| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Dr. Margarete Fischer-Bosch Institute of Clinical Pharmacology (M.B.B.), and Robert Bosch Hospital (M.B.B., C.K.), Department of Laboratory Medicine, 70376 Stuttgart, Germany; and Institute of Cell Biology and Immunology (K.P.), University of Stuttgart, 70569 Stuttgart, Germany
Address all correspondence and requests for reprints to: Cornelius Knabbe, MD, Department of Laboratory Medicine, Robert Bosch Hospital, Auerbachstrasse 110, 70376 Stuttgart, Germany. E-mail: cornelius.knabbe{at}rbk.de.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
We used specific MAPK inhibitors and MCF-7 carcinoma cells as a model to demonstrate that p38 MAPK is an important mediator of antiestrogen growth inhibition in breast cancer. A kinase assay showed that antiestrogens (4-hydroxytamoxifen and ICI 182.780) rapidly induce p38 activity. Overexpression of kinase-deficient mutants of p38 reduced the antiestrogen suppression of Cyclin A transcription.
TGFß, a negative regulator of breast cancer cell growth, is induced by antiestrogens; therefore, activation of p38 could have been mediated by TGFß. We used a TGFß and antiestrogen-sensitive reporter gene assay to show that p38 activation precedes TGFß activation. These results were further confirmed by quantitative RT-PCR analysis of the antiestrogen-induced transcription of TGFß2 and TGFß receptor II. Inhibition of p38 reduced the induction of both genes.
Finally, Western blot analysis shows that antiestrogens induce phosphorylation of Smad (mothers against decapentaplegic homolog) 2 via p38. Promoter assays with the Smad-dependent reporter p6SBE confirm participation of Smad3 and Smad4 in antiestrogen action.
Taken together, our data delineate an antiestrogen signal transduction pathway involving sequential activation of p38 and TGFß pathways to mediate growth inhibition.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The tumoristatic effect of antiestrogens is based on their ability to compete with estradiol for binding to the estrogen receptor (ER). Binding of antiestrogens induces conformational changes in the receptor that differ from those induced by estradiol (4). These changes lead to partial or complete inactivation of the receptor’s physiological transcriptional activation functions (TAF) and the recruitment of specific transcriptional coactivators or corepressors (5, 6). Tamoxifen inhibits solely TAF-2 and can act as an agonist when TAF-1 is needed for promoter activity. ICI 182.780 inhibits both TAFs and acts as a pure antiestrogen, independent of promoter context (7).
Antiestrogen-induced changes in cellular signaling downstream of the ER, which eventually cause inhibition of tumor growth, are less well characterized. We have previously shown that one important antiestrogen effect is an ER-dependent activation of the TGFß system. Treatment of breast cancer cells with antiestrogens leads to a release of active TGFß1 from a latent precursor molecule and to an increased transcription of TGFß2 and TGFß receptor type II (TßRII) (8, 9, 10, 11).
TGFß is a potent negative regulator of epithelial cell proliferation (12). Three isoforms of TGFß, TGFß 1, 2, and 3, have been described. TGFß1 and TGFß2 are equally effective in inhibiting breast cancer cell growth (13).
TGFß signals are mediated by specific transmembrane receptors. TGFß receptor type I (TßRI) and TßRII are serine-threonine kinases, which form a heteromeric signal transduction complex upon ligand binding. TßRII phosphorylates TßRI, which activates TßRI kinase and initiates downstream signaling (14). Intracellular TGFß signaling is complex, and many different pathways can be activated. These include the Smad (mothers against decapentaplegic homolog) pathway, MAPK pathways, phosphoinositol-3-kinase, and PP2A (15, 16, 17, 18, 19).
The Smad pathway appears to be the major TGFß signal transduction pathway. Smad2 and Smad3 are phosphorylated by the active TGFß receptor complex. Activated Smad2 or Smad3 form complexes with Smad4. The resulting heteromeric Smad complex is translocated to the nucleus where it binds to Smad-specific promoter elements and regulates transcription of TGFß-responsive genes (17).
MAPKs are involved in cellular signal transduction in response to various stimuli. So far, four different MAPKs have been described: the ERKs, the c-jun N-terminal kinases (JNKs), the p38 MAPKs, and the ERK5 or big MAPK 1. The ERK MAPKs are preferentially activated by mitogens, whereas the JNK and p38 MAPKs are responsive to stress and inflammatory signals (20). ERK, JNK, and p38 MAPKs have been implicated in TGFß signal transduction (15, 18, 21). Tamoxifen was shown to activate JNK and p38 MAPKs in connection with programmed cell death (22, 23). However, tamoxifen-induced apoptosis occurs only at high concentrations and seems to be ER independent as it is not reversible by addition of estradiol (24).
