell lines, PCR for each target was carried out in a single experiment and electrophoresed at the same time. As each target was amplified for each cell line in one PCR experiment under identical conditions, the resulting values were plotted against increasing concentrations of RNA, giving dose /response curves for each cell line that could then be compared for relative steady state mRNA levels. Each experiment was repeated three times and run on replicate gels. We observed no differences between the two gels. The results from one representative experiment are shown. 2.7. Proliferation studies: Methylene Blue assay We used a colorimetric technique (Oliver et al., 1989) to study cell growth. ISH and HEC-50 cells were grown in phenol red free DMEM supplemented with 2% FBS stripped of steroids in 96-well culture plates. The effects of E2 (Sigma, St Louis, MO) on ISH cell proliferation were evident at concentrations ranging from 10 9 to 10 7 M (data not shown). Then, we determined the kinetics of E2-stimulated cell growth. E2 (10 8 M) was added for 12, 24, 36, 48 and 72 h from a stock solution in ethanol. The final levels of ethanol in each well, including controls was 0.1%. After washing with 0.15 M saline, the cell layer was fixed by adding 100 ml of 10% formol saline to each well for at least 30 min. The fixative was shaken off each plate and 100 ml of filtered 1% (w/v) Methylene Blue in 0.01 M borate buffer (pH 8.5) was added to each well. After 30 min,
190
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
excess dye was removed. The absorbance at 650 nm (A 650) was measured from each well by a microplate photometer (Biorad Platerreader Model 405). In order to assess the accuracy of the assay, preliminary cultures were set up in replicate for a range of cell concentrations. The A650 was directly proportional to the cell number determined by haemocytometer count (data not shown; P B/0.01). Each experiment was repeated three times. Because ISH cell proliferation is estrogen-sensitive, the effects of activin (kindly provided by the National Hormone and Pituitary Program, Baltimore, USA) on E2 stimulated proliferation were also examined. Varying doses of activin (from 3 to 300 ng/ml) alone or in combination with E2 (10 8 M) were added to triplicate wells. Cells were treated for 48 h with the same drug concentrations and the medium was changed every 24 h. Results of cell growth assays are expressed as mean9/ S.E. of three different experiments. 2.8. Effect of activin on pro- and anti-apoptotic gene mRNA and proteins To examine the effect of activin on Bcl-2 and Bax mRNA and protein levels, the cells were treated on day 2 of culture with activin-A (3 /300 ng/ml) for 48 h in the presence or the absence of Estradiol (10 8 M). Cellular RNA was extracted for RT-PCR analyses and the results were expressed relative to b-actin for each time-point. Bcl-2 and Bax proteins were evaluated by cell ELISA (Del Papa et al., 1993; Xiao et al., 1997). On day 1 (24 h) and day 2 (48 h) of culture, the medium was removed and the cells were washed twice with DMEM and incubated for 60 min at room temperature with antihuman Bcl2 (10 mg/ml; clone 124; DAKO A/S, Glostrup, Denmark) and anti-human Bax (10 mg/ml; Santa Cruz Biotechnology, Santa Cruz, CA). After two more washes, the cells were fixed with paraformaldehyde for 15 min at room temperature. Cells were then washed twice and incubated for another 60 min at room temperature with 100 ml of alkaline phosphatase-conjugated affinity-purified anti-IgM or anti-IgG (Sigma/ Aldrich). After two washes with DMEM and two more with phosphate buffer, p -nitrophenylphosphate substrate was used to reveal IgG binding. The optical density (OD) values of the enzymatic reactions were read at 405 nm, after a 30-min incubation, with a microplate photometer (Platerreader, Bio-Rad, Milan, Italy). 2.9. Time-course study of the effect of estradiol on activin receptor mRNA levels ISH cells were passaged at dilution not greater than 1:5 after incubation for 5 min at 37 8C in 0.05% trypsin
and were maintained in culture for 72 h before testing the mRNA steady-state levels with or without E2 treatment. For time-course experiments, ISH cells were stimulated on day 2 of culture with E2 (10 8 M) for 12, 24 and 48 h. Cellular RNA was extracted for RT-PCR analyses with specific probes for activin receptors type I (Ia and I b) and type II (IIa and IIb). Results are the mean9/S.E. of three separate experiments, expressed relative to b-actin for each time point. 2.10. Statistical analyses Significant differences were determined using Student’s t -test, with P B/0.05 considered significant for all experiments.
3. Results 3.1. Steady-state messenger RNA levels Inhibin/activin subunit, FS and activin receptor mRNA expression was examined using RT-PCR. Amplification cycle number was varied to determine the exponential portion of the amplification curve for each target and the identity of the PCR products was verified by Southern blot either previously (inhibin/activin subunits and FS; Di Simone et al., 1996) or as demonstrated in Fig. 1 for activin receptors. Amplification of activin receptors required 28 cycles (Fig. 2) while for inhibin/ activin subunits and FS, 32 cycles were used. As indicated in Table 1, mRNAs encoding b-actin, activin b and bb subunits and FS were identified in both cell lines. A 823-bp inhibin a-subunit mRNA was detected in HEC-50 cells but was undetectable in the ISH cell line under identical conditions. 3.2. Activin receptors: semiquantitative analyses To compare mRNA levels for activin receptors in each cell line, the cells were harvested at approximately equal levels of confluence. After determination of RNA concentration, increasing amounts of RNA were reverse transcribed and amplified. Fig. 3 demonstrates that HEC-50 and ISH cells express mRNA for all four activin receptor subtypes: increasing RNA concentrations resulted in stronger PCR bands for each activin receptor subtype, creating a series of dose/response curves for each target cDNA. Of interest, expression of activin receptor mRNA IB and IIB was approximately three- to fourfold greater in ISH cells than in HEC-50 cells. On the other hand, activin receptor IA and IIA mRNA levels were approximately equal in both cell lines. Thus, ISH cells expressed higher IB/IA (1.039/0.14; P B/0.02) and IIB/IIA (2.429/
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
191
Fig. 1. Verification of activin receptors (Ia, Ib, IIa and IIb) by Southern blot. 32P-Labeled oligonucleotides for sequences internal to each pair of PCR primers were used to verify the identity of the PCR products. Ethidium bromide-stained agarose gels containing the PCR products were transferred to nitrocellulase and hybridized with labelled probe. Southern blot of the gels: specific recognition of the PCR products without detection of adjacent DNAs demonstrates the specificities of the PCR targets. Lane 1, activin receptor IIA; lane 2, activin receptor IIB; lane 3, activin receptor IA; lane 4, activin receptor IB.
mRNA extraction and expressed per 100 000 cells after 48 h of culture. Using a specific two-site activin-A immunoassay that measures total activin, HEC-50 and ISH cells secreted 119/1.5 and 49/0.3 ng/ml activin-A/100 000 cells per 48 h, respectively. No activin immunoactivity was detected in unconditioned medium. Inhibin and free FS were undetectable for both cell lines.
3.4. Cell proliferation studies
Fig. 2. Titration of PCR cycles using 0.2 mg total RNA from HEC-50 cells. Amplification is linear over the range of 23 /33 cycles for activin receptor IIa.
Table 1 Summary of comparative steady state mRNA levels determined from endometrial cell lines mRNA
HEC-50
ISH
a ba bb FS
/
, 100 /1000 cpm/min; , 1000 /10 000 cpm/min; , 10 000 cpm/min.
0.5; P B/0.01) activin receptor mRNA ratios compared to the HEC-50 cells (IB/IA 0.369/0.04 and IIB/IIA 0.299/0.03). 3.3. Hormonal secretion The ability of endometrial cell lines to secrete the hormones whose mRNA production was investigated by RT-PCR, was examined in the same cultures used for
The effect of E2 on the estrogen-responsive endometrial cancer cell line (ISH) was investigated. Linear growth was apparent from 12 h of culture but no difference in cell number was observed in the absence or the presence of E2 (10 8 M) until 48 h (Fig. 4). In contrast, there was no effect of estrogen on HEC-50 cells (data not shown). As both the cell lines expressed significant amounts of activin receptor mRNA, the effect of activin on cell growth was investigated. As shown in Fig. 5, activin caused a dose-dependent inhibition of ISH cell proliferation: a significant inhibition by activin on cell number was observed within 24 h of treatment for an activin dose of 300 ng/ml. This growth inhibition was more pronounced at 48 h with the lower dose (30 ng/ml) inhibiting cell proliferation. Because ISH cell proliferation is estrogen-sensitive, the effects of activin on estrogen-related proliferation were also examined. Interestingly, the anti-proliferative effect of activin on ISH cells was abolished in the presence of E2. In contrast to ISH cells, activin (30 ng/ml) significantly enhanced HEC-50 proliferation. Although proliferation of HEC-50 cells was enhanced by 30 ng/ml of activin, higher doses (300 ng/ml) did not significantly change cellular proliferation with respect to untreated cells. No effect of E2 on activin-induced HEC-50 cell proliferation was observed.
