This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager NURSA Molecule Pages Link Reprints, Permissions and Rights Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ma, Y. Articles by Rosen, E. M. Search for Related Content PubMed PubMed Citation Articles by Ma, Y. Articles by Rosen, E. M.

Growth Factor Signaling Pathways Modulate BRCA1 Repression of Estrogen Receptor- Activity

Department of Oncology, Lombardi Comprehensive Cancer Center, Georgetown University, Washington, D.C. 20057-1469

Address all correspondence and requests for reprints to: Dr. Eliot M. Rosen, Department of Oncology, Lombardi Comprehensive Cancer Center, Georgetown University, 3970 Reservoir Road NW, Preclinical Sciences Building, Room GM12B, Washington, D.C. 20057-1469. E-mail: emr36{at}georgetown.edu.

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
The breast cancer susceptibility gene BRCA1 is mutated in about one half of all hereditary breast cancer cases, and its expression is frequently decreased in sporadic cancers. Previously, we demonstrated a functional interaction between the BRCA1 and estrogen receptor- (ER-) proteins that causes inhibition of ER- signaling. Here, we examined the role of growth factor signaling pathways in modulating this interaction. We found that underexpression of BRCA1 caused ligand-independent activation of ER- that was mediated through phosphatidylinositol-3 kinase (PI3K)/c-Akt signaling. BRCA1 underexpression also enhanced estrogen-inducible ER- activity in a PI3K/Akt-dependent manner. Exogenous c-Akt conferred estrogen-independent ER- activation and rescued the BRCA1 repression of estrogen-stimulated ER- activity. BRCA1 knockdown stimulated c-Akt activity, in part, by inhibiting the activity of protein phosphatase 2A, an enzyme that dephosphorylates Akt. ERs with point mutations of several growth factor-targeted serine residues (S167A, S118A, and S118/167A) were resistant to repression by BRCA1, although the single point mutant receptors still associated with the BRCA1 protein. The enhanced ER- activity attributable to BRCA1 knockdown was dependent, in part, on serine residues 167 and 118 of ER-. BRCA1 knockdown caused an increase in ER- phosphorylation on serine-167 (but not serine-118 or serine-104/106) that was dependent on PI3K/Akt signaling and was mimicked by pharmacologic inhibition of protein phosphatase 2A. These findings suggest that BRCA1 regulates Akt signaling and the PI3K/Akt pathway modulates the ability of BRCA1 to repress ER-, in part through serine phosphorylation events in the activation function-1 domain of ER-.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
AN ESTIMATED 5–10% OF all breast cancer cases are attributable to inherited mutations of susceptibility genes. Among these genes, mutations of BRCA1 and BRCA2 account for about 50–85% of hereditary early-onset breast cancer cases (1, 2, 3). Inactivation of the BRCA1 gene may also contribute to the pathogenesis of the larger group of sporadic breast cancers, because BRCA1 expression is frequently decreased in sporadic cancers attributable to hypermethylation of the BRCA1 promoter, loss of one BRCA1 allele, a combination of the two, or other causes (4, 5, 6, 7). A variety of studies have implicated BRCA1 as a mediator of different aspects of DNA damage signaling, cell cycle checkpoints, and DNA repair (8 ; for review, see9, 10), and, accordingly, BRCA1 mutant breast cancer exhibits increased evidence of genomic instability compared with sporadic cancers (11, 12). Based on these findings, a tumor suppressor role for BRCA1 as a caretaker in preserving genomic integrity has been proposed (13).

Although the caretaker hypothesis for BRCA1 is attractive, it does not, by itself, explain why BRCA1 mutations are associated with specific tumor types, particularly breast cancer and gynecologic tumor types (ovarian, cervical, and endometrial cancers) (14), rather than a broad spectrum of cancer types. To address this issue, we and others identified a tissue-specific function for BRCA1 that may, in part, explain the predilection of BRCA1 mutation carriers to develop estrogen-responsive cancer types such as breast cancer. Thus, BRCA1 selectively inhibits the transcriptional activity of estrogen receptor- (ER-), in part, by a specific interaction with its carboxy-terminal transcriptional activation domain activation function 2 (AF-2) and, in part, by down-regulation of expression of a coactivator (p300) (15, 16, 17, 18 ; for review, see19). Additional studies have revealed that BRCA1 inhibits ER--mediated transcription and secretion of vascular endothelial growth factor via a direct interaction with the AF-2 domain of ER- (20). It has also been demonstrated that inactivation of BRCA1, either through a homozygous null mutation in mouse embryo fibroblasts or through RNA interference in human breast cancer cells, confers ligand-independent activation of ER- (21, 22). Based on these and other considerations, we postulated that, in the setting of a BRCA1 deficiency, loss of both the caretaker function and ER- regulation contribute to mammary carcinogenesis (19).

Over the past 10 or so years, various studies have identified mechanisms through which growth factor signal transduction pathways can crosstalk with and activate ER- signaling (23, 24, 25, 26, 27, 28, 29). These pathways involve phosphorylations that may be mediated by various kinases, including phosphatidylinositol-3 kinase (PI3K)/c-Akt (protein kinase B), MAPK, ribosomal S6 kinase 1, several cyclin-dependent kinases, protein kinase A, and possibly others. Activation of ER- through growth factor signaling may contribute to breast cancer development, progression, and resistance to therapy. In this study, we investigated the ability of growth factor signal transduction pathways to modulate the BRCA1-mediated repression of ER- activity. Our findings suggest that the ability of BRCA1 to repress ER- activity can be inactivated, in part, through a pathway involving PI3K c-Akt signaling that results in serine phosphorylation events within the AF-1 transcriptional activation domain of ER-.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Ligand-Independent Activation of ER- Is Dependent on PI3K Activity
Previously, we reported that knockdown of BRCA1 conferred estrogen-independent activation of ER- in MCF-7 cells, as measured using an estrogen-responsive reporter plasmid (ERE-TK-Luc) (30). The ability of the BRCA1 small interfering RNA (siRNA) to knock down BRCA1 protein levels is shown in Fig. 1A. Here, pretreatment with BRCA1-siRNA for 48 h caused about an 8- to 14-fold increase in ERE-TK-Luc activity (P < 0.001, two-tailed t test) in different transient transfection assays of MCF-7 cells in the absence of ligand [17ß-estradiol (E2)] (illustrated in Fig. 1B). Conversely, a control-siRNA had little or no effect on ERE-TK-Luc activity under the same conditions. In the same experiment, wortmannin, a selective inhibitor of PI3K (used at a nontoxic dose of 100 nM; see below) reduced the BRCA1-siRNA-stimulated ER- activity by more than 75% (P < 0.001), whereas vehicle only [dimethylsulfoxide (DMSO)] had little or no effect on ER- activity (Fig. 1B). Similarly, another selective PI3K inhibitor, LY294002 [2-(4-morpholinyl)-8-phenyl-1(4H)-benzopyran-4-one] (also used at a nontoxic dose), caused a similar reduction in BRCA1-siRNA-stimulated ligand-independent ER- activity (P < 0.001) (Fig. 1C). These findings suggest that E2-independent activation of ER- mediated through BRCA1-siRNA, in part, requires the enzymatic activity of PI3K.