To gain a better understanding of the growth-inhibitory effects of antiestrogens and the mechanisms leading to antiestrogen resistance, we investigated the involvement and interactions of MAPK and TGFß signal transduction pathways in antiestrogen-induced inhibition of breast cancer cell proliferation. We show that p38 MAPK is an important mediator of antiestrogen-induced growth inhibition and that p38 activation is required for antiestrogen induction of TGFß signal pathways.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
After 5 d, growth of cells treated with 4-hydroxytamoxifen was reduced to 50% as compared with cells treated with vehicle alone (Fig. 1A
). Growth inhibition was completely reversed by addition of 10 nM estradiol, ensuring the ER dependence of the antiestrogen effect. Estradiol by itself had no stimulatory or inhibitory effect on MCF-7 cell growth. In the presence of the specific p38 kinase inhibitors, SB203580 and SB202190, the inhibitory effect of 4-hydroxytamoxifen was almost completely abolished, resulting in nearly normal growth (85% and 87% of untreated controls, respectively). Reversal of the growth-inhibitory effect of 4-hydroxytamoxifen was highly significant for both p38 inhibitors (P < 0.0001). The MEK/ERK inhibitor PD98059 had no effect, whereas cotreatment with U0126 significantly enhanced the growth-inhibitory effect of 4-hydroxytamoxifen. The inactive control substances, SB202474 and U0124, had no influence on the growth-inhibitory effect of 4-hydroxytamoxifen (Fig. 1A). The continuous activity of SB203580 and PD98059 over the entire treatment period was verified by analysis of their effect on phosphorylation of specific p38 or MEK substrates (Elk-1 or p44/42, respectively; Fig. 1E).
|
). The growth-inhibitory effect was reversed by addition of estradiol. Again, treatment with p38 inhibitors significantly reverted the growth-inhibitory effect of ICI 182.780 although only a partial abrogation of growth inhibition was observed, reaching 24% and 26% of untreated control cells after cotreatment with SB203580 and SB202190, respectively (P < 0.0001). MEK/ERK inhibitors as well as the inactive control substances had no effect (Fig. 1B).
The results were confirmed in T47D cells, representing another ER-positive, hormone-responsive breast cancer cell line. T47D cells were less sensitive to antiestrogen growth inhibition than MCF-7 cells. After 5 d of treatment with 4-hydroxytamoxifen or ICI 182.780, cell growth was reduced to 73% or 53% of control, respectively (Fig. 1, C and D).
Cotreatment with SB203580 significantly reduced the growth-inhibitory effect of both antiestrogens as compared with controls (4-hydroxytamoxifen + SB203580: 88%, P = 0.0121; ICI 182.780 + SB203580: 74%, P = 0.0002). A significant reduction of the growth-inhibitory effect of both antiestrogens was also observed for SB202190 at a concentration of 5 µM. Inhibition by 4-hydroxytamoxifen was reduced to 88% of control (P = 0.0181), and inhibition by ICI 182.780 was reduced to 75% (P = 0.0011). At a concentration of 10 µM the growth-inhibitory effect of ICI 182.780 was nearly completely abolished (96% of control; P < 0.0001).
Both tamoxifen and ICI 182.780 have been previously shown to induce apoptosis at higher concentrations (22, 23, 24). Fluorescence-activated cell sorting analyses of MCF-7 cells labeled with propidium iodide indicate that the antiestrogen concentrations used in our experiments led to growth arrest and did not induce cell death. The observed reductions in cell number were caused by accumulation of cells in G1 phase (Table 1).
|
Antiestrogens Activate p38 Kinase in Breast Cancer Cells
To determine whether antiestrogens activate p38 kinase, MCF-7 cells were treated with 4-hydroxytamoxifen or ICI 182.780 for different time periods, and phosphorylated p38 was immunoprecipitated with a specific antibody. p38 activity in the precipitate was determined by an in vitro kinase assay using glutathione-S-transferase-activating transcription factor 2 (ATF-2) as substrate followed by Western blot detection of phosphorylated ATF-2. As shown in Fig. 2, treatment with antiestrogens induced a rapid and transient activation of p38 kinase activity, peaking around 1–1.5 h after stimulation with 4-hydroxytamoxifen (Fig. 2A) or ICI 182.780 (Fig. 2B) and declining to pretreatment levels after 2 h. In contrast, treatment with estradiol did not induce p38 kinase activity (Fig. 2C), underlining the specificity of the antiestrogen effect.
|
).
