Differential Regulation of Gene Expression by Estrogen in Estrogen Growth-Independent and -Dependent MCF-7 Human Breast Cancer Cell Sublines
Hyeseong Cho, Peter A. NG, and Benita S. Katzenellenbogen Department of Physiology and Biophysics University of Illinois University of Illinois College of Medicine Urbana, Illinois 61801
We have examined the ability of estradiol (E2) to regulate the expression of three mRNAs [for pS2, progesterone receptor (PR), and estrogen receptor (ER)], known to be under E2 regulation in the parental E2 growth-responsive MCF-7 cells, in an E2 growthindependent MCF-7 cell subline. This subline (denoted MCF-7 K3), previously isolated from the parental estrogen-dependent MCF-7 K1 human breast cancer cells after long term growth in vitro in the absence of estrogen, acquired estrogen-independent growth in vitro as well as the ability to form tumors in nude mice in vivo without estrogen. We find that the content of pS2 mRNA and the transcription rate of the pS2 gene, while being markedly increased by E2 in MCF-7 K1 cells, are no longer stimulated by E2 in this subline, although protein kinase activators tremendously increase (> 10-fold) pS2 mRNA in both K1 and K3 cells. In fact, basal pS2 mRNA levels are elevated 2.8 ± 0.4-fold in MCF7 K3 cells, and E2 evokes a concentration-dependent suppression of the pS2 mRNA level. In contrast, PR mRNA in the K3 subline, as in the parental K1 cells, is still up-regulated by E2, and ER mRNA content and the ER mRNA transcription rate are still downregulated by E2 and show normal E2 dose-response relationships, implying that the ER in this subline is functional. These results demonstrate that the progression to estrogen-independent growth in K3 cells is accompanied by a change in the regulation of some estrogen-induced genes by estrogen. While PR and ER retain normal patterns of regulation by E2, the pS2 gene in the estrogen growth-independent K3 subline is differentially affected and is no longer stimulated by E2. Our data suggest that this altered regulation of the pS2 gene is probably not caused by a defect of the ER or ER regulation in this subline. (Molecular Endocrinology 5: 1323-1330, 1991) 0888-8809/91/1323-1330$03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society
INTRODUCTION
A large proportion of breast cancer is regulated by sex steroid hormones, and the responsiveness of breast cancers to these steroid hormones is highly correlated with the presence of hormone receptors (1, 2). Hence, estrogen receptor (ER) status in breast cancer is used as an important indicator of endocrine therapy (3). During endocrine therapy, a progression from hormone dependence to independence often occurs, leading to resistance to hormonal therapy (4, 5). The ER-positive MCF-7 human breast cancer cell line (designated MCF7 K1) shows estrogen-dependent growth in vitro as well as estrogen-dependent tumorigenicity in vivo. We have previously isolated a subline of the MCF-7 cells (designated MCF-7 K3), which acquired estrogen-independent growth in vitro after prolonged growth in the absence of estrogen (6). In contrast to the parental MCF-7 K1 cells, this MCF-7 K3 subline is also capable of forming tumors in ovariectomized athymic nude mice without estrogen stimulation (5). Thus, this subline provides a unique system for studying the progression from hormone dependence to independence in human breast cancer cells and the changes that accompany this progression. In the present study we have compared the regulation of three estrogen-modulated genes in the estrogen growth-independent MCF-7 K3 subline: namely, pS2, ER, and progesterone receptor (PR). Since the expression of the pS2 gene is under the direct control of estrogen in MCF-7 cells (7-9) and is rapidly up-regulated by estrogen, its expression is a useful marker for investigating estrogen responsiveness in breast cancer cells. Although a portion of the primary structure of the pS2 protein resembles epidermal growth factor (10), and pS2 shows considerable homology to pancreatic spasmolytic polypeptide (11), the function of the pS2 protein remains unknown. The expression of pS2 mRNA is found in most ER-positive breast cancer cells, but not in estrogen-insensitive breast cancer cells such
1323
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 1324
Vol 5 No. 9
as T47D cells. Ninety-six percent of human breast tumors expressing the pS2 protein also contain ER (12). The presence of the pS2 protein is also considered an important variable for prediction of the relapse and survival rate in breast cancer patients (13). While pS2 gene expression is markedly stimulated by estrogen in wild-type parental MCF-7-K1 cells, we found that pS2 mRNA in the K3 subline is not induced, but, rather, is suppressed by estrogen. Similarly, the transcription rate of the pS2 gene in the MCF-7 K3 subline is not accelerated by estrogen. Interestingly, and in contrast, PR mRNA and ER mRNA levels in the K3 subline are still under the usual control of estrogen, implying that the ER in the K3 subline is biologically functional.