View larger version (28K): [in this window] [in a new window]   Fig. 1. Role of PI3K in Mediating ER- Activation in BRCA1 Knockdown Cells A, Knockdown of BRCA1 protein in MCF-7 cells by RNA interference. Subconfluent proliferating cells were treated with BRCA1-siRNA or control-siRNA (100 nM) for 48 h and subjected to Western blotting to detect BRCA1 and -actin (control for loading and transfer). B and C, PI3K inhibitors wortmannin (B) and LY294002 (C) inhibit ligand-independent ER- activity in BRCA1 knockdown cells. Subconfluent proliferating MCF-7 cells in 24-well dishes were pretreated with BRCA1-siRNA or control-siRNA (100 nM for 48 h) and then transfected overnight with the ERE-TK-Luc reporter (0.25 µg/well) in serum-free DMEM containing Lipofectamine. The cells were washed, incubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum without (–) or with (+) wortmannin (100 nM) (B), LY294002 (100 µM) (C), or an equal quantity of vehicle only (DMSO) for 24 h, and then harvested for luciferase assays. BRCA1-siRNA or control-siRNA was present during this incubation period, as before. The luciferase activity was calculated relative to the negative control value (ERE-TK-Luc only), and the values were expressed as means ± SEMs of four replicate wells. The data shown are representative of at least two independent experiments. D and E, PI3K inhibitors wortmannin (D) and LY294002 (E) inhibit ligand-dependent ER- activity in BRCA1 knockdown cells. Assays were performed as described above, except that, after transfection of ERE-TK-Luc reporter, the 24-h incubation was performed in the presence (+) or absence (–) of E2 (10 nM). The luciferase values were expressed relative to the negative control (ERE-TK-Luc only, no E2) and are means ± SEMs of quadruplicate wells. CON, Control.

c-Akt Signaling Is Required for Stimulation of ER- Activity by BRCA1 Knockdown
Previous studies have shown the tumor suppressor PTEN (phosphatase and tensin homolog) acts as an upstream inhibitor of c-Akt through its lipid phosphatase activity (31). Here, we found that a wild-type (wt) PTEN expression vector caused a more than 5-fold reduction of BRCA1-siRNA-stimulated ER- activity in the absence of ligand (P < 0.001, two-tailed t test), but an empty pcDNA3 vector had little or no effect on BRCA1-siRNA stimulated ligand-independent ER- activity (Fig. 2A). As observed previously, BRCA1-siRNA caused a significant (5-fold) increase in E2-stimulated ER- activity in MCF-7 cells (P < 0.001), and wtPTEN caused a 40% reduction in the E2-stimulated ER- activity in BRCA1-siRNA-treated cells (P < 0.001) (Fig. 2B). These findings suggest that PTEN is less efficient at inhibiting ligand-dependent than ligand-independent ER- activity attributable to BRCA1 knockdown.

View larger version (37K): [in this window] [in a new window]   Fig. 2. Role of c-Akt Signaling in Mediating ER- Activation in BRCA1 Knockdown Cells A, PTEN inhibits ligand-independent ER- activity in BRCA1 knockdown cells. Proliferating MCF-7 cells in 24-well dishes were pretreated with BRCA1-siRNA, control-siRNA (100 nM for 48 h), or no siRNA and then cotransfected overnight with ERE-TK-Luc and wtPTEN, empty pcDNA3 vector, or no vector (0.25 µg of each vector per well). The cells were washed, incubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum in the presence of BRCA1-siRNA, control-siRNA, or no siRNA for 24 h, and harvested for luciferase assays. The relative luciferase activity values are means ± SEMs of n = 4 wells. B, Effect of PTEN on ligand-stimulated ER- activity in BRCA1 knockdown cells. Assays were performed as described in A, except that, after the overnight transfection, the 24-h incubation was performed in the presence (+) or absence (–) of E2 (10 nM). C, Effect of Akt inhibition on ligand-independent ER- activity in BRCA1 knockdown cells. Assays were performed as above, except that, instead of wtPTEN, the cells were transfected with a DN-Akt vector, empty pcDNA3 vector, or no vector. D, Effect of Akt inhibition on ligand-stimulated ER- activity in BRCA1 knockdown cells. Assays were performed as described in C, except that, after the overnight transfection, the 24-h incubation was performed in the presence (+) or absence (–) of E2 (10 nM). CON, Control.

c-Akt Confers Ligand-Independent ER- Activation and Rescues BRCA1 Repression of ER-
In additional studies, we found that wtAkt (P < 0.001) but not the kinase-dead DN-Akt or the empty pcDNA3 vector caused about a 5-fold increase in ER- activity in the absence of ligand (Fig. 3A). The coexpression of wtBRCA1 did not inhibit the ability of wtAkt to mediate E2-independent activation of ER- (Fig. 3A). In the presence of E2, wtAkt had little effect on ER- activity, and DN-Akt caused only a modest (<20%) reduction in ER- activity (Fig. 3B). wtBRCA1 nearly completely repressed E2-stimulated ER- activity (P < 0.001), and the coexpression of wtAkt (but not DN-Akt) rescued the wtBRCA1-mediated repression of E2-stimulated ER- activity (P < 0.001) (Fig. 3B). The expression of the wtBRCA1 and wtAkt vectors in transfected MCF-7 cells is illustrated in Fig. 3, C and D, respectively. Thus, although c-Akt did not further enhance the ability of E2 to stimulate ER- activity, it did override the ability of BRCA1 to repress ER- signaling and it did activate the unliganded receptor, both in a manner dependent on its catalytic activity.

View larger version (38K): [in this window] [in a new window]   Fig. 3. c-Akt Causes Ligand-Independent ER- Activation and Rescues BRCA1 Repression of ER A, Wild-type but not kinase-dead Akt activates ER- in a BRCA1-resistant manner. Subconfluent proliferating MCF-7 cells in 24-well dishes were cotransfected overnight with ERE-TK-Luc and wtAkt, DN-Akt, empty pcDNA3 vector, or no vector (0.5 µg of each vector per well). The cells were then washed, postincubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum without (–) E2 for 24 h, and assayed for luciferase activity. The relative luciferase activity values are means ± SEMs of n = 4 wells. B, wtAkt but not kinase-dead Akt rescues BRCA1 repression of liganded ER-. Assays were performed as described in A, except that, after the overnight transfection, the 24-h incubation was performed in the presence (+) or absence (–) of E2 (10 nM). C, Expression of BRCA1 in wtBRCA1-transfected MCF-7 cells. Subconfluent proliferating cells were transfected overnight with the wtBRCA1 or empty pcDNA3 vector, washed, and postincubated for 24 h to allow gene expression. The cells were then harvested for Western blotting to detect BRCA1 or -actin (control for loading and transfer). D, Expression of Akt in wtAkt-transfected MCF-7 cells. Assays were performed as described above in C.

View larger version (35K): [in this window] [in a new window]   Fig. 4. BRCA1 Regulates the Activation of c-Akt in MCF-7 Cells A, wtBRCA1 inhibits and BRCA1-siRNA stimulates the Akt phosphorylation. Proliferating MCF-7 cells were pretreated with BRCA1-siRNA, control-siRNA (100 nM for 48 h), or no siRNA or transfected with wtBRCA1 or empty pcDNA3 vector and postincubated for 24 h to allow gene expression. They were then treated with or without E2 (10 nM for 20 min) or IGF-I (100 nM for 20 min) and harvested for Western blotting for phospho-Akt (serine-473), total Akt, or -actin. The right shows densitometry quantification of the phospho-Akt (serine-473) bands relative to -actin, based on four independent experiments. B, c-Akt phosphorylation induced by BRCA1 knockdown is blocked by PI3K inhibitors. Cells were treated with siRNAs and IGF-I as described above in the absence or presence of wortmannin (100 nM) or LY294002 (100 µM). They were then harvested for Western blotting to detect phospho-Akt (serine-473) or total Akt. C, Ability of BRCA1-siRNA to induce the phosphorylation of c-Akt substrate proteins. MCF-7 cells were treated with control-siRNA, BRCA1-siRNA (100 nM for 48 h), or no siRNA in the absence or presence of the PI3K inhibitor wortmannin (100 nM). The cells were then harvested for Western blotting to detect phospho (P)-GSK-3ß (serine-9) or total GSK-3ß. D, Effect of BRCA1 on c-Akt kinase activity. Cells were treated with siRNAs or transfected with expression vectors as indicated in the absence or presence of wortmannin (100 nM) and then harvested for assays of c-Akt kinase activity, as described in Materials and Methods. E, Ability of BRCA1-siRNA to induce PP2A activity. Cell treatments with siRNAs or expression vectors were performed as described above, except that, for the last 10 h, the cells were incubated without or with OA (5 nM). They were then harvested for measurements of PP2A enzymatic activity using a commercial assay kit, as described in Materials and Methods. The values plotted are the means ± SEMs of three independent experiments, and they are normalized to the control value observed in the absence of siRNA, vector transfection, OA treatment, or LY294002 treatment. F, OA induces or enhances the phosphorylation of c-Akt. Cells were treated as described for E and then harvested for Western blotting for phospho-Akt and total Akt. G, OA induces or enhances the phosphorylation of ATM. Cells were treated as in E and then harvested for Western blotting for phospho-ATM (serine-1981) or total ATM. CON, Control.