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
192
Fig. 3. The expression of activin receptor mRNA was quantitated by PCR as described in Section 2. One representative experiment is shown for each mRNA quantitation. RT reactions containing increasing amounts of RNA resulted in linearly increasing amounts of PCR products. Twenty-eight cycles of amplification were used for activin receptors. Analysis of each PCR product was carried out simultaneously for both cell lines. For each experiment, RT was omitted to exclude genomic or cDNA contamination; reactions without RNA were used to monitor contamination of RT or the primers (data not shown). All these controls gave negative results. ISH cells: I; 1, 2 and 3: 3, 50 and 200 ng/ml of RNA, respectively. HEC-50 cells: m; 4, 5 and 6: 3, 50 and 200 ng/ml of RNA, respectively.
ISH cells, the effect of activin was abolished in the presence of E2. The Bcl-2 protein was higher in HEC-50 than in ISH cells (7899/98 vs. 4159/54, P B/0.01). Treatment with activin had no effect on Bcl-2 protein in HEC-50 cells and on Bax protein for both cell lines (data not shown). In contrast, activin A (30 /300 ng/ml) induced a significant decrease in Bcl-2 protein of ISH cells (Fig. 6). The effect of activin was abolished in the presence of E2.
8
Fig. 4. Estradiol (10 M; I) induces proliferation of ISH cells. Cell proliferation was determined by a colorimetric technique (mean9/S.E. of triplicate culture wells). OD, optical density values. Detailed procedures are explained in Section 2. Each experiment was repeated three times. Significant differences from the control (ISH cells without E2; m) are indicated, *P B/0.02; **P B/0.01.
3.5. Regulation of pro- and anti-apoptotic Bax and Bcl-2 mRNAs and proteins The Bax and Bcl-2 were expressed at both messenger ribonucleic acid and protein levels in ISH and HEC-50 cells. The mRNA levels of Bax and Bcl-2 were investigated by RT-PCR analyses and the results were expressed relative to b-actin for each time point. No difference was observed in Bax mRNA between HEC-50 and ISH cells. The Bcl2 mRNA was higher in HEC-50 than in ISH cells (0.899/0.07 vs. 0.379/0.06, P B/0.01). As shown in Fig. 6, treatment with activin A (30, 300 ng/ml) induced a significant decrease in Bcl-2 mRNA in
3.6. Regulation of activin receptors mRNA levels by estradiol Because ISH proliferation is estrogen-sensitive and E2 eliminated activin’s antiproliferative effect, we analyzed the action of E2 treatment on activin receptor mRNA expression. As previously shown, ISH cells expressed transcripts for all four activin receptor subtypes. E2 (10 8 M) treatment significantly reduced mRNA levels for all four activin receptor subtypes, achieving suppression from 30 to 50% of mRNA levels at 24 and 48 h of incubation (Fig. 7). The levels of b-actin mRNA were unchanged during E2 treatment.
4. Discussion The present study demonstrates that activin subunit (bA and bB) mRNAs are expressed in endometrial cancer cell lines as determined by RT-PCR. In addition,
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
Fig. 5. Dose /response curve of activin-A on basal (I) and E2induced (10 8 M; m) endometrial cell proliferation. Significant differences from the untreated cells (0 ng/ml): *P B/0.02; **P B/0.01. After 48-h incubation, E2 (m) induces ISH cell proliferation (§P B/ 0.01). The concomitant treatment with activin-A (3 /300 ng/ml) does not significantly modify the E2-induced endometrial cell proliferation. Cell growth was determined by a colorimetric technique (mean9/S.E. of three separate experiments). OD, optical density values.
we observed that these cells express all four activin receptor subtypes. These data lend support to the hypothesis that locally synthesized cytokines, in particular members of the TGF-b superfamily, may influence endometrial function. In the ISH cell line, activin significantly inhibited cell proliferation and this inhibition was abolished in the presence of estrogen. Several studies have demonstrated how estrogen-dependent carcinomas regress after ovariectomy through a pathway involving programmed cell death or apoptosis, suggesting that estrogen may contribute to the survival of these cells (McGuire et al., 1975; Kyprianou et al., 1991). It was also recently reported that estrogen treatment induces Bcl-2 but not Bax mRNA in MCF-7 cells. This concept is further supported by the finding that estrogen may affect the efficacy of cancer treatments through modulation of Bcl-2 biosynthesis (Teixeira et al., 1995). The present study has demonstrated that mRNAs and proteins of Bax and Bcl-2 are expressed in endometrial adenocarcinoma cell lines. Activin-A resulted in a significant
193
Fig. 6. Effects of activin A on Bcl-2 mRNA (top) and protein (bottom) expression in the absence (no shading) or presence of estradiol (shading). The mRNA levels were normalized to b-actin. Protein levels were determined by ELISA. OD, optical density values at 405 nm. Significant differences from untreated cells (Activin-A 0 ng/ ml) are indicated: *P B/0.05; **P B/0.01. Each time point represents the mean9/S.E. of three independent experiments.
decrease in anti-apoptotic Bcl-2 protein level. Estradiol seems to promote resistance of estrogen-responsive endometrial cancer cells to the growth-retarding effects of activin and one mechanism of action seems to be the effect on Bcl-2 levels. The presence of effects of E2 on proliferation during the initial period of growth (48 h of culture) might be due to the experimental conditions of the present study. Although E2 (10 8 M) sustained ISH cell proliferation after about 10 days of culture (Holinda et al., 1986), under the chosen experimental condition (15% charcoaltreated fetal bovine serum renewed every 2 /3 days), Hamano et al. (1983) reported effects of E2, on growth rates of ISH cells within 2 days using less serum (1%) and lower initial cell number. In the estrogen-responsive endometrial cancer cell line (ISH), E2 treatment induced a state of resistance against activin’s growth inhibiting activity. Furthermore, when ISH cells were treated with E2, the mRNA levels for activin receptors decreased significantly, suggesting that the proliferative effect of estrogen in these cells could be mediated by a down-regulation of the activin signaling pathway. In contrast, activin-A stimulated HEC-50 cell growth. How the same growth factor, acting through the same specific receptors, is able to produce growth inhibition
194
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
The fact that HEC-50 cell growth can be stimulated by activin suggests that these cancer cells, through malignant transformation, have escaped from growth inhibition by the negative growth regulator thereby acquiring an altered responsiveness to this growth factor. In conclusion, the diversity of the observed effects is of interest, because activin is considered to be a potent growth regulator in most epithelial carcinoma cells (Ying and Zhang, 1996; Dalkin et al., 1996; Shikone et al., 1994), apparently acting as an autocrine regulator and with opposite effects in various cell types. Our results indicate that activin’s actions in endometrial tissue is susceptible to being subverted during oncogenesis. For this reason, elucidation of the regulatory pathways governing the expression and the actions of activin system is likely to be critical to our understanding of oncogenesis in a variety of tissues, including the endometrium.
References Fig. 7. Effects of incubation time in the absence (unshaded) or presence of E2 (10 8 M; shaded) on activin receptor mRNA expression. Levels of activin receptors Ia, Ib, IIa and IIb mRNA were significantly suppressed at 24 and 48 h (*P B/0.05; **P B/0.01) by E2. Each time or treatment point represents the mean9/S.E. of three independent experiments expressed relative to b-actin for each time point.
in some cell lines and stimulation in others is still unclear. Among the several possibilities that could be hypothesized is that activin type I and type II receptor heterogeneity underlies differential cellular signaling. Support for this possibility can be found in our observations: using a semiquantitative comparative RT-PCR assay, ISH cells show three- to fourfold higher expression of type IIB and IB activin receptor mRNA than HEC-50 cells. This could affect the number of IIA/ IA versus IIB/IB receptor complexes in different cell types, which may then produce different biological results. On the other hand, receptor function rather than receptor levels might be more important in carcinogenesis. Previously, TGF-b has been shown to be involved in the growth regulation of endometrial cells (Anzai et al., 1992): TGF-b inhibited ISH cell proliferation whereas it stimulated HEC-50 cell growth. Loss of sensitivity to growth inhibition by TGF-b is a phenomenon often observed in human epithelial tumor cells and is linked to malignant progression. Mutations of the activin type I receptor have been recently described by Su et al. (2001) in pancreatic carcinoma. These defects may impair the function of activin receptors with a secondary inactivation of the activin signals.