levels differed, with K3 cells containing 2.8 ± 0.4 times more pS2 mRNA (n = 4 experiments) than that found in MCF-7 K1 cells (data not shown). In these comparisons, 36B4 mRNA was used as an internal control, with pS2 being probed simultaneously in MCF-7 K1 and K3 cell RNAs in each experiment. The dose-response effect of E2 on pS2 mRNA levels in MCF-7 K1 and K3 cells was then examined. As shown in Fig. 1, 10"10 M E2 increased the pS2 mRNA level in the MCF-7 K1 cells nearly 10-fold. Somewhat lower stimulation was observed with 10 9 and 10~8 M E2 (Fig. 2A), although pS2 mRNA levels were still several-fold higher than the control values. In the MCF-7 K3 subline, 10 12 and 10 11 M E2 had no effect on pS2 mRNA levels (Figs. 1 and 2B), while higher concentra-
A
RESULTS PS2 mRNA Regulation by Estrogen in MCF-7 K1 and K3 Cell Sublines: Estradiol (E2) Dose Response The regulation of the pS2 mRNA level in MCF-7 K1 and K3 cells was examined by Northern blot analysis after 48-h exposure of cells to increasing concentrations of E2 (Fig. 1). Cells were switched to medium without phenol red and containing charcoal-dextran-stripped calf serum (CDCS) to eliminate estrogenic sources before treatment. We first compared the relative content of pS2 mRNA in control estrogen-withdrawn MCF-7 K1 and K3 cells and observed that the basal pS2 mRNA
Actin
pS2 C 101410121010109108ME2
pS2
Control
-12
-11
-10
-9
-8
Estradiol concentration (log M)
Fig. 1. Dose Response of pS2 mRNA to E2 in MCF-7 K1 and K3 Cell Sublines MCF-7 cell sublines were treated with 1O-12-1O** M E2. Total RNA was isolated using a guanidinium thiocyanate-phenolchloroform extraction method (see Materials and Methods). Twenty micrograms of total RNA were separated on 1.4% formaldehyde-denaturing agarose gel and transferred to GeneScreen Plus. Hybridization was performed with a 0.3-kb 32 P-labeled pS2 cDNA probe at 65 C. Autoradiographs were quantified using laser densitometry. The amount of mRNA loaded on the gel was normalized to the 36B4 or actin mRNA content. The same fold change in pS2 mRNA level was observed whether pS2 was quantitated relative to actin or 36B4 as the internal standard. Values represent the mean and SE of two to four determinations.
Fig. 2. A and B, Northern Blot Analysis of pS2 mRNA in MCF7 K1 and MCF-7 K3 Cell Sublines in Response to Various Concentrations of E2 MCF-7 K1 and K3 cells were treated with lO^-IO" 8 M E2 for 48 h. Total RNA was isolated and analyzed as described in Fig. 1. A, An autoradiograph of pS2 mRNA from MCF-7 K1 cells after treatment with E2. The mRNA content of each lane was normalized by the actin mRNA content. B, The levels of pS2 mRNA in MCF-7 K3 cells after the same treatments with E2. The mRNA content in each lane was normalized by the 36B4 mRNA content. The size of pS2 mRNA was 0.6 kb, that of 36B4 mRNA was 1.2 kb, and that of actin mRNA was 2.2 kb. The data in A and B are from separate experiments, using different preparations of radiolabeled pS2 cDNA probe and different times of autoradiogram development, so that information on control levels of pS2 mRNA in the two cell lines cannot be determined from these gels. In separate experiments in which pS2 mRNA from K1 and K3 cells was examined simultaneously using identical probe preparations for pS2 and 36B4, we determined that MCF-7 K3 cells contain 2.8 ± 0.4 (n = 4) times the level of pS2 mRNA found in MCF-7 K1 cells.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
Differential Regulation of Genes by Estrogen
1325
tions of E2 (10°° to 10--8 M) not only failed to increase expression of the pS2 mRNA, but markedly suppressed the pS2 mRNA level. Consequently, these results demonstrate that the expression of the pS2 gene in the estrogen growth-independent K3 subline is no longer up-regulated by estrogen. However, as shown below, pS2 mRNA in MCF-7 K3 cells is capable of being markedly up-regulated by other factors. Time Course of pS2 mRNA Response to Estrogen The time course of the change in pS2 mRNA measured after treatment with 10"9 M E2 is illustrated in Fig. 3. The pS2 mRNA content in the MCF-7 K1 cells showed a 5fold increase after 3 h of treatment with E2 and plateaued after 6 h at a level of 8-fold induction. Elevated levels were maintained until 48 h. In contrast, the level of pS2 mRNA in the MCF-7 K3 cells increased only transiently, then declined to below the initial control level. Effect of Estrogen on pS2 Run-On Transcription Rate