Next, we tested whether the BRCA1-siRNA-mediated activation of c-Akt is associated with the phosphorylation of a well-established kinase substrate of c-Akt, glycogen synthase kinase 3ß (GSK-3ß). This study was performed by treating the cells with BRCA1-siRNA vs. control-siRNA and analyzing the cells by Western blotting using phospho-specific antibodies. As shown in Fig. 4C, BRCA1-siRNA caused an increase in phospho-GSK-3ß (serine-9) that was blocked by the PI3K inhibitor wortmannin. Conversely, the control-siRNA had no effect on phospho-GSK-3ß levels, and neither BRCA1-siRNA nor Wortmannin altered the total GSK-3ß protein levels.

Finally, we directly tested the effect of altered BRCA1 levels on c-Akt kinase activity using a commercial c-Akt activity assay kit. This assay measures the ability of Akt immunoprecipitated from cell lysates to phosphorylate a model protein substrate (GSK-3) under standardized reaction conditions. Phosphorylated GSK-3 (serine-21) was detected by Western blotting and quantitated by densitometry. Here, knockdown of endogenous BRCA1 caused about a 4-fold increase in c-Akt activity that was blocked by Wortmannin, suggesting that this increase in activity requires PI3K (Fig. 4D).

Knockdown of BRCA1 Causes Inhibition of Protein Phosphatase 2A (PP2A) Activity
Akt can be dephosphorylated and inactivated by PP2A, a protein serine/threonine phosphatase that inhibits Akt activity by removing the phosphate from the key activating site at serine-473 (32, 33, 34). We tested the effect of BRCA1-siRNA on PP2A activity by an immunoprecipitation (IP)-phosphatase assay using a commercial assay kit. As a positive control, cells were treated with 5 nM okadaic acid (OA), a selective pharmacologic inhibitor of PP2A activity. Here, BRCA1-siRNA caused a large decrease in PP2A activity (P < 0.001), whereas the control-siRNA had little or no effect on PP2A activity (Fig. 4E). BRCA1 overexpression caused a modest increase in PP2A activity, whereas BRCA1-siRNA reduced the PP2A activity to levels only slightly higher than those observed using OA alone. The effects of BRCA1-siRNA plus OA were slightly greater than those of OA alone. Consistent with these findings, OA treatment caused an increase of phospho-Akt (serine-473) levels that was similar in magnitude to that caused by BRCA1-siRNA alone, and the combination of BRCA1-siRNA plus OA gave a somewhat greater effect than either agent alone (Fig. 4F).

Ionizing radiation causes autophosphorylation of ataxia-telangiectasia mutated protein (ATM) on serine-1981 by a process that is regulated by PP2A (35). As an additional demonstration of the ability of BRCA1-siRNA to inhibit PP2A activity, we showed that both BRCA1-siRNA and OA yielded significant increases in the phospho-ATM (serine 1981) protein levels (Fig. 4G). The effects of the combination of BRCA1-siRNA plus OA gave a slightly greater effect than either agent alone. Together, these findings suggest that ability of BRCA1-siRNA to up-regulate Akt activity is attributable, in part, to its ability to stimulate the activity of PP2A, an important regulator of protein kinase activity.

Role of Serine-167 and Serine-118 in BRCA1 Repression of ER- Activity
Previous studies have established that growth factor signaling pathways can stimulate the ligand-independent and/or ligand-dependent activity of ER-. The mechanism of growth factor and oncogene-stimulated ER- activity involves, in part, several phosphorylation events, including those on serine-118, serine-167, and other serine residues within the transcriptional AF-1 domain of ER- (23, 24, 25, 26, 27, 28, 29). In this regard, serine-118 is the target of an epidermal growth factor receptor Ras Raf1 MAPK pathway (23, 24), whereas serine-167 is the target of PI3K c-Akt pathway (28, 29). Here, we tested the roles of these two serine residues in BRCA1-mediated repression of E2-stimulated ER- activity using full-length mutant ERs with point mutations (serine alanine) in either or both of these serine residues. For these studies, transient transfection assays were performed in DU-145 human prostate cancer cells rather than MCF-7 cells, because DU-145 cells lack endogenous ER-.

As shown in Fig. 5A, E2 (10 nM) caused a 105- to 110-fold increase in ERE-TK-Luc activity in DU-145 cells that were cotransfected with a wt-ER- expression vector; and this activity was nearly abrogated by wtBRCA1 (P < 0.001). In cells transfected with the S167A mutant ER-, the fold stimulation of ERE-TK-Luc activity by E2 was about 60% of that observed in wt-ER--transfected cells. In contrast to the wt-ER-, the S167A mutant was quite resistant to repression by wtBRCA1 and retained about 60% of its activity in the presence of wtBRCA1 (P < 0.001 for comparison of E2 + S167A + wtBRCA1 vs. E2 + wt-ER- + wtBRCA1) (Fig. 5A). In another set of experiments, we found that E2 stimulated the activity of the S118A mutant ER- about as efficiently as it did the wt-ER-, but the S118A mutant ER- was also more refractory to repression of E2-stimulated ER- activity by wtBRCA1 than was the wt-ER- (P < 0.001) (Fig. 5B). In both sets of experiments, the empty pcDNA3 vector had little or no effect on ER- activity. Finally, we tested the sensitivity of double-mutant ER- (S118/167A) to repression by wtBRCA1 in DU-145 cells. The double-mutant ER- showed a modest reduction in E2-stimulated ERE-TK-Luc activity compared with the wt-ER-. However, strikingly, there was no repression of activity of the S118/167A receptor by wtBRCA1 (Fig. 5C). The increased expression of BRCA1 in DU-145 cells transfected with wtBRCA1 is shown in Fig. 5I. Together, these findings suggest that serine residues 167 and 118 within the AF-1 domain of ER- control the sensitivity of the liganded ER- to repression by ER-.

View larger version (32K): [in this window] [in a new window]   Fig. 5. Roles of Serine-167 and Serine-118 in BRCA1 Repression of Ligand Activated ER- A, Ability of wtBRCA1 to repress ER- with S167A mutation. Subconfluent proliferating DU-145 cells were cotransfected with ERE-TK-Luc and the indicated vectors (0.25 µg of each vector per well). The cells were then washed, postincubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum without (–) or with (+) E2 (10 nM) for 24 h, and assayed for luciferase activity. The relative luciferase activity values are means ± SEMs of n = 4 wells. B, Ability of wtBRCA1 to repress S118A mutant ER-. Assays were performed as above. C, Ability of wtBRCA1 to repress S118/167A double-mutant ER-. See above. D and E, In vivo association of BRCA1 with S167A and S118A mutant ER-. Subconfluent proliferating DU-145 cells were transfected overnight with wild-type ER-, S167A mutant ER-, S118A mutant ER-, or empty pcDNA3 vector (10 µg of plasmid DNA per 100-mm culture dish) and postincubated for 24 h to allow gene expression. The cells were harvested, subjected to anti-BRCA1 IP (D) or anti-ER-, and Western blotted to detect ER- and BRCA1. As a negative control, an IP was performed in wt-ER--transfected cells using an equal quantity of nonimmune IgG. Each IP used 500 µg whole-cell protein. The lysate lanes show Western blots of 50 µg unprecipitated whole-cell protein. F and G, In vivo association of BRCA1 with S118/167A double-mutant ER-. Assays were performed as described in D and E, except that the cells were transfected with wt-ER- or the S118/167A double-mutant ER-. Results of the BRCA1 IP are shown in F, and the results of the ER- IP are shown in G. H, Expression of wild-type and mutant ER- proteins in DU-145 cells. Subconfluent proliferating cells were transfected overnight with the indicated wt or mutant ER- expression vector, washed, and postincubated for 24 h to allow gene expression. The cells were harvested for Western blotting for ER- or -actin (control for loading and transfer). I, Expression of wtBRCA1 in DU-145 cells. To confirm the expression of the wtBRCA1 vector used in A–C, DU-145 cells were transfected with wtBRCA1, empty pcDNA3 vector, or no vector and Western blotted to detect BRCA1 or -actin (control for loading and transfer).