Anzai, Y., Gong, Y., Holinka, F.H., Murphy, L.J., Murphy, L.C., Kuramoto, H., Gurpide, E. 1992. Effects of transforming growth factors and regulation of their mRNA levels in two human endometrial adenocarcinoma cell lines. J. Steroid Biochem. Mol. Biol. 42, 449 /455. Bernstein, J.R., Crowley, W.F., Jr, Schneyer, A.L. 1990. An improved method of purifying inhibin radioligand for radioimmunoassay. Biol. Reprod. 43, 492 /496. Billestrup, N., Gonzalez-Marchon, C., Potter, E., Vale, W. 1990. Inhibition of somatotroph growth and growth hormone biosynthesis by activin in vitro. Mol. Endocrinol. 4, 356 /362. Bilezikjian, L.M., Blount, A.L., Campen, C.A., Gonzalez-Marchon, C., Vale, W. 1991. Activin-A inhibits proopimelanocortin messenger RNA accumulation and adrenocorticotropin secretion of AtT20 cells. Mol. Endocrinol. 5, 1389 /1395. Dalkin, A.C., Gilrain, J.T., Bradshaw, D., Myers, C.E. 1996. Activin inhibition of prostate cancer cell growth: selective actions on androgen-responsive LNCaP cells. Endocrinology 137, 5230 /5235. DeJong, F.H., de Winter, J.P., Wesseling, J.G., Timmerman, M.A., van Genesen, S., van den Eijnden-van Raaij, A.J., van Zoelen, E.J. 1993. Inhibin subunits, follistatin and activin receptors in the human teratocarcinoma cell line tera-2. Biochem. Biophys. Res. Commun. 192, 1334 /1339. Del Papa, N., Sheng, Y.H., Raschi, E., Kandiah, D.A., Tincani, A., Khamashta, M.A. 1993. Human b2-glycoprotein I binds to endothelial cells through a cluster of lysine residues that are critical for anionic phospholipids binding and offers epitopes for anti-b2glycoprotein I antibodies. J. Immunol. 160, 5572 /5578. Di Simone, N., Crowley, W.F., Jr, Wang, Q.F., Sluss, P.M., Schneyer, A.L. 1996. Characterization of inhibin/activin subunit, follistatin and activin type II receptors in human ovarian cancer cell lines: a potential role in autocrine growth regulation. Endocrinology 137, 486 /494. Di Simone, N., Hall, H.A., Welt, C., Schneyer, A.L. 1998. Activin regulates ba-subunit and activin receptor messenger ribonucleic acid and cellular proliferation in activin-responsive testicular tumor cells. Endocrinology 139, 1147 /1155. Gu, Y., Srivastava, R.K., Ou, J., Krett, N.L., Mayo, K.E., Gibori, G. 1995. Cell-specific expression of activin and its two binding
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195 proteins in the rat decidua: role of alpha 2-macroglobulin and follistatin. Endocrinology 136, 3815 /3822. Hamano, M., Kuramoto, K., Morisawa, T., Kato, Y., Arai, M. 1983. Study of the control mechanism of growth of endometrial carcinoma cells. Acta Obstet. Gynaecol. Jpn. 35, Abstr. No. 2183. Holinda, C.F., Hata, H., Gravanis, A., Kuramoto, H., Gurpide, E. 1986. Effects of estradiol on proliferation of endometrial adenocarcinoma cell line (Ishikawa line). J. Steroid. Biochem. 25, 781 / 786. Kassan, S., Mechanick, J.I., Gurpide, E. 1989. Altered estrogen receptor system in estrogen-unresponsive human endometrial adenocarcinoma cells. J. Steroid. Biochem. 33, 327 /333. Kuramoto, H., Hamano, M., Nishida, M., Taguchi, A., Jobo, T., Suzuki, M., Osanai, K. 1976. Establishment of a new human endometrial carcinoma cell line derived from ascites. Acta Obstet. Gynaecol. Jpn. 28, 1405 /1406. Kyprianou, N., English, H.F., Davidson, N.E., Isaacs, J.T. 1991. Programmed cell death during regression of the MCF-7 human breast cancer cell following estrogen ablation. Cancer Res. 51, 162 /166. Marone, M., Scambia, G., Mozzetti, S., Ferrandina, G., Iacovella, S., De Pasqua, A., Benedetti-Panici, P.L., Mancuso, S. 1998. Bcl-2, bax, and bcl-2 expression in normal and neoplastic ovarian tissue. Clin. Cancer Res. 4, 517 /524. Mathews, L.S. 1995. Activin receptors and cellular signaling by the receptor serine kinase family. Endocr. Rev. 15, 310 /341. Matzuk, M.M., Finegold, M.J., Su, J.J., Hsueh, A.J.W., Bradley, A. 1992. a-Inhibin is a tumor-suppressor gene with gonadal specificity in mice. Nature 360, 313 /319. Matzuk, M.M., Finegold, M.J., Mather, J.P., Krummen, L., Lu, L., Bradley, A. 1994. Development of cancer cachexia-like syndrome and adrenal tumors in inhibin deficient mice. Proc. Natl. Acad. Sci. USA 91, 8817 /8821. McGuire, W.L., Carbone, P.P., Vollmen, E.P. 1975. Estrogen Receptors in Human Breast Cancer. Raven Press, New York. McPherson, S.J., Thomas, T.Z., Wang, H., Gurusinghe, C.J., Rissbridger, G.P. 1997. Growth inhibitory response to Activin A and B by human prostate tumor cell lines, LNCaP and DU145. J. Endocrinol. 154, 535 /545. Nakamura, T., Sugino, K., Titani, K., Sugino, H. 1991. Follistatin, an activin-binding protein, associates with heparan sulfate chains of proteoglycans on follicular granulosa cells. J. Biol. Chem. 266, 19432 /19437. Nakamura, T., Taklo, K., Eto, Y., Shibai, H., Titani, K., Sugino, H. 1990. Activin-binding protein from rat ovary is follistatin. Science 247, 836 /838. Nishida, M., Kasahara, K., Kaneto, M., Iwasaki, H. 1985. Establishment of a new human endometrial adenocarcinoma cell line, Ishikawa cells, containing estrogen and progesterone receptors. Acta Obstet. Gynaecol. Jpn. 37, 1103 /1111. Oliver, M.H., Harrison, N.K., Bishop, J.E. 1989. A rapid and convenient assay for counting cells cultured in microwell plates: application for assessment of growth factors. J. Cell Sci. 92, 513 / 518.