To clarify the different effect of E2 on expression of the pS2 gene in MCF-7 K1 and K3 cells, the pS2 transcription rate was compared in both cells using a run-on transcription assay (Fig. 4). The pS2 transcription rate in MCF-7 K1 cells was dramatically increased (>7-fold) within 2 h by 10 9 M E2) before declining slightly to a rate still 4-fold elevated after 6-12 h of treatment. The pS2 transcription rate remained elevated until at least 24 h, when it was still neariy 3 times higher than that before E2 treatment. In contrast, the pS2 transcription rate in the MCF-7 K3 cells did not increase with the administration of E2. These results demonstrate that the different effects of E2 on pS2 mRNA levels in MCF7 K1 and K3 cells are at least partly explained by differential regulation at the transcriptional level.
10 20 30 40 50 Hours after estradiol treatment
60
Fig. 3. Time Course of Change in pS2 mRNA Levels after E2 Treatment of MCF-7 K1 and K3 Cell Sublines MCF-7 K1 and K3 cells were treated with 10 9 M E2 for 0 48 h. Total RNA was isolated and analyzed as described in Fig. 1. The pS2 mRNA content of samples was normalized by the 36B4 mRNA content. Values represent the mean of duplicate determinations from two separate experiments.
0
5
10
IS
20
25
Hours after estradiol treatment
Fig. 4. Effect of E2 on the Transcription Rate of pS2 mRNA MCF-7 K1 and K3 cells in T-150 flasks were treated with 10~9 M E2 or with ethanol vehicle alone for 0.5, 1, 2, 3, 6, 12, or 24 h before cells were harvested. [32P]UTP-labeled pS2 mRNA made from the run-on transcription assay (as described in Materials and Methods) was hybridized to 10 ^g pS2 cDNA cross-linked to a GeneScreen Plus membrane. Hybridization was performed at 55 C for 48 h. The resultant autoradiographs were quantified using Ultrascan XL laser densitometry. The pS2 transcription rate at each time point was normalized by the 36B4 transcription rate as a control for possible differences in incorporation of labeled uridine into nuclei. Basal pS2 transcription rates were similar in MCF-7 K1 and K3 cells. Values are expressed as the mean ± SE of at least three determinations.
Effects of Antiestrogen, Epidermal Growth Factor (EGF), and Protein Kinase Activators on pS2 mRNA Regulation In addition to estrogen, several mitogenic growth factors have recently been shown to modulate the pS2 mRNA content of MCF-7 K1 cells (14). Therefore, we examined the effects of EGF and protein kinase activators on pS2 mRNA levels as well as the effect of the antiestrogen ICI 164,384 (ICI; Fig. 5). With 10 7 M ICI, the basal pS2 mRNA level in both K1 and K3 sublines was suppressed. As seen in previous figures and shown here, 10 9 M E2 decreased the pS2 mRNA level in the K3 subline, while it increased pS2 mRNA in K1 cells, and the antiestrogen ICI suppressed this E2stimulated increase in pS2 mRNA in MCF-7 K1 cells. Treatment of MCF-7 K3 cells with 10 nM EGF decreased pS2 mRNA to 60% of the control level, while EGF increased the pS2 mRNA level 2-fold in MCF-7 K1 cells. 12-O-Tetradecanoyl phorbol-13-acetate (TPA), a direct stimulator of protein kinase-C (15), tremendously induced pS2 mRNA (>10-fold) in both K1 and K3 cells, and simultaneous treatment with cholera toxin (CT), which increases the intracellular cAMP concentration by activation of protein kinase-A, and 3-isobutyl-1methyl-xanthine (IBMX), which blocks cAMP phosphodiesterase activity and, hence, cAMP breakdown, likewise caused a dramatic increase in pS2 mRNA in both MCF-7 K1 and K3 cells. Effect of Estrogen and Antiestrogen on Progesterone Receptor (PR) mRNA Since PR in MCF-7 cells has been shown to be another target gene stimulated by estrogen (16), the expression
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 1326
Vol 5 No. 9
MCF-7 K1 MCF-7 K3 1 2
Control ICI
E,+ ICI
EGF
TPA
3
4
5
6
IBMX + CT
Fig. 5. Effects of Antiestrogen, EGF and Protein Kinase Activators on pS2 mRNA Regulation. MCF-7 K1 and K3 cells were treated as follows for 24 h: ethanol vehicle only; 1O~7 M ICI; 10"9 M E2; 10 9 M E2 plus 10-6 M ICI; and 10"8 M EGF; 10 7 M TPA; 10"4 M IBMX and CT (1 Mg/ml). Total RNA was isolated as described in Fig. 1. The amount of pS2 mRNA on each gel lane was evaluated after electrophoresis, autoradiography, and densitometry by comparison with the 36B4 mRNA content, which does not change in response to any of these treatments. Values represent the mean of two determinations from separate experiments.