Next, we examined the ability of the double point mutant ER- (S118/167A) to associate with BRCA1 in DU-145 cells. This mutant ER- was well expressed, and anti-ER- IPs of transfected cells revealed approximately equal quantities of wt-ER- vs. mutant ER- S118/167A (Fig. 5F). However, the quantity of BRCA1 in the ER- IP was substantially reduced in S118/167A-transfected cells compared with wt-ER--transfected cells. Conversely, the quantity of ER- in the BRCA1 IP was much less in cells transfected with S118/167A compared with wt-ER- (Fig. 5G). Each of the negative controls worked as expected. In addition, we used Western blotting to confirm that the levels of expression of the wild-type and mutant ER proteins in DU-145 cells were relatively similar in all cases (Fig. 5H). These findings suggest that, whereas the single mutations S167A and S118A have no effect or a modest effect on the ability of ER- to associate with the BRCA1 protein in vivo, the double point mutation of ER- significantly disrupts its ability to interact with BRCA1.

Roles of Serine-118 and Serine-167 in Stimulation of ER- Activity by BRCA1 Knockdown
We tested whether two serine residues of ER- targeted by growth factor signaling pathways (S167 and S118) contribute to stimulation of ER- activity by BRCA1-siRNA. These assays were performed using DU-145 rather than MCF-7 cells, because MCF-7 cells have endogenous ER-, which would interfere with interpretation of the results. Briefly, DU-145 cells were pretreated with BRCA1-siRNA or control-RNA, cotransfected with ERE-TK-Luc and different ER- expression vectors, treated with or without E2 for 24 h, and assayed for reporter activity. For DU-145 cells, little or no ERE-TK-Luc activity above baseline (1.0) was observed in the absence of E2 in cells transfected with empty vector, wt-ER-, or mutant ER- (S118A) (Fig. 6A). The control-siRNA had little or no effect on the activity of the nonliganded wt-ER- or S118A mutant ER-. However, BRCA1-siRNA caused a 9.8-fold increase in wt-ER- activity and a 5.1-fold increase in ER- (S118A) activity in the absence of E2 (P < 0.001). In the presence of E2, the ERE-TK-Luc activity remained at the baseline level in the absence of exogenous ER- but was increased by 52-fold in the presence of wt-ER- (P < 0.001) and 34-fold (P < 0.001) in the presence of ER- (S118A) (Fig. 6A). Here, BRCA1-siRNA increased the relative ERE-TK-Luc activity attributable to wt-ER- from 52- to 135-fold (ratio of 2.6) (P < 0.001) and increased the activity of ER- (S118A) from 34- to 57-fold (ratio of 1.7) (P < 0.001). These findings suggest that the degree to which BRCA1-siRNA can stimulate the activity of the unliganded or liganded ER is reduced but is not abrogated by the S118A mutation.

View larger version (25K): [in this window] [in a new window]   Fig. 6. Roles of Serine-167 and Serine-118 in Stimulation of ER- by BRCA1 Knockdown A, Effect of BRCA1-siRNA on activity of S118A mutant ER-. Subconfluent proliferating DU-145 cells were pretreated with BRCA1-siRNA or control-siRNA (100 nM for 48 h) and then cotransfected overnight with ERE-TK-Luc and the indicated expression vectors (0.25 µg of each vector per well). The cells were then washed, postincubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum with or without E2 (10 nM) for 24 h, and assayed for luciferase activity. The relative luciferase activity values are means ± SEMs of n = 4 wells. B, Effect of BRCA1-siRNA on activity of S167A mutant ER-. Assays were performed as described above for A. C, Effect of BRCA1-siRNA on activity of S118/167A double-mutant ER-. Assays were performed as described above for A. D, Knockdown of BRCA1 protein levels in DU-145 cells by RNA interference. Subconfluent proliferating cells were treated with BRCA1-siRNA or control-siRNA (100 nM) for 48 h and subjected to Western blotting to detect BRCA1 and -actin (loading control). Con, Control.

Finally, we tested the sensitivity of the S118/167A double-mutant ER- to stimulation by BRCA1-siRNA. In the absence of E2, BRCA1-siRNA conferred about a 4-fold increase (P < 0.001) in ER- (S118/167A) activity and an 8.7-fold increase (P < 0.001) in wt-ER- activity (Fig. 6C). In the presence of E2 (10 nM), BRCA1-siRNA enhanced the activity of wt-ER- by a factor of 2.5-fold (P < 0.001), whereas it only increased the activity of ER- (S118/167A) by a factor of 1.3-fold (Fig. 6C). These findings indicate that the activities of both the unliganded and liganded double-mutant ER- (S118/167A) are stimulated to a smaller extent by BRCA1 knockdown than are the correspondingly activities of the wt-ER-. However, the double mutation does not abrogate the stimulation of ER- activity by BRCA1-siRNA. The knockdown of BRCA1 protein levels attributable to BRCA1-siRNA is illustrated in the Western blot shown in Fig. 6D.

Ability of BRCA1 to Regulate the Phosphorylation Status of ER-
The studies described above suggest that two serine residues within the AF-1 domain of ER- (S118 and S167) that are known to be phosphorylation targets of several growth factor signal transduction pathways are determinants of the sensitivity of the receptor to transcriptional repression by BRCA1. Here, we tested the effect of BRCA1 on the phosphorylation state of these residues as well as serine-104/106. To do this, MCF-7 cells were transfected with wtBRCA1 (vs. empty pcDNA3 vector) or treated with BRCA1-siRNA (vs. control-siRNA) and Western blotted to detect various phosphorylated ER- species using commercial phospho-specific ER- antibodies. In the absence of E2 (to correspond with the ER- transcriptional assays performed in the absence of E2), we found that BRCA1-siRNA (but not control-siRNA) significantly enhanced the levels of phospho-ER- (serine-167), whereas wtBRCA1 had little or no effect on the levels of phospho-ER- (serine-167) (Fig. 7A). None of the treatments significantly altered the total quantity of ER- in the cells. Unlike serine-167, none of the treatments altered the levels of phospho-ER- (serine-118) (Fig. 7B) or phospho-ER- (serine-104/106) (Fig. 7C).

View larger version (36K): [in this window] [in a new window]   Fig. 7. Effect of BRCA1 on Phosphorylation State of ER- A–C, Without E2. Subconfluent proliferating MCF-7 cells were transfected with wtBRCA1 vs. empty pcDNA3 vector or pretreated with BRCA1-siRNA vs. control-siRNA. For transfections, the cells were transfected overnight using 10 µg plasmid DNA per 100-mm dish, washed, and postincubated for 48 h to allow gene expression. For siRNA treatments, the cells were treated with each siRNA (100 nM) for 48 h. The cells were then harvested, and equal aliquots of total cell protein (50 µg) were Western blotted to detect the indicated phosphorylated (P) species of ER- using commercial phospho-specific antibodies or to detect total ER-. D–F, With E2. Assays were performed as above except that, after transfection or siRNA treatment, the cells were treated with E2 (10 nM) for 24 h before being harvested for Western blotting. G, Effect of inhibition of PI3K/Akt signaling on ER- phosphorylation on serine-167. MCF-7 cells were pretreated with BRCA1-siRNA vs. control-siRNA (100 nM) for 48 h. They were then transfected overnight with wtBRCA1 or empty pcDNA3 vector as above, washed, postincubated for 24 h without or with LY294002 (100 µM), and Western blotted for phospho-ER- (serine-167) and total ER-. These assays were performed in the absence of E2. H, Effect of OA on ER- phosphorylation on serine-167. The cell treatments with siRNAs and expression vectors were performed as described above, except that, for the last 10 h of the postincubation period, the cells were incubated with or without OA (5 nM). They were then harvested and Western blotted as above. Assays were performed in the absence of E2. CON, Control.