195
Rose, P.G. 1996. Endometrial carcinoma. N. Engl. J. Med. 29, 640 / 649. Schneyer, A.L., Mason, A.J., Burton, L.E., Ziengner, J.R., Crowley, W.F. 1990. Immunoreactive inhibin a-subunit in human serum: implications for radioimmunoassay. J. Clin. Endocrinol. Metab. 70, 1208 /1212. Schneyer, A.L., Rzucidlo, D.A., Sluss, P.M., Crowley Jr, W.F. 1994. Characterization of unique binding kinetics of follistatin and activin or inhibin in serum. Endocrinology 135, 667 /674. Shikone, T., Matzuk, M.M., Perlas, E., Finegold, M.J., Lewis, K.A., Vale, W., Bradley, A., Hsueh, A.J.W. 1994. Characterization of gonadal sex cord-stromal tumor cell lines from inhibin-a and p53deficient mice: the role of activin as an autocrine growth factor. Mol. Endocrinol. 8, 983 /995. Su, G.H., Bansal, R., Murphy, K.M., Montgomery, E., Yeo, C.J., Hruban, R.H., Scott, E.K. 2001. ACVR1B (ALK4, activin receptor type 1B) gene mutations in pancreatic carcinoma. Proc. Natl. Acad. Sci. USA 98, 3254 /3257. Sugino, K., Kurosawa, N., Nakamura, T., Takio, K., Shimasaki, S., Ling, N., Titani, K., Sugino, H. 1993. Molecular heterogeneity of follistatin, an activin-binding protein: higher affinity of the carboxy-terminal truncated forms for heparin sulfate proteoglycans on the ovarian granulosa cell. J. Biol. Chem. 268, 15579 / 15587. Teixeira, C., Reed, J.C., Pratt, M.A.C. 1995. Estrogen promotes chemotherapeutic drug resistance by a mechanism involving Bcl-2 proto-oncogene expression in human breast cancer cells. Cancer Res. 55, 3902 /3907. Vale, W.W., Bilezikjian, L.M., Rivier, C. 1994. Reproductive and other roles of inhibin and activins. In: Knobil, E., Neill, J.D. (Eds.), The Physiology of Reproduction. Raven Press, New York, pp. 1861 /1878. Wang, Q.F., Khoury, R.H., Smith, P.C., McConnel, D.S., Padmanahban, V., Midgley, A.R., Schneyer, A.L., Crowley, W.F., Jr, Sluss, P.M. 1996. A two sites monoclonal antibody immunoradiometric assay for human follistatin: secretion by a human ovarian teratocarcinoma-derived cell line (PA-1). Endocrinology 81, 1434 / 1441. Wang, Q.F., Tilly, K.I., Tilly, J.L., Preffer, F., Schneyer, A.L., Crowley, W.F., Sluss, P.M. 1996. Activin inhibits basal and androgen-stimulated proliferation and induces apoptosis in the human prostatic cancer cell line, LNCaP. Endocrinology 137, 5476 /5483. Welt, C.K., Lambert-Messerlian, G., Zheng, W., Crowley, W.F., Jr, Schneyer, A.L. 1997. Presence of activin, inhibin and follistatin in epithelial ovarian carcinoma. J. Clin. Endocrinol. Metab. 82, 3720 /3727. Xiao, J., Garcia-Lloret, M., Winkler-Lowen, B., Miller, R., Simpson, K., Guilbert, L.J. 1997. ICAM-1-mediated adhesion of peripheral blood monocytes to the maternal surface of placental syncytiotrophoblast: implications for placental villitis. Am. J. Pathol. 150, 1845 /1860. Ying, S.H., Zhang, Z. 1996. Expression and localization of inhibin/ activin subunits and activin receptors in MCF-7 cells, a human breast cancer cell line. Breast Cancer Res. 37, 151 /160.
Regulation of endometrial adenocarcinoma cell proliferation by Activin-A and its modulation by 17b-estradiol Nicoletta Di Simone a, Alan L. Schneyer b, Dario Caliandro a, Roberta Castellani a, Alessandro Caruso a,* b
a Department of Obstetrics and Gynecology, Universita’ Cattolica del S. Cuore, Largo Gemelli 8, 00168 Rome, Italy National Center for Infertility Research and the Reproductive Endocrine Sciences Center, Department of Medicine, Massachusetts General Hospital, Boston, MA 02114, USA
Received 12 March 2001; accepted 27 August 2001
Abstract A role for activins in regulating cellular transformation is suggested by the a-inhibin knockout mouse in which development of gonadal tumors is associated with elevated activin levels. It was the purpose of the current study to determine whether activin had similar actions on endometrial cell lines, specifically on a well differentiated estrogen-responsive endometrial adenocarcinoma cell line (ISH) and estrogen-unresponsive cells (HEC-50) obtained from a poorly differentiated endometrial adenocarcinoma. Activin was secreted by both adenocarcinoma cell lines. Using reverse transcription-PCR, messenger RNA type I and type II activin receptor subtypes were detected in both cell lines: expression of IB and IIB was approximately three- to fourfold greater in ISH cells than in HEC-50 cells, while activin receptor IA and IIA messenger RNA levels were approximately equal in both cell lines. Activin treatment (30 /300 ng/ml) caused a dose- and time-dependent inhibition of ISH cells proliferation and resulted in a significant decrease in Bcl-2 protein and mRNA levels. No difference was observed in Bax expression. There was no significant effect of activin when the cultures of ISH cells were exposed to 17b-estradiol. In contrast, activin showed a weak, but significant, mitogenic effect on HEC-50 cells without modifications in Bax and Bcl-2 mRNA and protein levels. The results demonstrate that activin is a regulator of endometrial cancer cell growth. 17b-Estradiol may promote resistance of estrogen-responsive endometrial cancer cells to the growth-retarding effects of activin and one of the mechanisms might be a down-regulation of the activin receptors. # 2002 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Activin; Cancer; Endometrium
1. Introduction Activins are members of the transforming growth factor-b (TGF-b) superfamily and are involved with regulation of proliferation, differentiation and/or disorders of growth regulation such as oncogenesis (Ying and Zhang, 1996; Dalkin et al., 1996; Di Simone et al., 1996, 1998; Welt et al., 1997; Vale et al., 1994). The potential involvement of inhibin and activin in cellular transformation is suggested by a-inhibin subunit-deficient mutant mice, in which nearly 100% of males and females develop gonadal and adrenal tumors in early adulthood (Matzuk et al., 1992, 1994). Whether * Corresponding author. Tel./fax: /39-6-3551-0031. E-mail address: [email protected] (A. Caruso).
this is the result of reduced inhibin or increased activin remains unknown. The ability of activin to enhance proliferation in certain cell lines has been reported, suggesting that activin can act as a growth promoter and perhaps play a role in tumor formation and proliferation (Di Simone et al., 1996; Shikone et al., 1994). In contrast, activin can also inhibit proliferation and induce apoptosis in LnCAP prostatic cancer cells (Dalkin et al., 1996; Wang et al., 1996b; McPherson et al., 1997), rat pituitary (Billestrup et al., 1990) and corticotroph cells (Bilezikjian et al., 1991). The molecular basis for whether activin is stimulatory or inhibitory is incompletely understood. However, it is likely to be multifaceted because the end effect of activin would be influenced by factors such as the local levels of
0303-7207/02/$ - see front matter # 2002 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 3 - 7 2 0 7 ( 0 1 ) 0 0 6 4 8 - 7
188
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
follistatin (FS; Sugino et al., 1993; Nakamura et al., 1991) and the level and type of activin receptor expression (DeJong et al., 1993). FS is a monomeric glycoprotein whose only known actions are mediated through nearly irreversible binding and neutralization of activin (Nakamura et al., 1990; Schneyer et al., 1994). Members of the TGF-b superfamily act at the cellular level through a complex of two related serine/threonine kinase receptor subfamilies known as type I and type II receptors (Mathews, 1995). Although type I/II receptor complexes may exist in the unligated state, complex formation is necessary for signal transduction, so that the relative number of each type of activin receptor on the cell surface at any moment may be a critical factor in determining the cell’s responsiveness to activin. Activin subunits and activin receptors were observed in human endometrium (Gu et al., 1995), suggesting that this tissue might represent a source as well as a target of activin, with actions similar to those summarized above for prostatic or gonadal tumors. Endometrial carcinoma is thought to exist in two principal forms. The first is responsive to estrogens, occurs in relatively young perimenopausal women and is characterized by predominantly low grade, minimally invasive tumors and usually a positive prognosis (Rose, 1996). On the other hand, estrogen-independent endometrial tumors occur in relatively older postmenopausal women and are typically high grade, invasive tumors having a progressive clinical course and a poor prognosis. The availability of estrogen responsive and unresponsive endometrial adenocarcinoma cell lines allows investigation of factors that regulate growth in endometrial tumors. Such cell lines include the Ishikawa (ISH) cell line (Nishida et al., 1985), which originated from a well differentiated endometrial adenocarcinoma and contains estrogen and progesterone receptors. This cell line is responsive to estrogens (Holinda et al., 1986). Estrogen-unresponsive HEC-50 cells (Kuramoto et al., 1976; Kassan et al., 1989) were established from the ascites of a patient with poorly differentiated endometrial cancer. Using these two distinct cell lines, the biosynthesis and actions of inhibin/activin subunits and activin receptors were compared. In addition, potential regulation of activin action by 17b-estradiol (E2) was investigated in the estrogen-responsive ISH cells.