of PR mRNA in K1 and K3 cells was compared (Fig. 6). PR mRNA was not detectable in either K1 or K3 control cells. However, 10~9 M E2 dramatically increased PR mRNA levels in both K1 and K3 cells (Fig. 6), giving the expected five species (17). In addition, accumulation of PR mRNA by 10~9 M E2 was totally abolished by 1 0 6 M ICI in both cells. Dose-response curves for E2 stimulation of PR (10"11 M half-maximal and 10 10 M E2 maximal; not shown) were also identical for the MCF-7 K1 and K3 cells. Thus, PR mRNA in the K3 subline is still markedly induced by E2, while pS2 mRNA is no longer up-regulated by estrogen in K3 cells, highlighting the differential regulation of these two genes here by E2.
Fig. 6. Effects of Estrogen and Antiestrogen on PR mRNA in MCF-7 K1 and K3 Cell Sublines MCF-7 K1 and K3 cells were treated with ethanol vehicle only, 1CT9 M E2, or 10~9 M E2 plus 10"6 M ICI for 72 h. Total RNA was isolated by guanidinium-phenol-chloroform extraction, and poly(A)+ RNA was further isolated from 1 mg total RNA, using a batch method described in Materials and Methods. Isolated poly(A)+ RNA was separated on a 1 % formaldehyde-denaturing gel and transferred to GeneScreen Plus membrane. 32P-Labeled PR cDNA probe was prepared by single primer labeling and isolated from the urea-denaturing polyacrylamide gel. Prehybridization and hybridization were performed at 65 C. The amount of PR mRNA was evaluated by comparison with the content of 36B4 mRNA (not shown) that was very similar in the six samples.
Down-Regulation of ER mRNA by Estrogen The ER mRNA and ER protein in MCF-7 K1 cells are known to be down-regulated by estrogen (18, 19). In this study we observed that ER mRNA in the MCF-7 K3 cells was also down-regulated by estrogen (Fig. 7). While 10 12 and 10 11 M E2 did not suppress ER mRNA significantly, higher concentrations (10 10 -10 8 M) reduced ER mRNA levels to 30-50% of the control level, the same extent of reduction as that achieved by E2 in MCF-7 K1 cells (18).
showed a somewhat slower response to E2 treatment, with a maximum ER transcription rate suppression by 3-6 h, but the magnitude of suppression was similar to that in K1 cells, namely a decrease to approximately 50% of the control level. By 12 h of exposure to E2, the ER transcription rate in the K3 subline returned to or slightly exceeded the initial control level. At this time, the ER transcription rate in the MCF-7 K1 cells had also returned to its initial level.