We tested the effects of inhibiting the PI3K/Akt pathway on phospho-ER- (serine-167) levels in MCF-7 cells. Both DN-Akt and the PI3K inhibitor LY294002 reduced the basal levels of phospho-ER- (serine-167), and both agents also reduced the BRCA1-siRNA-stimulated levels of phospho-ER- (serine-167) (Fig. 7G). This observation correlates with the ability of DN-Akt and LY294002 to inhibit the BRCA1-siRNA-stimulated increase in ER- transcriptional activity (Figs. 2C and 1C, respectively). Finally, we tested whether OA, which mimics the effects of BRCA1-siRNA on PP2A and c-Akt activity, could also enhance phospho-ER- (serine-167) levels. Here, a 10-h treatment with OA caused increased phosphorylation of ER- at serine-167 that was approximately similar in magnitude to that induced by BRCA1-siRNA treatment (Fig. 7H). The effect of the combination of BRCA1-siRNA and OA appeared to be a little greater than that of either agent alone. These findings suggest that PP2A inhibition has a similar effect to BRCA1 knockdown on ER- phosphorylation at serine-167.

Contribution of Other Signaling Pathways to Activation of ER- by BRCA1 Knockdown
Inhibition of the PI3K/Akt signaling pathway by the use of pharmacologic inhibitors (wortmannin or LY294002), a DN-Akt protein, or overexpression of wtPTEN significantly inhibited activation of both the unliganded and liganded ER- by BRCA1-siRNA. Here, we used pharmacologic inhibitors of other growth factor-activated signal transduction pathways to test the role of additional pathways in ER- activation by BRCA1-siRNA. Inhibitors were tested at concentrations reported to be active in other cell systems. First, we tested the ability of a group of inhibitors on the activity of the unliganded ER- in MCF-7 cells treated with BRCA1-siRNA vs. control-siRNA. In the two experiments shown in Fig. 8A, BRCA1-siRNA caused about a 7- to 11-fold increase in ER- activity. Inhibitors of MAPK kinase 1/2 (PD98059 [2-(2-amino-3-methyoxyphenyl)-4H-1-benzopyran-4-one] at 30 µM), protein kinase C [GF109302X (2-[1-(3-dimethylaminopropyl)-1H-indol-3-yl]-3-(1H-indol-3-yl)maleimide) at 30 µM], mammalian target of rapamycin (rapamycin at 10 ng/ml), Src family kinases (PP1 [4-amino-5-(4-methylphenyl)-7-(t-butyl) pyrazolo(3,4-d)-pyrimidine] at 10 µM), and p38 MAPK (SB202190 [4-(4-fluorophenyl)-2-(4-hydroxyphenyl)-5-(4-pyridyl)1H-imidazole] at 10 µM), had little or no effect on BRCA1-siRNA stimulation of ER- activity. However, as in previous experiments, the PI3K inhibitor LY294002 (100 µM) blocked most of the ER- activation secondary to BRCA1-siRNA treatment.

View larger version (35K): [in this window] [in a new window]   Fig. 8. Role of Other Signaling Pathways in BRCA1-siRNA-Mediated Activation of ER- A, Effect of signaling inhibitors on ligand-independent ER- activation by BRCA1-siRNA. MCF-7 cells were pretreated with BRCA1 or control siRNA (100 nM for 48 h), transfected overnight with the ERE-TK-Luc reporter, washed, incubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum with the indicated inhibitor or an equal quantity of vehicle only (DMSO) for 24 h, and harvested for luciferase assays. The two panels shown show two different experiments. The luciferase values are means ± SEMs of four replicate wells. The inhibitors tested (concentrations) were as follows: GF109302X (30 µM), PD98059 (30 µM), rapamycin (10 ng/ml), PP1 (10 µM), SB202190 (10 µM), and LY294002 (100 µM). B, Effect of inhibitors on stimulation of the liganded ER- by BRCA1-siRNA. Assays were performed as above except that, after transfection of ERE-TK-Luc, the 24-h incubation was performed in the presence of E2 (10 nM), along with the indicated pharmacologic inhibitors. Con, Control.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
We studied interactions between growth factor signaling pathways and the tumor suppressor protein BRCA1 that modulate ER- activity using cell lines that do (MCF-7) or do not (DU-145) express endogenous ER-. A major finding of this study is that repression of ER- activity by the endogenous BRCA1 protein is antagonized by PI3K/c-Akt signaling. Thus, knockdown of BRCA1 by RNA interference conferred ligand-independent activation of ER- that was substantially attenuated by pharmacologic inhibitors of PI3K (wortmannin and LY294002). Although it is possible that PI3K inhibition could have a global transcriptional effect that is not specific for BRCA1 or ER-, we note that neither wortmannin nor LY294002 caused cytotoxicity (assessed using MTT assays), and this effect would not explain why the fold reduction of ER- activity was greater in the presence of BRCA1-siRNA. Furthermore, overexpression of PTEN [an upstream inhibitor of Akt activation (31)] and expression of a DN-Akt protein yielded comparable effects with PI3K inhibition. Similar studies in the presence of a physiological concentration of E2 (10 nM) revealed that inhibition of PI3K c-Akt signaling blocked the ability of BRCA1-siRNA to further stimulate the activity of the liganded ER-. However, pharmacologic inhibitors of several other signal transduction pathways failed to block the activation of ER- attributable to knockdown of endogenous BRCA1.

Consistent with the idea that PI3K/c-Akt signaling is required for ligand-independent activation of ER- attributable to BRCA1 knockdown, BRCA1-siRNA caused the phosphorylation and activation of c-Akt and enhanced the phosphorylation of a known c-Akt substrate (GSK-3). The phosphorylation and activation of c-Akt induced by BRCA1-siRNA requires PI3K activity, because it was blocked by pharmacologic inhibitors of PI3K. To our knowledge, the ability of BRCA1 to regulate c-Akt activity has not been described previously. However, several previous studies suggest that heregulin can induce phosphorylation of BRCA1 and down-regulation of its activity and expression through a mechanism involving PI3K and c-Akt (36, 37). These findings raise the possibility of a reciprocal negative regulatory cycle involving BRCA1 and c-Akt. Our findings do not rule out the possibility that PI3K could contribute to ER- activation by BRCA1-siRNA independently of c-Akt. Thus, it was reported that the regulatory subunit of PI3K (p85) interacts directly with ER-, and the interaction mediates some of the rapid nongenomic actions of estrogen (38). A subsequent study described the ability of PI3K to activate ER- by both Akt-dependent and Akt-independent mechanisms (28).

The ability of BRCA1-siRNA to stimulate the phosphorylation and, thus, activation of Akt appears to be attributable, in part, to its ability to inhibit the activity of PP2A, a major cytoplasmic serine/threonine phosphatase that is known to regulate the activity of c-Akt and other protein kinases by dephosphorylating the active kinase (32, 33, 34). Thus, BRCA1-siRNA treatment inhibited the catalytic activity of PP2A to a degree that was almost as great as the selective pharmacologic PP2A inhibitor OA. Consistent with these findings, OA by itself induced the phosphorylation of Akt on serine-473 to an extent that was generally similar to that of BRCA1-siRNA. The mechanism by which BRCA1 knockdown inhibits PP2A activity deserves additional study, because the inhibition of PP2A may have important implications for a variety of cellular signaling pathways. In addition, our findings do not rule out the possibility that BRCA1 knockdown alters the activity of other components or regulators of the PI3K/Akt signaling pathway. Such alterations remain to be determined.