Antiestrogen Inhibition of Cyclin A Is p38 Dependent
The reduction of Cyclin A mRNA expression is an event that can be detected shortly after antiestrogen treatment. We used kinase-deficient p38
or p38ß mutants (p38AGF or p38ßAGF) to investigate whether p38 was involved in the antiestrogen regulation of Cyclin A. Cells were treated with antiestrogens and transiently transfected with the respective expression vectors. RNA was isolated 24 h after transfection, and Cyclin A expression was analyzed by quantitative LightCycler RT-PCR. In MCF-7 cells transfected with empty vector (pUB6), Cyclin A mRNA was reduced to 61% of control after treatment with 4-hydroxytamoxifen and to 48% of control after treatment with ICI 182.780. Expression of p38AGF or p38ßAGF completely abolished the reduction of Cyclin A mRNA by 4-hydroxytamoxifen (Fig. 3A) and partially reverted the inhibitory effect of ICI 182.780 (70% and 63% of control for p38AGF or p38ßAGF, respectively, Fig. 3B).
|
AGF or p38ßAGF (data not shown).
These results strengthen the observations made in the proliferation assays described above. They show that the two isoforms of p38 that are inhibited by SB203580, namely p38 and p38ß (26), are involved in antiestrogen growth inhibition. Interestingly, the growth-inhibitory effect of ICI 182.780 was only partially reverted in both assays, pointing to the participation of other pathways in the growth-inhibitory effect of the pure antiestrogen.
Antiestrogen Activation of p38 Kinase Is Not Mediated by TGFß
Antiestrogens are potent activators of TGFß (9, 10). In recent studies p38 kinase pathways were shown to regulate gene expression in response to TGFß (15). We therefore investigated whether the observed activation of a p38 pathway after antiestrogen treatment was upstream or downstream of TGFß induction. A reporter gene assay was used to address this question. The reporter plasmid p3TP-lux (27), which contains activator protein 1 as well as Smad binding elements, is TGFß sensitive and has been extensively used to study TGFß-dependent effects.
TGFß sensitivity of the reporter construct p3TP-lux was verified in the MCF-7 cells studied here, revealing a 3.4-fold activation by treatment with 100 pM TGFß1 (Fig. 4A). In the same system, both 4-hydroxytamoxifen and ICI 182.780 strongly induced p3TP-lux reporter gene activity, yielding a 7.5-fold and a 11.2-fold induction, respectively, as compared with cells treated with vehicle alone (Fig. 4, B and C).
|
).
Likewise, ICI 182.780-induced activation of p3TP-lux appeared strongly reduced upon overexpression of dKTßRI and dKTßRII, respectively (Fig. 4C). However, due to a high variability in p3TP-lux induction in the control cells transfected with empty expression vector, statistical significance was not reached in this experimental group.
Taken together, these results indicate that antiestrogens are able to activate p3TP-lux and that this activation is mediated, at least partially, through activation of TGFß signal transduction pathways.
Next we examined whether antiestrogen-induced activation of p3TP-lux was p38 kinase dependent. MCF-7 cells were treated with antiestrogens in the presence or absence of the p38 kinase inhibitor SB203580 and transiently transfected with p3TP-lux. Blockade of p38 kinase activity strongly interfered with the antiestrogen-dependent activation of the TGFß reporter plasmid, with a reduction of the 4-hydroxytamoxifen response from 3.9-fold to 1.8-fold (P = 0.0078) upon cotreatment with SB203580 (Fig. 4D), and an even more pronounced inhibition in the case of ICI 182.780 from a 5.1-fold to a 1.5-fold activation (P = 0.0271, Fig. 4E). Of note, TGFß-induced activation of p3TP-lux was not attenuated by inhibition of p38 kinase with SB203580 (Fig. 4A), placing p38 upstream of TGFß signals. In addition, the results imply that the SB203580 concentration used in the experiments (10 µM) did not influence TGFß receptor kinase (28), as TGFß induction of reporter gene activity was even enhanced in the presence of the inhibitor (Fig. 4A).
p38 Kinase Activity Is Needed for the Antiestrogen Induction of TGFß2 and TßRII
We have shown previously that treatment of breast cancer cells with antiestrogens induces the transcription of TGFß2 and TßRII (9, 10, 11). The above data indicate that p38 plays a role in induction of TGFß signal pathways in response to antiestrogens. To evaluate the contribution of p38 kinase in antiestrogen induction of TGFß2 and TßRII, MCF-7 cells were treated with antiestrogens in the presence or absence of p38 kinase inhibitor SB203580, and total RNA was isolated from the cells. Quantitative LightCycler RT-PCRs were used to analyze the expression of TGFß2 and TßRII, with GAPDH serving as control.