poorly differentiated endometrial adenocarcinoma (Kuramoto et al., 1976). The two cell lines were generously provided by Dr De Cicco, Department of Obstetrics and Gynecology, Catholic University, Rome, Italy. Cells used for these studies were cultured for 7/10 days, with several changes of medium, in phenol red free DMEM containing 2% FBS (Life Technologies, Grand Island, NY) stripped of steroids by absorption to dextran-coated charcoal suspension (0.025% Dextran/ 0.25% charcoal wt/vol, DCC), 2 mM L-glutamine and antibiotics (100 IU/ml penicillin and 100 mg/ml streptomycin sulfate). Cultures were maintained at 37 8C in a humidified atmosphere of 95% air /5% carbon dioxide. Media were changed every 2 days and the cells were subcultured once a week by replating trypsinized cell suspensions. The cells were grown to near confluence. After 48 h, the conditioned medium was collected and stored at /20 8C until assayed for protein hormones. 2.2. Immunoassays
2. Material and methods
Total activin-A was measured using a solid phase sandwich enzyme-linked immunoadsorbent assay (Elisa; Serotec, Turin, Italy). Wells of the microplate were dry coated with monoclonal antibody specific for the b-Asubunit of activin. Samples were incubated for 3 min at 100 8C with an equal volume of 20% SDS. Hydrogen peroxide solution (30% v/v) was then added and the samples allowed to stand at room temperature for 30 min. Then the samples were added along with assay diluent to the coated microplate followed by biotinylated monoclonal detection antibody. The plate was covered and incubated overnight at room temperature on a plate shaker. The following day the plate was washed and streptavidin /alkaline phosphatase added for 1 h. After additional washes, p -nitrophenylphosphate in 0.05 M ethanolamine buffer (pH 9.8) was added and incubated at room temperature until color developed in control wells, at which time the assay was read in a microtiter plate reader (Biorad Platerreader Model 405). Total immunoreactive inhibin was assessed as previously described (Bernstein et al., 1990) using the Monash Inhibin RIA (Schneyer et al., 1990), radioiodinated inhibin-A (Genentech), and a human follicular fluid assay standard that was calibrated at 22 pg/ml using recombinant inhibin-A. FS was quantitated by two-site monoclonal antibody for free FS as previously described (Wang et al., 1996a).
2.1. Cells and culture conditions
2.3. RNA extraction
The experiments reported here were carried out in a well differentiated estrogen-responsive endometrial adenocarcinoma cell line (ISH; Nishida et al., 1985) and in estrogen-unresponsive cells, HEC-50, obtained from a
Confluent cells remaining after medium collection were treated with Tri Reagent (Molecular Research Center, Cincinnati, OH) according to the manufacturer’s instructions. RNA integrity was confirmed by
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
agarose gel electrophoresis and ethidium bromide staining as well as by monitoring absorbance at A260/280. The RNA concentration was determined in all samples in two independent determinations and at two dilutions before RT-PCR experiments. All samples were stored at /80 8C. 2.4. RT-PCR procedures Complementary DNA (cDNA) was obtained by reverse transcription at 42 8C for 45 min in a 20-ml reaction mixture containing 200 ng of RNA, first treated with Dnase (Life Technologies), 0.5 mM of each deoxyNTP (Promega, Life Science, Florence, Italy), 2.5 mM oligodT (Life Technologies), and 2.5 U SuperScript (Life Technologies). The PCR reaction was carried out in a volume of 25 ml containing 2.5 ml 10 / PCR buffer, 5 mM MgCl2, 0.2 mM of each deoxy-NTP, 0.5 mM of each specific primer, 0.25 ml Taq polymerase (Promega) and 0.05 mCi nucleotide triphosphate ([32P]dCTP; NEN, Dreieich, Germany) to which 1 ml of the RT reaction product was added. The sequences of the specific primers were described previously (Di Simone et al., 1996, 1998; Marone et al., 1998). Amplification was achieved using a thermal cycler (Perkin Elmer GeneAmp, PCR System 9600). The amplification profile involved preincubation at 95 8C for 3 min, denaturation at 94 8C for 0.5 min, primer annealing at 57 8C for 1.3 min and extension at 72 8C for 1.3 min. In initial experiments, cycle number and RNA concentrations were optimized for each target, so that signals were always in the exponential portion of the amplification curve (35 cycles for Bcl-2 and Bax; 32 cycles for a, b subunits and FS; 28 cycles for activin receptors and 22 cycles for b-actin). Three microliters of the PCR reaction were electrophoresed in 5% polyacrylamide gels in Tris /borate / EDTA buffer. The concentration of each messenger RNA was determined by counting the radioactivity of each target by Instant/Imager (Camberra/Packard S.r.l., Milan, Italy) and was expressed as counts per min. To monitor DNA contamination in the reagents or samples, RT was omitted from each reaction to exclude genomic or cDNA contamination. Reactions without RNA was used to monitor contamination of RT or the primers. All of these controls gave negative results. 2.5. Southern blot analyses Previously, we verified the b-actin, a, b subunits and FS PCR products identities by Southern blot (Di Simone et al., 1996), using synthetic oligos located internal to the two PCR primers. In the present study, activin receptors, Bcl-2 and Bax identities were demonstrated with the same method (data not shown). PCR-amplified products without
189
radionuclide were fractionated on a 1.5% agarose gel, visualized by UV transillumination and the product sizes were estimated by comparison to molecular weight markers (PGEM, Promega). The gels were then denatured in 50 mM NaOH /1.5 M NaCl for 30 min and neutralized in 1 M Tris /1.5 M NaCl. The DNA was transferred by capillary action to a nylon membrane (Biorad S.r.l., Milan, Italy) in 20 / SSC (3 M NaCl and 0.3 M sodium citrate, pH 7.0) and subjected to Southern blot analyses using [32P]dATP (NEN)-labeled oligonucleotide probes. 2.6. Semiquantitative analyses of activin receptors messenger RNA expression Semiquantitative analysis of messenger RNA (mRNA) expression was carried out as previously described for ovarian cancer cell lines (Di Simone et al., 1996). RNAs prepared from endometrial cells were diluted in water to concentrations ranging from 200 to 3 ng per sample. The cDNA was amplified for activin receptors type I (Ia and Ib) and type II (IIa and IIb). Three microliters of the PCR reactions were electrophoresed in 5% polyacrylamide gels in Tris /borate / EDTA buffer. To allow direct comparison of RNA concentrations between cell lines, PCR for each target was carried out in a single experiment and electrophoresed at the same time. As each target was amplified for each cell line in one PCR experiment under identical conditions, the resulting values were plotted against increasing concentrations of RNA, giving dose /response curves for each cell line that could then be compared for relative steady state mRNA levels. Each experiment was repeated three times and run on replicate gels. We observed no differences between the two gels. The results from one representative experiment are shown. 2.7. Proliferation studies: Methylene Blue assay We used a colorimetric technique (Oliver et al., 1989) to study cell growth. ISH and HEC-50 cells were grown in phenol red free DMEM supplemented with 2% FBS stripped of steroids in 96-well culture plates. The effects of E2 (Sigma, St Louis, MO) on ISH cell proliferation were evident at concentrations ranging from 10 9 to 10 7 M (data not shown). Then, we determined the kinetics of E2-stimulated cell growth. E2 (10 8 M) was added for 12, 24, 36, 48 and 72 h from a stock solution in ethanol. The final levels of ethanol in each well, including controls was 0.1%. After washing with 0.15 M saline, the cell layer was fixed by adding 100 ml of 10% formol saline to each well for at least 30 min. The fixative was shaken off each plate and 100 ml of filtered 1% (w/v) Methylene Blue in 0.01 M borate buffer (pH 8.5) was added to each well. After 30 min,
190
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
excess dye was removed. The absorbance at 650 nm (A 650) was measured from each well by a microplate photometer (Biorad Platerreader Model 405). In order to assess the accuracy of the assay, preliminary cultures were set up in replicate for a range of cell concentrations. The A650 was directly proportional to the cell number determined by haemocytometer count (data not shown; P B/0.01). Each experiment was repeated three times. Because ISH cell proliferation is estrogen-sensitive, the effects of activin (kindly provided by the National Hormone and Pituitary Program, Baltimore, USA) on E2 stimulated proliferation were also examined. Varying doses of activin (from 3 to 300 ng/ml) alone or in combination with E2 (10 8 M) were added to triplicate wells. Cells were treated for 48 h with the same drug concentrations and the medium was changed every 24 h. Results of cell growth assays are expressed as mean9/ S.E. of three different experiments. 2.8. Effect of activin on pro- and anti-apoptotic gene mRNA and proteins To examine the effect of activin on Bcl-2 and Bax mRNA and protein levels, the cells were treated on day 2 of culture with activin-A (3 /300 ng/ml) for 48 h in the presence or the absence of Estradiol (10 8 M). Cellular RNA was extracted for RT-PCR analyses and the results were expressed relative to b-actin for each time-point. Bcl-2 and Bax proteins were evaluated by cell ELISA (Del Papa et al., 1993; Xiao et al., 1997). On day 1 (24 h) and day 2 (48 h) of culture, the medium was removed and the cells were washed twice with DMEM and incubated for 60 min at room temperature with antihuman Bcl2 (10 mg/ml; clone 124; DAKO A/S, Glostrup, Denmark) and anti-human Bax (10 mg/ml; Santa Cruz Biotechnology, Santa Cruz, CA). After two more washes, the cells were fixed with paraformaldehyde for 15 min at room temperature. Cells were then washed twice and incubated for another 60 min at room temperature with 100 ml of alkaline phosphatase-conjugated affinity-purified anti-IgM or anti-IgG (Sigma/ Aldrich). After two washes with DMEM and two more with phosphate buffer, p -nitrophenylphosphate substrate was used to reveal IgG binding. The optical density (OD) values of the enzymatic reactions were read at 405 nm, after a 30-min incubation, with a microplate photometer (Platerreader, Bio-Rad, Milan, Italy). 2.9. Time-course study of the effect of estradiol on activin receptor mRNA levels ISH cells were passaged at dilution not greater than 1:5 after incubation for 5 min at 37 8C in 0.05% trypsin
and were maintained in culture for 72 h before testing the mRNA steady-state levels with or without E2 treatment. For time-course experiments, ISH cells were stimulated on day 2 of culture with E2 (10 8 M) for 12, 24 and 48 h. Cellular RNA was extracted for RT-PCR analyses with specific probes for activin receptors type I (Ia and I b) and type II (IIa and IIb). Results are the mean9/S.E. of three separate experiments, expressed relative to b-actin for each time point. 2.10. Statistical analyses Significant differences were determined using Student’s t -test, with P B/0.05 considered significant for all experiments.