Effect of Estrogen on ER Run-On Transcription Rate
DISCUSSION
As shown in Fig. 8, the transcription rate of the ER gene was reduced by E2 in both the MCF-7 K1 and K3 cell sublines. In MCF-7 K1 cells, maximum ER transcription rate suppression (by ~60%) was seen within 2 h of E2 treatment (Fig. 8). This was followed by a return to near the initial rate by 6-12 h. The MCF-7 K3 subline
Although the pS2 gene was discovered in the process of screening genes that showed a rapid response to the addition of E2 in MCF-7 cells, further studies have shown that its regulation is complex, being regulated by estrogen as well as other factors (14, 20). We have investigated here whether the regulation by estrogen
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
1327
(% control)
Differential Regulation of Genes by Estrogen
120
100-
A
Iptlon
S
5 80-
n
40-
fti >
Hyeseong Cho, Peter A. NG, and Benita S. Katzenellenbogen Department of Physiology and Biophysics University of Illinois University of Illinois College of Medicine Urbana, Illinois 61801
We have examined the ability of estradiol (E2) to regulate the expression of three mRNAs [for pS2, progesterone receptor (PR), and estrogen receptor (ER)], known to be under E2 regulation in the parental E2 growth-responsive MCF-7 cells, in an E2 growthindependent MCF-7 cell subline. This subline (denoted MCF-7 K3), previously isolated from the parental estrogen-dependent MCF-7 K1 human breast cancer cells after long term growth in vitro in the absence of estrogen, acquired estrogen-independent growth in vitro as well as the ability to form tumors in nude mice in vivo without estrogen. We find that the content of pS2 mRNA and the transcription rate of the pS2 gene, while being markedly increased by E2 in MCF-7 K1 cells, are no longer stimulated by E2 in this subline, although protein kinase activators tremendously increase (> 10-fold) pS2 mRNA in both K1 and K3 cells. In fact, basal pS2 mRNA levels are elevated 2.8 ± 0.4-fold in MCF7 K3 cells, and E2 evokes a concentration-dependent suppression of the pS2 mRNA level. In contrast, PR mRNA in the K3 subline, as in the parental K1 cells, is still up-regulated by E2, and ER mRNA content and the ER mRNA transcription rate are still downregulated by E2 and show normal E2 dose-response relationships, implying that the ER in this subline is functional. These results demonstrate that the progression to estrogen-independent growth in K3 cells is accompanied by a change in the regulation of some estrogen-induced genes by estrogen. While PR and ER retain normal patterns of regulation by E2, the pS2 gene in the estrogen growth-independent K3 subline is differentially affected and is no longer stimulated by E2. Our data suggest that this altered regulation of the pS2 gene is probably not caused by a defect of the ER or ER regulation in this subline. (Molecular Endocrinology 5: 1323-1330, 1991) 0888-8809/91/1323-1330$03.00/0 Molecular Endocrinology Copyright © 1991 by The Endocrine Society
INTRODUCTION
A large proportion of breast cancer is regulated by sex steroid hormones, and the responsiveness of breast cancers to these steroid hormones is highly correlated with the presence of hormone receptors (1, 2). Hence, estrogen receptor (ER) status in breast cancer is used as an important indicator of endocrine therapy (3). During endocrine therapy, a progression from hormone dependence to independence often occurs, leading to resistance to hormonal therapy (4, 5). The ER-positive MCF-7 human breast cancer cell line (designated MCF7 K1) shows estrogen-dependent growth in vitro as well as estrogen-dependent tumorigenicity in vivo. We have previously isolated a subline of the MCF-7 cells (designated MCF-7 K3), which acquired estrogen-independent growth in vitro after prolonged growth in the absence of estrogen (6). In contrast to the parental MCF-7 K1 cells, this MCF-7 K3 subline is also capable of forming tumors in ovariectomized athymic nude mice without estrogen stimulation (5). Thus, this subline provides a unique system for studying the progression from hormone dependence to independence in human breast cancer cells and the changes that accompany this progression. In the present study we have compared the regulation of three estrogen-modulated genes in the estrogen growth-independent MCF-7 K3 subline: namely, pS2, ER, and progesterone receptor (PR). Since the expression of the pS2 gene is under the direct control of estrogen in MCF-7 cells (7-9) and is rapidly up-regulated by estrogen, its expression is a useful marker for investigating estrogen responsiveness in breast cancer cells. Although a portion of the primary structure of the pS2 protein resembles epidermal growth factor (10), and pS2 shows considerable homology to pancreatic spasmolytic polypeptide (11), the function of the pS2 protein remains unknown. The expression of pS2 mRNA is found in most ER-positive breast cancer cells, but not in estrogen-insensitive breast cancer cells such
1323
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 1324
Vol 5 No. 9
as T47D cells. Ninety-six percent of human breast tumors expressing the pS2 protein also contain ER (12). The presence of the pS2 protein is also considered an important variable for prediction of the relapse and survival rate in breast cancer patients (13). While pS2 gene expression is markedly stimulated by estrogen in wild-type parental MCF-7-K1 cells, we found that pS2 mRNA in the K3 subline is not induced, but, rather, is suppressed by estrogen. Similarly, the transcription rate of the pS2 gene in the MCF-7 K3 subline is not accelerated by estrogen. Interestingly, and in contrast, PR mRNA and ER mRNA levels in the K3 subline are still under the usual control of estrogen, implying that the ER in the K3 subline is biologically functional.