Treatment with 4-hydroxytamoxifen resulted in an 18-fold induction of TGFß2 and 2-fold induction of TßRII expression (Fig. 5A). Inhibition of p38 kinase activity nearly completely abolished the induction of both mRNAs. In the presence of SB203580, 4-hydroxytamoxifen induction of TGFß2 and TßRII was only 2-fold (P = 0.0027) and 1.4-fold (P = 0.0069), respectively (Fig. 5A).
|
). GAPDH expression was slightly reduced by both antiestrogens; however, no further modulation of GAPDH expression was observed after cotreatment with antiestrogens and SB203580 (data not shown).
Taken together, these results suggest that a p38 kinase-dependent pathway is needed for the antiestrogen-mediated induction of TGFß2 and TßRII.
Antiestrogen-Induced TGFß Effects Are Mediated by a Smad-Dependent Pathway
TGFß has been shown to activate many different signal transduction pathways (15, 16, 17, 18, 19). The canonical TGFß pathway is mediated by Smad2 and Smad3, which are phosphorylated by the active TGFß receptor complex and by Smad4, which heterodimerizes with activated Smad2/3 (17). To establish whether a Smad pathway participates in antiestrogen signaling, we investigated the involvement of Smad2, Smad3, and Smad4.
Smad2 is directly phosphorylated by activated TßRI. The TGFß-dependent phosphorylation of Smad2 in the MCF-7 cells used in this study was verified by treatment with 100 pM TGFß1 for 3 h, which induced a 3-fold increase in phosphorylation (Fig. 6A).
|
, B and C). However, the inhibitor had no effect on the TGFß-induced phosphorylation (Fig. 6A), supporting the notion that p38 kinase is needed for the antiestrogen induction of TGFß. Furthermore, the results confirm that TGFß receptor activity and signal transduction are not influenced by the p38 inhibitor.
Of note, the expression level of Smad2 was strongly influenced by treatment: TGFß, 4-hydroxytamoxifen, and ICI 182.780 induced a distinct down-regulation of Smad2 (Fig. 6). Consequently, induction of Smad2 phosphorylation by antiestrogens becomes even more pronounced, when normalized for Smad2 expression levels (i.e. 
40-fold for both antiestrogens).
To address the effect of antiestrogens on a Smad-dependent promoter, we examined the induction of p6SBE-luc, a reporter vector that contains six copies of a Smad-binding element (29). p6SBE-luc was activated 1.7-fold or 2.1-fold after treatment with 4-hydroxytamoxifen or ICI 182.780, respectively. Cotransfections of p6SBE-luc with expression vectors for Smad3 or Smad4 significantly enhanced the antiestrogen activation of the reporter (Fig. 7, A and B). Smad4 had a slightly stronger effect than Smad3, as Smad3 also increased p6SBE activity in control cells treated with vehicle alone. Cotransfection of Smad4 increased activation of p6SBE-luc 4.3-fold (P < 0.0001) after treatment with 4-hydroxytamoxifen and 6.1-fold (P = 00079) after treatment with ICI 182.780, whereas Smad3 led to an 2.9-fold (P = 0.0051) or 5.3-fold (P = 00079) induction after treatment with 4-hydroxytamoxifen or ICI 182.780, respectively (Fig. 7, A and B).
|
, A and B) (30).
Next we examined whether antiestrogen-induced activation of p6SBE-luc was p38 kinase dependent. MCF-7 cells were cotransfected with Smad4 to achieve a high level of activation. Before transfection cells were treated with antiestrogens in the presence or absence of the p38 kinase inhibitor SB203580. Inhibition of p38 kinase activity strongly interfered with the antiestrogen-dependent activation of the 6SBE-promoter, with a reduction of the 4-hydroxytamoxifen response from 2.9-fold to 1.5-fold (P = 0.0131) upon cotreatment with the inhibitor (Fig. 7C), and from 3.6-fold to 1.9-fold in the case of ICI 182.780 (Fig. 7D).
To confirm that a Smad pathway is also involved in the antiestrogen activation of the more complex 3TP-promoter, cotransfections were carried out with p3TP-lux and a dominant negative mutant of Smad4 (Smad4dM), which lacks the Smad-dependent activation domain (31). Overexpression of Smad4dM attenuated the antiestrogen-induced activation of p3TP-lux by 4-hydroxytamoxifen from 7.4-fold to 4.4-fold and by ICI 182.780 from 7.3-fold to 5.1-fold (Fig. 8, A and B).
|
). Taken together, the results show that Smad2, Smad3, and Smad4 participate in antiestrogen signal transduction and that p38 is needed for antiestrogen activation of this pathway.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The purpose of this study was to investigate the involvement of MAPK and TGFß signal transduction pathways in partial (tamoxifen) and pure (ICI 182.780) antiestrogen-induced growth inhibition to obtain a better understanding of the mechanisms of antiestrogen resistance and to establish a basis for the development of new diagnostic and therapeutic strategies.