3. Results 3.1. Steady-state messenger RNA levels Inhibin/activin subunit, FS and activin receptor mRNA expression was examined using RT-PCR. Amplification cycle number was varied to determine the exponential portion of the amplification curve for each target and the identity of the PCR products was verified by Southern blot either previously (inhibin/activin subunits and FS; Di Simone et al., 1996) or as demonstrated in Fig. 1 for activin receptors. Amplification of activin receptors required 28 cycles (Fig. 2) while for inhibin/ activin subunits and FS, 32 cycles were used. As indicated in Table 1, mRNAs encoding b-actin, activin b and bb subunits and FS were identified in both cell lines. A 823-bp inhibin a-subunit mRNA was detected in HEC-50 cells but was undetectable in the ISH cell line under identical conditions. 3.2. Activin receptors: semiquantitative analyses To compare mRNA levels for activin receptors in each cell line, the cells were harvested at approximately equal levels of confluence. After determination of RNA concentration, increasing amounts of RNA were reverse transcribed and amplified. Fig. 3 demonstrates that HEC-50 and ISH cells express mRNA for all four activin receptor subtypes: increasing RNA concentrations resulted in stronger PCR bands for each activin receptor subtype, creating a series of dose/response curves for each target cDNA. Of interest, expression of activin receptor mRNA IB and IIB was approximately three- to fourfold greater in ISH cells than in HEC-50 cells. On the other hand, activin receptor IA and IIA mRNA levels were approximately equal in both cell lines. Thus, ISH cells expressed higher IB/IA (1.039/0.14; P B/0.02) and IIB/IIA (2.429/
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
191
Fig. 1. Verification of activin receptors (Ia, Ib, IIa and IIb) by Southern blot. 32P-Labeled oligonucleotides for sequences internal to each pair of PCR primers were used to verify the identity of the PCR products. Ethidium bromide-stained agarose gels containing the PCR products were transferred to nitrocellulase and hybridized with labelled probe. Southern blot of the gels: specific recognition of the PCR products without detection of adjacent DNAs demonstrates the specificities of the PCR targets. Lane 1, activin receptor IIA; lane 2, activin receptor IIB; lane 3, activin receptor IA; lane 4, activin receptor IB.
mRNA extraction and expressed per 100 000 cells after 48 h of culture. Using a specific two-site activin-A immunoassay that measures total activin, HEC-50 and ISH cells secreted 119/1.5 and 49/0.3 ng/ml activin-A/100 000 cells per 48 h, respectively. No activin immunoactivity was detected in unconditioned medium. Inhibin and free FS were undetectable for both cell lines.
3.4. Cell proliferation studies
Fig. 2. Titration of PCR cycles using 0.2 mg total RNA from HEC-50 cells. Amplification is linear over the range of 23 /33 cycles for activin receptor IIa.
Table 1 Summary of comparative steady state mRNA levels determined from endometrial cell lines mRNA
HEC-50
ISH
a ba bb FS
/
, 100 /1000 cpm/min; , 1000 /10 000 cpm/min; , 10 000 cpm/min.
0.5; P B/0.01) activin receptor mRNA ratios compared to the HEC-50 cells (IB/IA 0.369/0.04 and IIB/IIA 0.299/0.03). 3.3. Hormonal secretion The ability of endometrial cell lines to secrete the hormones whose mRNA production was investigated by RT-PCR, was examined in the same cultures used for
The effect of E2 on the estrogen-responsive endometrial cancer cell line (ISH) was investigated. Linear growth was apparent from 12 h of culture but no difference in cell number was observed in the absence or the presence of E2 (10 8 M) until 48 h (Fig. 4). In contrast, there was no effect of estrogen on HEC-50 cells (data not shown). As both the cell lines expressed significant amounts of activin receptor mRNA, the effect of activin on cell growth was investigated. As shown in Fig. 5, activin caused a dose-dependent inhibition of ISH cell proliferation: a significant inhibition by activin on cell number was observed within 24 h of treatment for an activin dose of 300 ng/ml. This growth inhibition was more pronounced at 48 h with the lower dose (30 ng/ml) inhibiting cell proliferation. Because ISH cell proliferation is estrogen-sensitive, the effects of activin on estrogen-related proliferation were also examined. Interestingly, the anti-proliferative effect of activin on ISH cells was abolished in the presence of E2. In contrast to ISH cells, activin (30 ng/ml) significantly enhanced HEC-50 proliferation. Although proliferation of HEC-50 cells was enhanced by 30 ng/ml of activin, higher doses (300 ng/ml) did not significantly change cellular proliferation with respect to untreated cells. No effect of E2 on activin-induced HEC-50 cell proliferation was observed.
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
192
Fig. 3. The expression of activin receptor mRNA was quantitated by PCR as described in Section 2. One representative experiment is shown for each mRNA quantitation. RT reactions containing increasing amounts of RNA resulted in linearly increasing amounts of PCR products. Twenty-eight cycles of amplification were used for activin receptors. Analysis of each PCR product was carried out simultaneously for both cell lines. For each experiment, RT was omitted to exclude genomic or cDNA contamination; reactions without RNA were used to monitor contamination of RT or the primers (data not shown). All these controls gave negative results. ISH cells: I; 1, 2 and 3: 3, 50 and 200 ng/ml of RNA, respectively. HEC-50 cells: m; 4, 5 and 6: 3, 50 and 200 ng/ml of RNA, respectively.
ISH cells, the effect of activin was abolished in the presence of E2. The Bcl-2 protein was higher in HEC-50 than in ISH cells (7899/98 vs. 4159/54, P B/0.01). Treatment with activin had no effect on Bcl-2 protein in HEC-50 cells and on Bax protein for both cell lines (data not shown). In contrast, activin A (30 /300 ng/ml) induced a significant decrease in Bcl-2 protein of ISH cells (Fig. 6). The effect of activin was abolished in the presence of E2.
8
Fig. 4. Estradiol (10 M; I) induces proliferation of ISH cells. Cell proliferation was determined by a colorimetric technique (mean9/S.E. of triplicate culture wells). OD, optical density values. Detailed procedures are explained in Section 2. Each experiment was repeated three times. Significant differences from the control (ISH cells without E2; m) are indicated, *P B/0.02; **P B/0.01.