levels differed, with K3 cells containing 2.8 ± 0.4 times more pS2 mRNA (n = 4 experiments) than that found in MCF-7 K1 cells (data not shown). In these comparisons, 36B4 mRNA was used as an internal control, with pS2 being probed simultaneously in MCF-7 K1 and K3 cell RNAs in each experiment. The dose-response effect of E2 on pS2 mRNA levels in MCF-7 K1 and K3 cells was then examined. As shown in Fig. 1, 10"10 M E2 increased the pS2 mRNA level in the MCF-7 K1 cells nearly 10-fold. Somewhat lower stimulation was observed with 10 9 and 10~8 M E2 (Fig. 2A), although pS2 mRNA levels were still several-fold higher than the control values. In the MCF-7 K3 subline, 10 12 and 10 11 M E2 had no effect on pS2 mRNA levels (Figs. 1 and 2B), while higher concentra-
A
RESULTS PS2 mRNA Regulation by Estrogen in MCF-7 K1 and K3 Cell Sublines: Estradiol (E2) Dose Response The regulation of the pS2 mRNA level in MCF-7 K1 and K3 cells was examined by Northern blot analysis after 48-h exposure of cells to increasing concentrations of E2 (Fig. 1). Cells were switched to medium without phenol red and containing charcoal-dextran-stripped calf serum (CDCS) to eliminate estrogenic sources before treatment. We first compared the relative content of pS2 mRNA in control estrogen-withdrawn MCF-7 K1 and K3 cells and observed that the basal pS2 mRNA
Actin
pS2 C 101410121010109108ME2
pS2
Control
-12
-11
-10
-9
-8
Estradiol concentration (log M)
Fig. 1. Dose Response of pS2 mRNA to E2 in MCF-7 K1 and K3 Cell Sublines MCF-7 cell sublines were treated with 1O-12-1O** M E2. Total RNA was isolated using a guanidinium thiocyanate-phenolchloroform extraction method (see Materials and Methods). Twenty micrograms of total RNA were separated on 1.4% formaldehyde-denaturing agarose gel and transferred to GeneScreen Plus. Hybridization was performed with a 0.3-kb 32 P-labeled pS2 cDNA probe at 65 C. Autoradiographs were quantified using laser densitometry. The amount of mRNA loaded on the gel was normalized to the 36B4 or actin mRNA content. The same fold change in pS2 mRNA level was observed whether pS2 was quantitated relative to actin or 36B4 as the internal standard. Values represent the mean and SE of two to four determinations.
Fig. 2. A and B, Northern Blot Analysis of pS2 mRNA in MCF7 K1 and MCF-7 K3 Cell Sublines in Response to Various Concentrations of E2 MCF-7 K1 and K3 cells were treated with lO^-IO" 8 M E2 for 48 h. Total RNA was isolated and analyzed as described in Fig. 1. A, An autoradiograph of pS2 mRNA from MCF-7 K1 cells after treatment with E2. The mRNA content of each lane was normalized by the actin mRNA content. B, The levels of pS2 mRNA in MCF-7 K3 cells after the same treatments with E2. The mRNA content in each lane was normalized by the 36B4 mRNA content. The size of pS2 mRNA was 0.6 kb, that of 36B4 mRNA was 1.2 kb, and that of actin mRNA was 2.2 kb. The data in A and B are from separate experiments, using different preparations of radiolabeled pS2 cDNA probe and different times of autoradiogram development, so that information on control levels of pS2 mRNA in the two cell lines cannot be determined from these gels. In separate experiments in which pS2 mRNA from K1 and K3 cells was examined simultaneously using identical probe preparations for pS2 and 36B4, we determined that MCF-7 K3 cells contain 2.8 ± 0.4 (n = 4) times the level of pS2 mRNA found in MCF-7 K1 cells.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
Differential Regulation of Genes by Estrogen
1325
tions of E2 (10°° to 10--8 M) not only failed to increase expression of the pS2 mRNA, but markedly suppressed the pS2 mRNA level. Consequently, these results demonstrate that the expression of the pS2 gene in the estrogen growth-independent K3 subline is no longer up-regulated by estrogen. However, as shown below, pS2 mRNA in MCF-7 K3 cells is capable of being markedly up-regulated by other factors. Time Course of pS2 mRNA Response to Estrogen The time course of the change in pS2 mRNA measured after treatment with 10"9 M E2 is illustrated in Fig. 3. The pS2 mRNA content in the MCF-7 K1 cells showed a 5fold increase after 3 h of treatment with E2 and plateaued after 6 h at a level of 8-fold induction. Elevated levels were maintained until 48 h. In contrast, the level of pS2 mRNA in the MCF-7 K3 cells increased only transiently, then declined to below the initial control level. Effect of Estrogen on pS2 Run-On Transcription Rate To clarify the different effect of E2 on expression of the pS2 gene in MCF-7 K1 and K3 cells, the pS2 transcription rate was compared in both cells using a run-on transcription assay (Fig. 4). The pS2 transcription rate in MCF-7 K1 cells was dramatically increased (>7-fold) within 2 h by 10 9 M E2) before declining slightly to a rate still 4-fold elevated after 6-12 h of treatment. The pS2 transcription rate remained elevated until at least 24 h, when it was still neariy 3 times higher than that before E2 treatment. In contrast, the pS2 transcription rate in the MCF-7 K3 cells did not increase with the administration of E2. These results demonstrate that the different effects of E2 on pS2 mRNA levels in MCF7 K1 and K3 cells are at least partly explained by differential regulation at the transcriptional level.