In place of tamoxifen, we used the metabolite 4-hydroxytamoxifen, which proved to be more potent than tamoxifen in inhibiting breast cancer cell growth (25).
Our results with specific kinase inhibitors indicate that MEK/ERK activity slightly interferes with antiestrogen growth inhibition. This is in accordance with recent results showing that activation of ERK-dependent signal transduction pathways is involved in antiestrogen resistance (32).
p38 MAPK, on the other hand, was required for 4-hydroxytamoxifen- and ICI 182.780-induced growth inhibition. p38 was previously shown to participate in 4-hydroxytamoxifen-induced apoptosis of HeLa cells stably transfected with ER (22). We used antiestrogens at concentrations that induced growth arrest rather than apoptosis. Our results suggest that p38 activation patterns associated with apoptosis differ from those leading to growth inhibition: antiestrogens at low, growth-inhibitory concentrations induce a rapid and transient activation of p38, whereas high-dose, apoptosis-inducing 4-hydroxytamoxifen treatment was connected with a lasting activation of p38 (33).
The participation of p38 in cell growth regulation has been described previously. For example, sodium arsenite-induced growth inhibition was shown to be mediated by p38-dependent induction of the cyclin-dependent kinase inhibitor p21CIP1/WAF1 (34). In this study, inhibition of Cyclin A transcription, which represents an early marker of antiestrogen effects on proliferation (35), was shown to be regulated via a p38-dependent pathway. Our results thus provide a direct connection between p38 activity and antiestrogen growth inhibition.
Recent reports indicate that p38 is also involved in TGFß signal transduction. p38 was shown to be activated in response to TGFß in several cell types (15, 36). TGFß rapidly induced p38 activity in MDA-MB-231 breast cancer cells and was required for TGFß-dependent transcription of collagenase-3 and PTHrP (37, 38).
We have shown previously that breast cancer cells treated with antiestrogens secrete active TGFß (8). Therefore, it seemed possible that activation of p38 observed after treatment with antiestrogens was mediated via an autocrine TGFß signal pathway. However, our results revealed that the converse is true, i.e. that antiestrogen activation of p38 precedes and critically contributes to activation of TGFß by antiestrogens in MCF-7 cells. We provide two lines of evidence to support this conclusion. Phosphorylation of Smad2 was induced by treatment with antiestrogens or TGFß. Inhibition of p38 abolished the antiestrogen effect on Smad2 phosphorylation but not the TGFß effect. In addition, p3TP-lux, a TGFß-sensitive reporter gene, was activated by TGFß as well as by antiestrogens. The antiestrogen activation was mediated through TGFß as it could be blocked by overexpression of dominant negative TGFß receptors. Inhibition of p38 abolished antiestrogen-, but not TGFß-dependent activation of p3TP-lux. Taken together, these results show that TGFß activation of p3TP-lux is independent of p38, whereas antiestrogen activation of p3TP-lux is dependent on both p38 and TGFß.
Differences between our observations and those made in other model systems (15, 36, 37, 38) might result from a putative cross-talk between ER and TGFß/p38 signal pathways.
The critical role of antiestrogen-induced p38 activity for induction of autocrine TGFß signal pathways was further underlined by transcriptional analyses of TGFß2 and TßRII expression. Transcription of both genes is induced by antiestrogens, and the level of induction correlates directly with the growth-inhibitory potential of the respective antiestrogen (9, 11). Our results show that activation of p38 is a prerequisite for antiestrogen induction of TGFß2 and TßRII as inhibition of p38 completely abolished or strongly reduced 4-hydroxytamoxifen- and ICI 182.780-induced transcription, respectively.
This finding is in accordance with previous results showing that the promoter of TGFß2 contains an ATF-2 binding site (39), that ATF-2 is a substrate of p38 (40), and that p38-activated ATF-2 participates in regulation of TGFß2 promoter activity (41). Our data thus suggest that treatment with antiestrogens leads to a p38-dependent activation of ATF-2 and subsequent increased transcription of TGFß2.
Several transcription factors were shown to participate in the transcriptional control of TßRII (42, 43, 44, 45, 46). Some of them appear to be targets of p38 and might be involved in antiestrogen induction of TßRII transcription (47, 48). Ongoing work in our group is focused on the identification of transcription factors involved in the regulation of TGFß2 and TßRII by antiestrogens.