3.5. Regulation of pro- and anti-apoptotic Bax and Bcl-2 mRNAs and proteins The Bax and Bcl-2 were expressed at both messenger ribonucleic acid and protein levels in ISH and HEC-50 cells. The mRNA levels of Bax and Bcl-2 were investigated by RT-PCR analyses and the results were expressed relative to b-actin for each time point. No difference was observed in Bax mRNA between HEC-50 and ISH cells. The Bcl2 mRNA was higher in HEC-50 than in ISH cells (0.899/0.07 vs. 0.379/0.06, P B/0.01). As shown in Fig. 6, treatment with activin A (30, 300 ng/ml) induced a significant decrease in Bcl-2 mRNA in
3.6. Regulation of activin receptors mRNA levels by estradiol Because ISH proliferation is estrogen-sensitive and E2 eliminated activin’s antiproliferative effect, we analyzed the action of E2 treatment on activin receptor mRNA expression. As previously shown, ISH cells expressed transcripts for all four activin receptor subtypes. E2 (10 8 M) treatment significantly reduced mRNA levels for all four activin receptor subtypes, achieving suppression from 30 to 50% of mRNA levels at 24 and 48 h of incubation (Fig. 7). The levels of b-actin mRNA were unchanged during E2 treatment.
4. Discussion The present study demonstrates that activin subunit (bA and bB) mRNAs are expressed in endometrial cancer cell lines as determined by RT-PCR. In addition,
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
Fig. 5. Dose /response curve of activin-A on basal (I) and E2induced (10 8 M; m) endometrial cell proliferation. Significant differences from the untreated cells (0 ng/ml): *P B/0.02; **P B/0.01. After 48-h incubation, E2 (m) induces ISH cell proliferation (§P B/ 0.01). The concomitant treatment with activin-A (3 /300 ng/ml) does not significantly modify the E2-induced endometrial cell proliferation. Cell growth was determined by a colorimetric technique (mean9/S.E. of three separate experiments). OD, optical density values.
we observed that these cells express all four activin receptor subtypes. These data lend support to the hypothesis that locally synthesized cytokines, in particular members of the TGF-b superfamily, may influence endometrial function. In the ISH cell line, activin significantly inhibited cell proliferation and this inhibition was abolished in the presence of estrogen. Several studies have demonstrated how estrogen-dependent carcinomas regress after ovariectomy through a pathway involving programmed cell death or apoptosis, suggesting that estrogen may contribute to the survival of these cells (McGuire et al., 1975; Kyprianou et al., 1991). It was also recently reported that estrogen treatment induces Bcl-2 but not Bax mRNA in MCF-7 cells. This concept is further supported by the finding that estrogen may affect the efficacy of cancer treatments through modulation of Bcl-2 biosynthesis (Teixeira et al., 1995). The present study has demonstrated that mRNAs and proteins of Bax and Bcl-2 are expressed in endometrial adenocarcinoma cell lines. Activin-A resulted in a significant
193
Fig. 6. Effects of activin A on Bcl-2 mRNA (top) and protein (bottom) expression in the absence (no shading) or presence of estradiol (shading). The mRNA levels were normalized to b-actin. Protein levels were determined by ELISA. OD, optical density values at 405 nm. Significant differences from untreated cells (Activin-A 0 ng/ ml) are indicated: *P B/0.05; **P B/0.01. Each time point represents the mean9/S.E. of three independent experiments.
decrease in anti-apoptotic Bcl-2 protein level. Estradiol seems to promote resistance of estrogen-responsive endometrial cancer cells to the growth-retarding effects of activin and one mechanism of action seems to be the effect on Bcl-2 levels. The presence of effects of E2 on proliferation during the initial period of growth (48 h of culture) might be due to the experimental conditions of the present study. Although E2 (10 8 M) sustained ISH cell proliferation after about 10 days of culture (Holinda et al., 1986), under the chosen experimental condition (15% charcoaltreated fetal bovine serum renewed every 2 /3 days), Hamano et al. (1983) reported effects of E2, on growth rates of ISH cells within 2 days using less serum (1%) and lower initial cell number. In the estrogen-responsive endometrial cancer cell line (ISH), E2 treatment induced a state of resistance against activin’s growth inhibiting activity. Furthermore, when ISH cells were treated with E2, the mRNA levels for activin receptors decreased significantly, suggesting that the proliferative effect of estrogen in these cells could be mediated by a down-regulation of the activin signaling pathway. In contrast, activin-A stimulated HEC-50 cell growth. How the same growth factor, acting through the same specific receptors, is able to produce growth inhibition
194
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195
The fact that HEC-50 cell growth can be stimulated by activin suggests that these cancer cells, through malignant transformation, have escaped from growth inhibition by the negative growth regulator thereby acquiring an altered responsiveness to this growth factor. In conclusion, the diversity of the observed effects is of interest, because activin is considered to be a potent growth regulator in most epithelial carcinoma cells (Ying and Zhang, 1996; Dalkin et al., 1996; Shikone et al., 1994), apparently acting as an autocrine regulator and with opposite effects in various cell types. Our results indicate that activin’s actions in endometrial tissue is susceptible to being subverted during oncogenesis. For this reason, elucidation of the regulatory pathways governing the expression and the actions of activin system is likely to be critical to our understanding of oncogenesis in a variety of tissues, including the endometrium.
References Fig. 7. Effects of incubation time in the absence (unshaded) or presence of E2 (10 8 M; shaded) on activin receptor mRNA expression. Levels of activin receptors Ia, Ib, IIa and IIb mRNA were significantly suppressed at 24 and 48 h (*P B/0.05; **P B/0.01) by E2. Each time or treatment point represents the mean9/S.E. of three independent experiments expressed relative to b-actin for each time point.
in some cell lines and stimulation in others is still unclear. Among the several possibilities that could be hypothesized is that activin type I and type II receptor heterogeneity underlies differential cellular signaling. Support for this possibility can be found in our observations: using a semiquantitative comparative RT-PCR assay, ISH cells show three- to fourfold higher expression of type IIB and IB activin receptor mRNA than HEC-50 cells. This could affect the number of IIA/ IA versus IIB/IB receptor complexes in different cell types, which may then produce different biological results. On the other hand, receptor function rather than receptor levels might be more important in carcinogenesis. Previously, TGF-b has been shown to be involved in the growth regulation of endometrial cells (Anzai et al., 1992): TGF-b inhibited ISH cell proliferation whereas it stimulated HEC-50 cell growth. Loss of sensitivity to growth inhibition by TGF-b is a phenomenon often observed in human epithelial tumor cells and is linked to malignant progression. Mutations of the activin type I receptor have been recently described by Su et al. (2001) in pancreatic carcinoma. These defects may impair the function of activin receptors with a secondary inactivation of the activin signals.