10 20 30 40 50 Hours after estradiol treatment
60
Fig. 3. Time Course of Change in pS2 mRNA Levels after E2 Treatment of MCF-7 K1 and K3 Cell Sublines MCF-7 K1 and K3 cells were treated with 10 9 M E2 for 0 48 h. Total RNA was isolated and analyzed as described in Fig. 1. The pS2 mRNA content of samples was normalized by the 36B4 mRNA content. Values represent the mean of duplicate determinations from two separate experiments.
0
5
10
IS
20
25
Hours after estradiol treatment
Fig. 4. Effect of E2 on the Transcription Rate of pS2 mRNA MCF-7 K1 and K3 cells in T-150 flasks were treated with 10~9 M E2 or with ethanol vehicle alone for 0.5, 1, 2, 3, 6, 12, or 24 h before cells were harvested. [32P]UTP-labeled pS2 mRNA made from the run-on transcription assay (as described in Materials and Methods) was hybridized to 10 ^g pS2 cDNA cross-linked to a GeneScreen Plus membrane. Hybridization was performed at 55 C for 48 h. The resultant autoradiographs were quantified using Ultrascan XL laser densitometry. The pS2 transcription rate at each time point was normalized by the 36B4 transcription rate as a control for possible differences in incorporation of labeled uridine into nuclei. Basal pS2 transcription rates were similar in MCF-7 K1 and K3 cells. Values are expressed as the mean ± SE of at least three determinations.
Effects of Antiestrogen, Epidermal Growth Factor (EGF), and Protein Kinase Activators on pS2 mRNA Regulation In addition to estrogen, several mitogenic growth factors have recently been shown to modulate the pS2 mRNA content of MCF-7 K1 cells (14). Therefore, we examined the effects of EGF and protein kinase activators on pS2 mRNA levels as well as the effect of the antiestrogen ICI 164,384 (ICI; Fig. 5). With 10 7 M ICI, the basal pS2 mRNA level in both K1 and K3 sublines was suppressed. As seen in previous figures and shown here, 10 9 M E2 decreased the pS2 mRNA level in the K3 subline, while it increased pS2 mRNA in K1 cells, and the antiestrogen ICI suppressed this E2stimulated increase in pS2 mRNA in MCF-7 K1 cells. Treatment of MCF-7 K3 cells with 10 nM EGF decreased pS2 mRNA to 60% of the control level, while EGF increased the pS2 mRNA level 2-fold in MCF-7 K1 cells. 12-O-Tetradecanoyl phorbol-13-acetate (TPA), a direct stimulator of protein kinase-C (15), tremendously induced pS2 mRNA (>10-fold) in both K1 and K3 cells, and simultaneous treatment with cholera toxin (CT), which increases the intracellular cAMP concentration by activation of protein kinase-A, and 3-isobutyl-1methyl-xanthine (IBMX), which blocks cAMP phosphodiesterase activity and, hence, cAMP breakdown, likewise caused a dramatic increase in pS2 mRNA in both MCF-7 K1 and K3 cells. Effect of Estrogen and Antiestrogen on Progesterone Receptor (PR) mRNA Since PR in MCF-7 cells has been shown to be another target gene stimulated by estrogen (16), the expression
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
MOL ENDO-1991 1326
Vol 5 No. 9
MCF-7 K1 MCF-7 K3 1 2
Control ICI
E,+ ICI
EGF
TPA
3
4
5
6
IBMX + CT
Fig. 5. Effects of Antiestrogen, EGF and Protein Kinase Activators on pS2 mRNA Regulation. MCF-7 K1 and K3 cells were treated as follows for 24 h: ethanol vehicle only; 1O~7 M ICI; 10"9 M E2; 10 9 M E2 plus 10-6 M ICI; and 10"8 M EGF; 10 7 M TPA; 10"4 M IBMX and CT (1 Mg/ml). Total RNA was isolated as described in Fig. 1. The amount of pS2 mRNA on each gel lane was evaluated after electrophoresis, autoradiography, and densitometry by comparison with the 36B4 mRNA content, which does not change in response to any of these treatments. Values represent the mean of two determinations from separate experiments.