Furthermore, our study shows that a Smad pathway is involved in antiestrogen action downstream of p38 and TGFß. After activation by the TGFß receptors and heteromerization, Smads bind to specific DNA sequences and induce transcription of TGFß-sensitive genes (49, 50). The Smad pathway is an important mediator of TGFß-induced growth inhibition (51, 52). We were able to show in this study that Smad2 is phosphorylated in response to antiestrogen treatment and that Smad3 and Smad4 participate in antiestrogen action.
The lower degree of activation of the strictly Smad-dependent promoter of p6SBE in comparison with the more complex promoter of p3TP-lux probably results from the fact that Smad3 and Smad4 bind DNA with low affinity. Additional DNA contacts are necessary for specific and high-affinity binding of Smad complexes to target genes. This is achieved by cooperation of Smads with a multitude of different transcription factors (53, 54). Therefore, our results suggest that further transcription factors are involved in mediating antiestrogen-induced TGFß responses.
In conclusion, we propose the following signal transduction pathway leading to antiestrogen-induced growth inhibition (Fig. 9). Antiestrogens bind to the ER, which induces two effects: on the one hand, release of active TGFß from its latent precursor molecule (8, 9) and, on the other hand, activation of p38. The p38 signal cascade participates in the induction of TGFß2 and TßRII. Together these effects lead to increased TGFß signaling via a Smad-dependent pathway resulting in inhibition of proliferation.
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The following concentrations were used if not otherwise indicated: 4-hydroxytamoxifen, 100 nM; ICI 182.780, 1 nM; estradiol, 10 nM; SB203580, 10 µM; SB202190, 1 µM; SB202474, 5 µM; PD98059, 10 µM; U0126, 1 µM; U0124, 1 µM; TGFß, 100 pM.
Cell Culture
MCF-7 and T47D cells were maintained in DMEM (Invitrogen, Karlsruhe, Germany) containing 4.5 g glucose/liter supplemented with 50 µg/ml gentamicin and 10% fetal calf serum (FCS). Cells were passaged twice per week. Before their use in experiments, cells were maintained for one passage in the same medium as described above but with 5% of steroid-depleted FCS (sulfatase and charcoal-treated FCS). All media contained phenol red, which is known to have a weak estrogenic effect (55). Antiestrogens and estradiol were added in ethanol, giving a final ethanol concentration of 0.1% (vol/vol). MAPK inhibitors were added in dimethylsulfoxide, giving a final dimethylsulfoxide concentration of 0.1% (vol/vol).
Proliferation Assay
MCF-7 cells were plated in triplicate into 24-well cell culture dishes at a density of 2 x 104 cells per well. T47D cells were plated at a density of 5 x 104 cells per well. On the next day medium was changed, and fresh medium with antiestrogens and/or MAPK inhibitors or control vehicle was added. Cells were harvested 5 d after treatment began and were counted using a cell counter (Casy, Schärfe Systems, Reutlingen, Germany).
Fluorescence-Activated Cell Sorting Analysis
MCF-7 cells were plated in six-well cell culture dishes at a density of 2 x 105 cells per well and treated with antiestrogens at the indicated concentrations. Cells were harvested after the indicated time periods, fixed by slow addition of 70% ethanol, and stored overnight at 4 C. After fixation cells were pelleted, washed with PBS, and then stained with 1 ml of 50 µg/ml propidium iodide, 1 mg/ml RNase A in PBS for 20 min. Cells were analyzed by flow cytometry by gating on an area vs. width dot plot to exclude debris and aggregates, and 15,000 cells were examined on 585 nm fluorescence vs. cell number histogram.
p38 MAPK Assay
The nonradioactive p38 MAPK Assay from Cell Signaling [New England Biolabs (NEB), Frankfurt am Main, Germany] was used according to the manufacturer’s instructions. Briefly, MCF-7 or T47D cells were plated into 100-mm cell culture dishes at a density of 4 x 106 cells per well. On the next day cells were treated with antiestrogens or estradiol for the indicated time periods. Cells were lysed in 500 µl cell lysis buffer (20 mM Tris, pH 7.5; 150 mM NaCl; 1 mM EDTA; 1 mM EGTA; 1% Triton X-100; 2.5 mM sodium pyrophosphate; 1 mM ß-glycerolphosphate; 1 mM Na3VO4; 1 µg/ml leupeptin; 1 mM PMSF; 4.6 µM Cantharidin; 20 nM (–)-p-Bromotetramisole; 4.5 nM Microcystin) and 500 µg of total protein were incubated overnight with immobilized phospho-p38(Thr180/Tyr182) monoclonal antibody. Immunoprecipitated phospho-p38 was incubated with 200 µM ATP and 2 µg of a ATF-2 glutathione-S-transferase fusion protein at 30 C for 30 min in kinase buffer (25 mM Tris, pH 7.5; 5 mM ß-glycerolphosphate; 2 mM dithiothreitol; 0.1 mM Na3VO4; 10 mM MgCl2). The samples were separated on a 12% sodium dodecyl sulfate gel and blotted to a nitrocellulose membrane. Phosphorylation of ATF-2 at Thr71 was measured using a specific phospho-ATF-2 antibody.