Anzai, Y., Gong, Y., Holinka, F.H., Murphy, L.J., Murphy, L.C., Kuramoto, H., Gurpide, E. 1992. Effects of transforming growth factors and regulation of their mRNA levels in two human endometrial adenocarcinoma cell lines. J. Steroid Biochem. Mol. Biol. 42, 449 /455. Bernstein, J.R., Crowley, W.F., Jr, Schneyer, A.L. 1990. An improved method of purifying inhibin radioligand for radioimmunoassay. Biol. Reprod. 43, 492 /496. Billestrup, N., Gonzalez-Marchon, C., Potter, E., Vale, W. 1990. Inhibition of somatotroph growth and growth hormone biosynthesis by activin in vitro. Mol. Endocrinol. 4, 356 /362. Bilezikjian, L.M., Blount, A.L., Campen, C.A., Gonzalez-Marchon, C., Vale, W. 1991. Activin-A inhibits proopimelanocortin messenger RNA accumulation and adrenocorticotropin secretion of AtT20 cells. Mol. Endocrinol. 5, 1389 /1395. Dalkin, A.C., Gilrain, J.T., Bradshaw, D., Myers, C.E. 1996. Activin inhibition of prostate cancer cell growth: selective actions on androgen-responsive LNCaP cells. Endocrinology 137, 5230 /5235. DeJong, F.H., de Winter, J.P., Wesseling, J.G., Timmerman, M.A., van Genesen, S., van den Eijnden-van Raaij, A.J., van Zoelen, E.J. 1993. Inhibin subunits, follistatin and activin receptors in the human teratocarcinoma cell line tera-2. Biochem. Biophys. Res. Commun. 192, 1334 /1339. Del Papa, N., Sheng, Y.H., Raschi, E., Kandiah, D.A., Tincani, A., Khamashta, M.A. 1993. Human b2-glycoprotein I binds to endothelial cells through a cluster of lysine residues that are critical for anionic phospholipids binding and offers epitopes for anti-b2glycoprotein I antibodies. J. Immunol. 160, 5572 /5578. Di Simone, N., Crowley, W.F., Jr, Wang, Q.F., Sluss, P.M., Schneyer, A.L. 1996. Characterization of inhibin/activin subunit, follistatin and activin type II receptors in human ovarian cancer cell lines: a potential role in autocrine growth regulation. Endocrinology 137, 486 /494. Di Simone, N., Hall, H.A., Welt, C., Schneyer, A.L. 1998. Activin regulates ba-subunit and activin receptor messenger ribonucleic acid and cellular proliferation in activin-responsive testicular tumor cells. Endocrinology 139, 1147 /1155. Gu, Y., Srivastava, R.K., Ou, J., Krett, N.L., Mayo, K.E., Gibori, G. 1995. Cell-specific expression of activin and its two binding
N. Di Simone et al. / Molecular and Cellular Endocrinology 192 (2002) 187 /195 proteins in the rat decidua: role of alpha 2-macroglobulin and follistatin. Endocrinology 136, 3815 /3822. Hamano, M., Kuramoto, K., Morisawa, T., Kato, Y., Arai, M. 1983. Study of the control mechanism of growth of endometrial carcinoma cells. Acta Obstet. Gynaecol. Jpn. 35, Abstr. No. 2183. Holinda, C.F., Hata, H., Gravanis, A., Kuramoto, H., Gurpide, E. 1986. Effects of estradiol on proliferation of endometrial adenocarcinoma cell line (Ishikawa line). J. Steroid. Biochem. 25, 781 / 786. Kassan, S., Mechanick, J.I., Gurpide, E. 1989. Altered estrogen receptor system in estrogen-unresponsive human endometrial adenocarcinoma cells. J. Steroid. Biochem. 33, 327 /333. Kuramoto, H., Hamano, M., Nishida, M., Taguchi, A., Jobo, T., Suzuki, M., Osanai, K. 1976. Establishment of a new human endometrial carcinoma cell line derived from ascites. Acta Obstet. Gynaecol. Jpn. 28, 1405 /1406. Kyprianou, N., English, H.F., Davidson, N.E., Isaacs, J.T. 1991. Programmed cell death during regression of the MCF-7 human breast cancer cell following estrogen ablation. Cancer Res. 51, 162 /166. Marone, M., Scambia, G., Mozzetti, S., Ferrandina, G., Iacovella, S., De Pasqua, A., Benedetti-Panici, P.L., Mancuso, S. 1998. Bcl-2, bax, and bcl-2 expression in normal and neoplastic ovarian tissue. Clin. Cancer Res. 4, 517 /524. Mathews, L.S. 1995. Activin receptors and cellular signaling by the receptor serine kinase family. Endocr. Rev. 15, 310 /341. Matzuk, M.M., Finegold, M.J., Su, J.J., Hsueh, A.J.W., Bradley, A. 1992. a-Inhibin is a tumor-suppressor gene with gonadal specificity in mice. Nature 360, 313 /319. Matzuk, M.M., Finegold, M.J., Mather, J.P., Krummen, L., Lu, L., Bradley, A. 1994. Development of cancer cachexia-like syndrome and adrenal tumors in inhibin deficient mice. Proc. Natl. Acad. Sci. USA 91, 8817 /8821. McGuire, W.L., Carbone, P.P., Vollmen, E.P. 1975. Estrogen Receptors in Human Breast Cancer. Raven Press, New York. McPherson, S.J., Thomas, T.Z., Wang, H., Gurusinghe, C.J., Rissbridger, G.P. 1997. Growth inhibitory response to Activin A and B by human prostate tumor cell lines, LNCaP and DU145. J. Endocrinol. 154, 535 /545. Nakamura, T., Sugino, K., Titani, K., Sugino, H. 1991. Follistatin, an activin-binding protein, associates with heparan sulfate chains of proteoglycans on follicular granulosa cells. J. Biol. Chem. 266, 19432 /19437. Nakamura, T., Taklo, K., Eto, Y., Shibai, H., Titani, K., Sugino, H. 1990. Activin-binding protein from rat ovary is follistatin. Science 247, 836 /838. Nishida, M., Kasahara, K., Kaneto, M., Iwasaki, H. 1985. Establishment of a new human endometrial adenocarcinoma cell line, Ishikawa cells, containing estrogen and progesterone receptors. Acta Obstet. Gynaecol. Jpn. 37, 1103 /1111. Oliver, M.H., Harrison, N.K., Bishop, J.E. 1989. A rapid and convenient assay for counting cells cultured in microwell plates: application for assessment of growth factors. J. Cell Sci. 92, 513 / 518.
195
Rose, P.G. 1996. Endometrial carcinoma. N. Engl. J. Med. 29, 640 / 649. Schneyer, A.L., Mason, A.J., Burton, L.E., Ziengner, J.R., Crowley, W.F. 1990. Immunoreactive inhibin a-subunit in human serum: implications for radioimmunoassay. J. Clin. Endocrinol. Metab. 70, 1208 /1212. Schneyer, A.L., Rzucidlo, D.A., Sluss, P.M., Crowley Jr, W.F. 1994. Characterization of unique binding kinetics of follistatin and activin or inhibin in serum. Endocrinology 135, 667 /674. Shikone, T., Matzuk, M.M., Perlas, E., Finegold, M.J., Lewis, K.A., Vale, W., Bradley, A., Hsueh, A.J.W. 1994. Characterization of gonadal sex cord-stromal tumor cell lines from inhibin-a and p53deficient mice: the role of activin as an autocrine growth factor. Mol. Endocrinol. 8, 983 /995. Su, G.H., Bansal, R., Murphy, K.M., Montgomery, E., Yeo, C.J., Hruban, R.H., Scott, E.K. 2001. ACVR1B (ALK4, activin receptor type 1B) gene mutations in pancreatic carcinoma. Proc. Natl. Acad. Sci. USA 98, 3254 /3257. Sugino, K., Kurosawa, N., Nakamura, T., Takio, K., Shimasaki, S., Ling, N., Titani, K., Sugino, H. 1993. Molecular heterogeneity of follistatin, an activin-binding protein: higher affinity of the carboxy-terminal truncated forms for heparin sulfate proteoglycans on the ovarian granulosa cell. J. Biol. Chem. 268, 15579 / 15587. Teixeira, C., Reed, J.C., Pratt, M.A.C. 1995. Estrogen promotes chemotherapeutic drug resistance by a mechanism involving Bcl-2 proto-oncogene expression in human breast cancer cells. Cancer Res. 55, 3902 /3907. Vale, W.W., Bilezikjian, L.M., Rivier, C. 1994. Reproductive and other roles of inhibin and activins. In: Knobil, E., Neill, J.D. (Eds.), The Physiology of Reproduction. Raven Press, New York, pp. 1861 /1878. Wang, Q.F., Khoury, R.H., Smith, P.C., McConnel, D.S., Padmanahban, V., Midgley, A.R., Schneyer, A.L., Crowley, W.F., Jr, Sluss, P.M. 1996. A two sites monoclonal antibody immunoradiometric assay for human follistatin: secretion by a human ovarian teratocarcinoma-derived cell line (PA-1). Endocrinology 81, 1434 / 1441. Wang, Q.F., Tilly, K.I., Tilly, J.L., Preffer, F., Schneyer, A.L., Crowley, W.F., Sluss, P.M. 1996. Activin inhibits basal and androgen-stimulated proliferation and induces apoptosis in the human prostatic cancer cell line, LNCaP. Endocrinology 137, 5476 /5483. Welt, C.K., Lambert-Messerlian, G., Zheng, W., Crowley, W.F., Jr, Schneyer, A.L. 1997. Presence of activin, inhibin and follistatin in epithelial ovarian carcinoma. J. Clin. Endocrinol. Metab. 82, 3720 /3727. Xiao, J., Garcia-Lloret, M., Winkler-Lowen, B., Miller, R., Simpson, K., Guilbert, L.J. 1997. ICAM-1-mediated adhesion of peripheral blood monocytes to the maternal surface of placental syncytiotrophoblast: implications for placental villitis. Am. J. Pathol. 150, 1845 /1860. Ying, S.H., Zhang, Z. 1996. Expression and localization of inhibin/ activin subunits and activin receptors in MCF-7 cells, a human breast cancer cell line. Breast Cancer Res. 37, 151 /160.