Previous studies establish that wtBRCA1 represses ligand (E2)-stimulated ER- activity and blocks the induction of multiple E2-responsive genes (15, 16, 17, 18, 39, 40). As expected, wtBRCA1 nearly abrogated the E2-induced ER- activity. Here, the repression of E2-induced ER- activity was rescued by exogenous Akt in a manner dependent on its kinase activity. However, Akt had little or no effect on E2-induced ER- activity in the absence of wtBRCA1, and a kinase-dead DN-Akt only modestly reduced the ER- activity, suggesting that, in the presence of E2, Akt is not required for ER- activity, except when the receptor is in a BRCA1-repressed state. In this case, the rescue activity of Akt might be attributable to functional inactivation of BRCA1 through the Akt kinase activity (36) or to an effect of the Akt kinase or kinases downstream of Akt on the ER- (see below) that might render the receptor resistant to repression by BRCA1. Consistent with the latter idea, wtBRCA1 did not block the ability of exogenous Akt to induce ligand-independent activation of ER-.

Serine-118 within the AF-1 domain of ER- has been found to be a phosphorylation target for the ERK1/2 MAPKs, cyclin-dependent kinase 7, and Pak1 (p21/Cdc42/Rac1-activated kinase 1) that is required for maximal activity of ER- (27, 41, 42, 43). This residue may also regulate the ubiquitin-mediated degradation of ER- (44), and it may be a target for Akt-mediated phosphorylation, along with serine-167, which appears to be the major site for phosphorylation mediated by c-Akt (28, 45). Although the liganded mutant ER- (S118A) showed about two thirds of the activity of the liganded wt-ER-, wtBRCA1 caused significantly greater fold repression and BRCA1-siRNA caused significantly greater fold activation in the wild-type receptor compared with the mutant receptor in DU-145 cells. Similar findings were obtained using the S167A mutant ER-, and the double-mutant ER- (S118/S167) was even more refractory to regulation by BRCA1. IP-Western blot experiments revealed that the S118A and S167A mutants retained the ability to associate with BRCA1, whereas the S118/167A double mutant showed significantly reduced association with the BRCA1 protein. These findings suggest that the physical association of BRCA1 with ER- contributes to the crosstalk between BRCA1 and ER- in modulating the activity of ER-.

It was also apparent that the various mutant receptors were not fully refractory to regulation by BRCA1, except for the finding that exogenous BRCA1 did not inhibit the activity of the liganded double-mutant receptor. These findings are consistent with a model in which additional ER- phosphorylation sites (e.g. serine-104 and serine-106) (26, 46) are also regulated through BRCA1 or a model in which BRCA1 can only partially control the phosphorylation status of serine-118 and serine-167. Interestingly, whereas the activity of the liganded S118/167A double-mutant ER- was not inhibited by wtBRCA1, the activity of the unliganded receptor was enhanced by BRCA1 knockdown, although to a lesser degree than the wt-ER-. These findings suggest that the pathways through which BRCA1 regulates the liganded ER- are not identical to those through which it regulates the unliganded receptor. They further suggest that phosphorylation events within the AF-1 domain of ER-, at least in part, control the sensitivity of the receptor to repression by BRCA1.

In previous studies, we established that BRCA1 binding to ER- involves one (or possibly two) contact sites within the AF-2 domain but does not involve the AF-1 domain or the DNA-binding domain of ER- (16, 17). We further showed that wtBRCA1 represses the E2-inducible activity of the AF-2 domain but does not inhibit the constitutively active AF-1 domain, in assays using galactosidase-4-ER- AF-1 or AF-2 constructs and a galactosidase-4 luciferase reporter (15, 18). We hypothesized that, in the intact ER-, the BRCA1-bound AF-2 domain mediates the repression of AF-1, which is not observed when AF-1 activity is tested in the absence of AF-2. The observations in the present study are consistent with this hypothesis and with the idea that conformational changes in the AF-1 domain caused by phosphorylation mediated by growth factor signal transduction pathways may render ER- more refractory to repression by BRCA1.

Using phospho-specific antibodies against ER-, we found that endogenous BRCA1 negatively regulates the phosphorylation of ER- on serine-167 within the AF-1 activation domain but does not regulate phosphorylation on serine residues 118, 104, or 106. Knockdown of BRCA1 stimulated ER- phosphorylation on serine-167, the major serine residue targeted by c-Akt, in both the absence and presence of E2. Our findings are consistent with a model in which the loss of BRCA1 induces or stimulates ER- activity predominantly through enhanced phosphorylation of ER- on serine-167 attributable to an increase in the activity of the c-Akt kinase. Consistent with this model, blockade of the PI3K/Akt signaling pathway using the PI3K inhibitor LY294002 or using a DN-Akt expression vector reduced the levels of phospho-ER- (serine-167). Conversely, treatment with OA, an inhibitor of PP2A activity, enhanced the levels of phospho-ER- (serine-167) similar to the effect of BRCA1-siRNA. This effect may have been attributable to increased Akt activity attributable to inhibition of PP2A but also could have resulted, in part, in direct regulation of the ER- phosphorylation state by PP2A.

The findings from this study have potential clinical implications. Thus, although most BRCA1 mutant breast cancers are ER- negative, a small percentage are ER- positive, and the growth of these tumors might be driven by a hyperactive ER- attributable to the absence of functional BRCA1 (19). Probably more importantly, a significant percentage of sporadic breast cancers (30–40%) exhibit absent or decreased BRCA1 expression (4, 6), raising the possibility that patients with "BRCA1-deficient" ER--positive cancers might benefit from inhibition of the PI3K/Akt pathway. However, whether the deficiency of BRCA1 observed in a subset of sporadic breast cancers is clinically significant and can be targeted therapeutically remains to be determined.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 
Cell Lines and Culture
Human breast cancer (MCF-7) and prostate cancer (DU-145) cell lines were obtained from the American Type Culture Collection (Manassas, VA) and cultured as described previously (16, 17). Briefly, the cells were grown in DMEM supplemented with 5% (vol/vol) fetal calf serum, L-glutamine (5 mM), nonessential amino acids (5 mM), penicillin (100 U/ml), and streptomycin (100 µg/ml) (all obtained from BioWhittaker, Walkersville, MD).

Reagents
E2 and MTT dye were obtained from the Sigma (St. Louis, MO). The signal transduction inhibitors used in this study and their sources were as follows: wortmannin (Biomol Research Laboratories, Plymouth Meeting, PA); LY294002 (Sigma); PD98059 (Biomol); GF109302X (LC Laboratories, Woburn, MA); rapamycin (Biomol); PP1 (Biomol); and SB202190 (Calbiochem, La Jolla, CA). These agents were dissolved in DMSO and diluted into culture medium at the time of experiments. The antibodies used for IP and Western blotting are described below.

Expression Vectors and Reporters
The wtBRCA1 expression vector was created by cloning the BRCA1 cDNA into the pcDNA3 vector (Invitrogen, Carlsbad, CA) using artificially engineered 5' HindIII and 3' NotI sites (47). The ER- expression vector pCMV-ER- was used to express wt-ER-. Expression vectors for point mutant ER- proteins (S167A, S118A, and S118/167A) were created by site-directed mutagenesis of wt-ER-, using the QuikChange Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA). The estrogen-responsive reporter ERE-TK-Luc is composed of the vitellogenin A2 estrogen-responsive enhancer (ERE) controlling a minimal thymidine kinase promoter (TK81) and luciferase, in plasmid pGL2 (30). Assays of ER- transcriptional activity are described below.

wtAkt and DN-Akt mutant (K179A) in the pCIS2 expression vector were provided by Dr. M. J. Quon (National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, MD) (48). Human wtPTEN in the pFLAG-CMV vector was provided by Dr. M. Georgescu (Rockefeller University, New York, NY) (49).