Plasmids
Expression plasmids for kinase-truncated TßRI and TßRII (dKTßRI and dKTßRII) were constructed by PCR. Primers with EcoRI and NotI restriction sites were used to amplify nucleotides 8–726 of TßRI (GenBank accession no. L11695). Primers with EcoRI and XhoI restriction sites were used to amplify nucleotides 306-1237 of TßRII (GenBank accession no. M85079). The corresponding cDNAs were cloned into the EcoRI and NotI/XhoI site of a pcDNA3-flag vector (Invitrogen, Karlsruhe, Germany).
Kinase-deficient mutants of p38 and p38ß were made by site-directed mutagenesis using QuikChange (Stratagene, Amsterdam, The Netherlands) and specific primers resulting in exchange of threonine 180 with alanine and tyrosine 182 with phenylalanine in the phosphorylation motif TGY. The resulting mutants were named p38AGF and p38ßAGF. The nucleotide exchange was verified by sequencing. p38AGF and p38ßAGF were cloned into the expression vector pUB6/V5-His (Invitrogen).
The expression plasmid for the Smad4 deletion construct (pcDNA3-Smad4dM) was kindly provided by Mark de Caestecker. The reporter plasmid p3TP-lux was kindly provided by Jens Würthner. p6SBE was a gift from Werner Hilgers.
Transient Transfections and Luciferase Assay
MCF-7 cells were plated in triplicate into 24-well cell culture dishes at a density of 5 x 104 cells per well. MAPK inhibitors were added directly after plating. On the next day, medium was changed and fresh medium with 100 pM TGFß and MAPK inhibitor or control vehicle was added. Directly after treatment began, cells were transfected with 300 ng/well p3TP-lux, 150 ng/well pcDNA3-Smad4dM-Flag, or pcDNA3 (to adjust the transfected amount of DNA to 500 ng) and 5 ng/well phRL-TK. FuGene6 (Roche, Mannheim, Germany) was used as transfection reagent. Cells were harvested 24 h later and assayed for luciferase activity using the dual luciferase assay system (Promega, Mannheim, Germany).
Antiestrogen-mediated induction of TGFß requires 3–5 d to reach its highest level; therefore, a different transfection scheme was adopted: MCF-7 cells were plated in duplicate into 24-well cell culture dishes at a density of 2 x 104 cells per well. On the next day, the medium was changed and fresh medium with antiestrogens and/or MAPK inhibitors or control vehicle was added. Cells were transfected 3 d after treatment began. In the case of p3TP-lux, cells were cotransfected with 300 ng/well p3TP-lux, 150 ng/well of the dominant negative constructs (pcDNA3-dKTßRI-Flag, pcDNA3-dKTßRII-Flag, pcDNA3-Smad4dM-Flag) or empty expression vector, and 5 ng/well phRL-TK. In the case of p6SBE, cells were transfected with 200 ng p6SBE, 600 ng Smad expression constructs (pcDNA3-Smad2-Flag, pcDNA-Smad3-Flag, pcDNA-Smad4-Flag), 1 ng/well phRL-TK, and varying amounts of pcDNA3 (final DNA amount was 1000 ng). Cells were harvested 48 h later and assayed for luciferase activity using the dual luciferase assay system (Promega).
Total light emission was measured during the initial 10 sec of the reaction using a luminometer (Autolumat Plus, Berthold Technologies, Bad Wildbad, Germany). Firefly-luciferase activities were corrected using Renilla-luciferase activities.
Western Blot Analysis
MCF-7 cells were plated at a density of 2.5 x 105 cells in 100-mm cell culture dishes. On the next day cells were treated as indicated. Five days after treatment cells were lysed in 500 µl cell lysis buffer (20 mM Tris, pH 7.5; 150 mM NaCl; 1 mM EDTA; 1 mM EGTA; 1% Triton X-100; 2.5 mM sodium pyrophosphate; 1 mM ß-glycerolphosphate; 1 mM Na3VO4; 1 µg/ml leupeptin; 1 mM PMSF; 4.6 µM Cantharidin; 20 nM (–)-p-Bromotetramisole; 4.5 nM Microcystin). Proteins (100 µg) were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked with 5% nonfat dry milk in Tris-buffered saline containing 0.05% Tween 20 and then incubated with antibodies raised against phospho-Smad2 (Ser465/467) (Cell Signaling, NEB), phospho-p44/42 MAPK (Thr202/Tyr204) (Cell Signaling, NEB) or phospho-Elk-1 (Santa Cruz, Heidelberg, Germany) overnight at 4 C. Blots were developed by peroxidase-conjugated secondary antibody, and proteins were visualized using SuperSignal West Dura Extended Duration Substrate (Pierce, Bonn, Germany) according to the manufacturer’s recommendation.