of PR mRNA in K1 and K3 cells was compared (Fig. 6). PR mRNA was not detectable in either K1 or K3 control cells. However, 10~9 M E2 dramatically increased PR mRNA levels in both K1 and K3 cells (Fig. 6), giving the expected five species (17). In addition, accumulation of PR mRNA by 10~9 M E2 was totally abolished by 1 0 6 M ICI in both cells. Dose-response curves for E2 stimulation of PR (10"11 M half-maximal and 10 10 M E2 maximal; not shown) were also identical for the MCF-7 K1 and K3 cells. Thus, PR mRNA in the K3 subline is still markedly induced by E2, while pS2 mRNA is no longer up-regulated by estrogen in K3 cells, highlighting the differential regulation of these two genes here by E2.
Fig. 6. Effects of Estrogen and Antiestrogen on PR mRNA in MCF-7 K1 and K3 Cell Sublines MCF-7 K1 and K3 cells were treated with ethanol vehicle only, 1CT9 M E2, or 10~9 M E2 plus 10"6 M ICI for 72 h. Total RNA was isolated by guanidinium-phenol-chloroform extraction, and poly(A)+ RNA was further isolated from 1 mg total RNA, using a batch method described in Materials and Methods. Isolated poly(A)+ RNA was separated on a 1 % formaldehyde-denaturing gel and transferred to GeneScreen Plus membrane. 32P-Labeled PR cDNA probe was prepared by single primer labeling and isolated from the urea-denaturing polyacrylamide gel. Prehybridization and hybridization were performed at 65 C. The amount of PR mRNA was evaluated by comparison with the content of 36B4 mRNA (not shown) that was very similar in the six samples.
Down-Regulation of ER mRNA by Estrogen The ER mRNA and ER protein in MCF-7 K1 cells are known to be down-regulated by estrogen (18, 19). In this study we observed that ER mRNA in the MCF-7 K3 cells was also down-regulated by estrogen (Fig. 7). While 10 12 and 10 11 M E2 did not suppress ER mRNA significantly, higher concentrations (10 10 -10 8 M) reduced ER mRNA levels to 30-50% of the control level, the same extent of reduction as that achieved by E2 in MCF-7 K1 cells (18).
showed a somewhat slower response to E2 treatment, with a maximum ER transcription rate suppression by 3-6 h, but the magnitude of suppression was similar to that in K1 cells, namely a decrease to approximately 50% of the control level. By 12 h of exposure to E2, the ER transcription rate in the K3 subline returned to or slightly exceeded the initial control level. At this time, the ER transcription rate in the MCF-7 K1 cells had also returned to its initial level.
Effect of Estrogen on ER Run-On Transcription Rate
DISCUSSION
As shown in Fig. 8, the transcription rate of the ER gene was reduced by E2 in both the MCF-7 K1 and K3 cell sublines. In MCF-7 K1 cells, maximum ER transcription rate suppression (by ~60%) was seen within 2 h of E2 treatment (Fig. 8). This was followed by a return to near the initial rate by 6-12 h. The MCF-7 K3 subline
Although the pS2 gene was discovered in the process of screening genes that showed a rapid response to the addition of E2 in MCF-7 cells, further studies have shown that its regulation is complex, being regulated by estrogen as well as other factors (14, 20). We have investigated here whether the regulation by estrogen
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 26 May 2015. at 14:10 For personal use only. No other uses without permission. . All rights reserved.
1327
(% control)
Differential Regulation of Genes by Estrogen
120
100-
A
Iptlon
S
5 80-
n
40-
fti >