Transient Transfections
For transient transfections, subconfluent proliferating cells were transfected overnight with the vector of interest or with the empty pcDNA3 vector (Invitrogen) (10 µg of plasmid DNA per 100-mm dish) using Lipofectamine (Invitrogen) and then washed to remove the excess vector and Lipofectamine. To determine the transfection efficiency, cultures were cotransfected with plasmid pRSV-ß-gal (Promega, Madison, WI) to allow staining with X-gal reagent and visualization of transfected (blue staining) cells.

siRNAs
Double-stranded siRNA to knock down BRCA1 protein levels (BRCA1-siRNA) and a control-siRNA that is not homologous to human DNA sequences were chemically synthesized by Dharmacon (Lafayette, CO). The efficacy of the BRCA1-siRNA was validated previously (22, 50). In this study, we used a combination of four different BRCA1-siRNAs at a total concentration of 100 nM. The sequences of these BRCA1-siRNAs were as follows: BRCA1-siRNA-1, 5'-CAGCTACCCTTCCATCATA-3'; BRCA1-siRNA-2, 5'-GGGATACCATGCAACATAA-3'; BRCA1-siRNA-3, 5'-GAAGGAGCTTTCATCATTC-3'; and BRCA1-siRNA-4, 5' CTAGAAATCTG TTGCTATG-3'. For transfection of siRNAs, subconfluent proliferating cells were treated with a BRCA1-siRNA-1,2,3,4 mixture (25 nM per siRNA, total concentration of 100 nM) or control-siRNA (100 nM) using siPORT Amine transfection reagent (Ambion, Austin, TX) as per the instructions of the manufacturer. Western blotting experiments revealed that a minimum of 48-h exposure to siRNAs was required to obtain a substantial reduction of the BRCA1 protein levels.

Assays of ER Transcriptional Activity
Subconfluent proliferating cells in 24-well dishes were incubated overnight with 0.25 µg of each indicated vector in serum-free DMEM containing Lipofectamine (Invitrogen). The total transfected DNA was kept constant by addition of the appropriate control vector. The cells were washed, incubated in phenolphthalein-free DMEM containing 5% charcoal-stripped serum (obtained from the Tissue Culture Shared Resource of the Lombardi Comprehensive Cancer Center) (0.2 ml per well) with or without E2 (10 nM) for 24 h, and harvested for luciferase assays. For each assay condition in each experiment, four replicate wells were tested.

MTT Assay of Cell Viability
The MTT assay is based on the ability of viable mitochondria to convert MTT, a soluble tetrazolium salt, into an insoluble formazan precipitate, which is dissolved in DMSO and quantitated by spectrophotometry (51). After the indicated treatment, cells in 96-well dishes were tested for MTT dye conversion. The cell viability was calculated as the amount of MTT dye conversion relative to sham-treated control cells and expressed as the mean ± SEM of 10 replicate wells for each cell treatment.

Measurement of Akt Kinase Activity
Akt kinase activity was measured using an Akt Activity Assay Kit purchased from Calbiochem. Assays were performed according to the instructions supplied by the manufacturer. This assay is based on the ability of Akt immunoprecipitated from whole-cell lysates (200 µg of cell protein per assay) to phosphorylate a protein substrate (GSK-3) in vitro under standardized reaction conditions. The phosphorylated GSK-3 is detected by Western blotting using a phospho-specific antibody directed against phospho-GSK-3 (serine-21).

PP2A Activity Assay
PP2A activity was measured using a serine/threonine phosphatase assay kit (Upstate Biotechnology, Lake Placid, NY), according to the instructions of the manufacturer. Briefly, subconfluent proliferating MCF-7 cells were treated with BRCA1 (vs. control) siRNA or transfected with wtBRCA1 (vs. empty pcDNA3 vector) and then treated with or without OA (5 nM) (a selective pharmacologic inhibitor of PP2A) or with or without LY294002 (100 µM) for the last 10 h, as described in the figure legends. After these treatments, the cells were harvested, and whole lysates were prepared using radioimmunoprecipitation assay buffer. The total PP2A enzymatic activity was measured in aliquots of whole-cell lysate containing 400 µg of protein, after IP of the PP2A catalytic subunit using anti-PP2A antibody clone 1D6. A specific phospho-Akt/PKB peptide was used as the substrate. The phosphatase reactions were performed for 15 min at 30 C. The release of phosphate from the added phosphopeptide was quantified using a malachite green reagent. Changes in absorbance were measured at 650 nm in a Universal Microplate Reader (Bio-Tek Instruments, Winooski, VT). PP2A activity was calculated as the ratio between released free phosphate measured as absorbance at 650 nm with the phosphatase assay kit. The PP2A activity values were normalized to the control values (no siRNA, expression vector, OA, or LY294002) and expressed as the means ± SEMs of three independent experiments.

IP
To study the in vivo association of BRCA1 with different ER- proteins, subconfluent proliferating DU-145 cells were transfected overnight with wt-ER-, S167A mutant ER-, S118A mutant ER-, S118/167A double-mutant ER-, or empty pcDNA3 vector (10 µg of plasmid DNA per 100-mm culture dish) and postincubated for 24 h to allow gene expression. The cells were then harvested, and whole-cell extracts were prepared as described previously, using radioimmunoprecipitation assay buffer (16, 17). Each IP was performed using 2 µg antibody and 500 µg extract protein. The extracts were incubated with a combination of anti-BRCA1 mouse monoclonals (Ab-1, Ab-2, and Ab-3; Oncogene Research Products, San Diego, CA) or anti-ER- H184 (rabbit polyclonal IgG, sc-7207; Santa Cruz Biotechnology, Santa Cruz, CA). The precipitated proteins were collected using protein A/G agarose (Santa Cruz Biotechnology). After low-speed centrifugation to remove the supernatants, the agarose was washed with PBS, incubated in boiling Laemli’s sample buffer, and subjected to SDS-PAGE and Western blotting.

Western Blotting
Equal aliquots of whole-cell protein (either 50 µg unprecipitated whole-cell extract or the immunoprecipitated protein from 500 µg whole-cell protein) were electrophoresed on 4–12% SDS-polyacrylamide gradient gels, transferred to nitrocellulose membranes (Millipore, Bedford, MA), and blotted using primary antibodies directed against the following: BRCA1 (C-20, rabbit polyclonal, 1:200 dilution; Santa Cruz Biotechnology); ER- (F10, mouse monoclonal, sc-8002, 1:500 dilution; Santa Cruz Biotechnology); phospho-ER- (serine-118) (1:300; Santa Cruz Biotechnology); phospho-ER- (serine-167) (1:1000; Santa Cruz Biotechnology); phospho-ER- (serine-104/106) (1:800; Santa Cruz Biotechnology); phospho-Akt (serine-473) (catalog no. 9271S, 1:500; Cell Signaling Technology, Beverly, MA); total Akt (catalog no. 9271, 1:500; Cell Signaling Technology); phospho-GSK-3ß (serine-9) (1:600; Cell Signaling Technology); total GSK-3ß (1:300; Cell Signaling Technology); phospho-ATM (serine-1981) (sc-47739, 1:200; Santa Cruz Biotechnology); total ATM (mouse monoclonal; Santa Cruz Biotechnology); or -actin (goat polyclonal, 1:400; Santa Cruz Biotechnology). Methodologic details have been published previously (17, 22). The membranes were blotted with appropriate secondary antibodies (1:1000; Santa Cruz Biotechnology); and blotted proteins were visualized using the enhanced chemiluminescence detection system (GE Healthcare, Little Chalfont, UK), with colored size markers (Bio-Rad, Hercules, CA).

Statistical Methods
When appropriate, statistical comparisons were made using the two-tailed Student’s t test.

	FOOTNOTES

 
This work was supported in part by United States Public Health Service Grants R01-CA82599 and R01-CA80000 and by Susan G. Komen Breast Cancer Foundation Grant BCTR0503799 (to E.M.R.).

Disclosure statement: The authors have nothing to disclose.

First Published Online May 15, 2007

Abbreviations: AF, Activation function; ATM, ataxia-telangiectasia mutated protein; BRCA, breast cancer gene; DMSO, dimethylsulfoxide; DN, dominant negative; E2, 17ß-estradiol; ER, estrogen receptor; ERE, estrogen-responsive enhancer; GSK-3ß, glycogen synthase kinase 3ß; IP, immunoprecipitation; OA, okadaic acid; PI3K, phosphatidylinositol-3 kinase; PP2A, protein phosphatase 2A; siRNA, small interfering RNA.