Membranes were stripped for 30 min at 50 C in buffer containing 62.5 mM Tris HCl (pH 6.7), 2% sodium dodecyl sulfate, and 100 mM mercaptoethanol and rehybridized with Smad2 antibody (BD Biosciences, Heidelberg, Germany) as described above. Membranes were stripped again for rehybridization with GAPDH antibody (Biodesign, Asbach, Germany).
RNA Isolation
MCF-7 cells were seeded at a density of 1 x 105 cells in six-well-plates. the medium was exchanged 24 h later for medium containing antiestrogens and/or MAPK inhibitors or control vehicle. Cells were harvested 120 h after treatment. For detection of Cyclin A expression, cells were seeded at a density of 5 x 105. On the next day medium was exchanged for medium containing antiestrogens and cells were transfected with 1000 ng of pUB6, pUB6-p38AGF, or pUB6p38ßAGF using FuGene6 (Roche) and harvested 24 h after transfection. Total RNA was isolated with RNeasy spin columns (QIAGEN, Hilden, Germany) as described by the supplier, followed by treatment with DNase I.
Quantitative LightCycler RT-PCR
Total RNA (80 ng) was amplified with the LightCycler-RNA Amplification Kit Hybridization Probes (Roche) in a one-tube RT-PCR. The sequences of primers and hybridization probes were as follows:
TGFß2: TGFß2 sense (se) 5'-GTT TTT CTG TTG GGC ATT GA-3', TGFß2 antisense (as) 5'-TCT TCT GGG GGA CTG GTG AG-3', TGFß 2 FL 5'-GCG CTT TTC TGA TCC TGC ATC TGG-3', TGFß 2 LC 5'-ACG GTC GCG CTC AGC CTG TCT AC-3';
TßRII: TßRII se 5'-CCT CCA CAG TGA TCA CAC TC-3', TßRII as 5'-TCG GTC TGC TTG AAG GAC TC-3', TGFßRII FL 5'-GAC CTA ACC TGC TGC CTG TGT GAC TTT-3', TGFßRII LC 5'-CTT TCC CTG CGT CTG GAC CCT ACT-3';
Cyclin A: Cyclin A se 5'-GCC TGC GTT CAC CAT TCA TG-3', Cyclin A as 5'-CCA GTC CAC GAG GAT AGC TC-3', CycA FL 5'-TGG GTC CAG GTA AAC TAA TGG CTG AAT-3', CycA LC 5'-AAG CCA GGG CAT CTT CAC GCT CTA TT-3';
GAPDH: GAPDH se 5'-CGG AGT CAA CGG ATT TGG TCG TAT-3', GAPDH as 5'-AGC CTT CTC CAT GGT GGT GAA GAC-3', GAPDH FL 5'-AGG GGT CAT TGA TGG CAA CAA TAT CCA-3', GAPDH LC 5'-TTT ACC AGA GTT AAA AGC AGC CCT GGT G-3'.
Samples were incubated at 55 C for 10 min for reverse transcription. The cycling parameters were one denaturation cycle at 95 C for 2 sec and 40 amplification cycles (temperature transition rate of 20 C/sec) at 94 C for 2 sec, 62 C (TGFß2, Cyclin A, GAPDH) or 60 C (TßRII) for 10 sec, and 72 C for 18 sec (Cyclin A), 14 sec (TGFß2, GAPDH), or 11 sec (TßRII). Fluorescence readings were taken after the annealing step. For each target sequence, a serial dilution of a target-specific standard cRNA (10 pg to 1 fg) was amplified at the same time as the samples. The concentrations of the target mRNAs were extrapolated from the standard curve. Induction of mRNA was calculated in comparison with control cells treated with vehicle.
Statistical Analysis
Data are given as mean ± SEM. Statistical comparisons were performed by unpaired Student’s t test. A value of P < 0.05 was considered significant.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: ATF-2, Activating transcription factor 2; ER, estrogen receptor; FCS, fetal calf serum; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; JNK, c-jun N-terminal kinase; MEK, MAPK kinase; Smad, mothers against decapentaplegic homolog; TAF, transcriptional activation function; TßRI and -II, TGFß receptors I and II.
Received for publication July 16, 2003. Accepted for publication March 23, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