Received for publication September 22, 2006. Accepted for publication May 11, 2007.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS REFERENCES

 

1. Miki Y, Swensen J, Shattuck-Eidens D, Futreal PA, Harshman K, Tavtigian S, Liu Q, Cochran C, Bennett LM, Ding W, Bell R, Rosenthal J, Hussey C, Tran T, McClure M, Frye C, Hattier T, Phelps R, Haugen-Strano A, Katcher Y, Yakumo K, Gholami Z, Shaffer D, Stone S, Bayer S, Wray C, Bogden R, Dayananth P, Ward J, Tonin P, Narod S, Bristow PK, Norris FH, Helvering L, Morrison P, Rosteck P, Lai M, Barrett JC, Lewis C, Neuhausen S, Cannon-Albright L, Goldgar D, Wiseman R, Kamb A, Skolnick MH 1994 A strong candidate for the breast and ovarian cancer susceptibility gene BRCA1. Science 266:66–71[Abstract/Free Full Text]
2. Tavtigian SV, Simard J, Rommens J, Couch F, Shattuck-Eidens D, Neuhausen S, Merajver S, Thorlacius S, Offit K, Stoppa-Lyonnet D, Belanger C, Bell R, Berry S, Bogden R, Chen Q, Davis T, Dumont M, Frye C, Hattier T, Jammulapati S, Janecki T, Jiang P, Kehrer R, Leblanc JF, Mitchell JT, McArthur-Morrison J, Nguyen K, Peng Y, Samson C, Schroeder M, Snyder SC, Steele L, Stringfellow M, Stroup C, Swedlund B, Swense J, Teng D, Thomas A, Tran T, Tranchant M, Weaver-Feldhaus J, Wong AK, Shizuya H, Eyfjord JE, Cannon-Albright L, Tranchant M, Labrie F, Skolnick MH, Weber B, Kamb A, Goldgar DE 1996 The complete BRCA2 gene and mutations in chromosome 13q-linked kindreds. Nat Genet 1:333–337
3. Antoniou AC, Easton DF 2006 Risk prediction models for familial breast cancer. Future Oncol 2006 2:257–274[CrossRef][Medline]
4. Wilson CA, Ramos L, Villasenor MR, Anders KH, Press MF, Clarke K, Karlan B, Chen JJ, Scully R, Livingston D, Zuch RH, Kanter MH, Cohen S, Calzone FJ, Slamon DJ 1999 Localization of human BRCA1 and its loss in high-grade, non-inherited breast carcinomas. Nat Genet 21:236–240[CrossRef][Medline]
5. Esteller M, Silva JM, Dominguez G, Bonilla F, Matias-Guiu X, Lerma E, Bussaglia E, Prat J, Harkes IC, Repasky EA, Gabrielson E, Schutte M, Baylin SB, Herman JG 2000 Promoter hypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors. J Natl Cancer Inst 92:564–569[Abstract/Free Full Text]
6. Staff S, Isola J, Tanner M 2003 Haplo-insufficiency of BRCA1 in sporadic breast cancer. Cancer Res 63:4978–4983[Abstract/Free Full Text]
7. Wei M, Grushko TA, Dignam J, Hagos F, Nanda R, Sveen L, Xu J, Fackenthal J, Tretiakova M, Das S, Olopade OI 2005 BRCA1 promoter methylation in sporadic breast cancer is associated with reduced BRCA1 copy number and chromosome 17 aneusomy. Cancer Res 65:10692–10699[Abstract/Free Full Text]
8. Polanowska J, Martin JS, Garcia-Muse T, Petalcorin MI, Boulton SJ 2006 A conserved pathway to activate BRCA1-dependent ubiquitylation at DNA damage sites. EMBO J 25:2178–2188[CrossRef][Medline]
9. Rosen EM, Fan S, Pestell RG, Goldberg ID 2003 BRCA1 gene in breast cancer. J Cell Physiol 196:19–41[CrossRef][Medline]
10. Scully R, Xie A, Nagaraju G 2004 Molecular functions of BRCA1 in the DNA damage response. Cancer Biol Ther 3:521–527[Medline]
11. Tirkkonen M, Johannsson O, Agnarsson BA, Olsson H, Ingvarsson S, Karhu R, Tanner M, Isola J, Barkardottir RB, Borg A, Kallioniemi OP 1997 Distinct somatic genetic changes associated with tumor progression in carriers of BRCA1 and BRCA2 germ-line mutations. Cancer Res 57:1222–1227[Abstract/Free Full Text]
12. Weaver Z, Montagna C, Xu X, Howard T, Gadina M, Brodie SG, Deng CX, Ried T 2002 Mammary tumors in mice conditionally mutant for Brca1 exhibit gross genomic instability and centrosome amplification yet display a recurring distribution of genomic imbalances that is similar to human breast cancer. Oncogene 21:5097–5107[CrossRef][Medline]
13. Levitt NC, Hickson ID 2002 Caretaker tumour suppressor genes that defend genome integrity. Trends Mol Med 8:179–186[CrossRef][Medline]
14. Thompson D, Easton DF; Breast Cancer Linkage Consortium 2002 Cancer incidence in BRCA1 mutation carriers. J Natl Cancer Inst 94:1358–1365[Abstract/Free Full Text]
15. Fan S, Wang J, Yuan R, Ma Y, Meng Q, Erdos MR, Pestell RG, Yuan F, Auborn KJ, Goldberg ID, Rosen EM 1999 BRCA1 inhibition of estrogen receptor signaling in transfected cells. Science 284:1354–1356[Abstract/Free Full Text]
16. Fan S, Ma YX, Wang C, Yuan RQ, Meng Q, Wang JA, Erdos M, Goldberg ID, Webb P, Kushner PJ, Pestell RG, Rosen EM 2001 Role of direct interaction in BRCA1 inhibition of estrogen receptor activity. Oncogene 20:77–87[CrossRef][Medline]
17. Ma YX, Tomita Y, Fan S, Wu K, Tong Y, Zhao Z, Song LN, Goldberg ID, Rosen EM 2005 Structural determinants of the BRCA1:estrogen receptor interaction. Oncogene 24:1831–1846[CrossRef][Medline]
18. Fan S, Ma YX, Wang C, Yuan RQ, Meng Q, Wang JA, Erdos M, Goldberg ID, Webb P, Kushner PJ, Pestell RG, Rosen EM 2002 p300 Modulates the BRCA1 inhibition of estrogen receptor activity. Cancer Res 62:141–151[Abstract/Free Full Text]
19. Rosen EM, Fan S, Isaacs C 2005 BRCA1 in hormonal carcinogenesis: basic and clinical research. Endocr Relat Cancer 12:533–548[Abstract/Free Full Text]
20. Kawai H, Li H, Chun P, Avraham S, Avraham HK 2002 Direct interaction between BRCA1 and the estrogen receptor regulates vascular endothelial growth factor (VEGF) transcription and secretion in breast cancer cells. Oncogene 21:7730–7739[CrossRef][Medline]
21. Zheng L, Annab LA, Afshari CA, Lee WH, Boyer TG 2001 BRCA1 mediates ligand-independent transcriptional repression of the estrogen receptor. Proc Natl Acad Sci USA 98:9587–9592[Abstract/Free Full Text]
22. Jones LP, Li M, Halama ED, Ma Y, Lubet R, Grubbs CJ, Deng CX, Rosen EM, Furth PA 2005 Promotion of mammary cancer development by tamoxifen in a mouse model of Brca1-mutation-related breast cancer. Oncogene 24:3554–3562[CrossRef][Medline]
23. Kato S, Endoh H, Masuhiro Y, Kitamoto T, Uchiyama S, Sasaki H, Masushige S, Gotoh Y, Nishida E, Kawashima H, Metzger D, Chambon P 1995 Activation of the estrogen receptor through phosphorylation by mitogen-activated protein kinase. Science 270:1491–1494[Abstract/Free Full Text]
24. Bunone G, Briand PA, Miksicek RJ, Picard D 1996 Activation of the unliganded estrogen receptor by EGF involves the MAP kinase pathway and direct phosphorylation. EMBO J 15:2174–2183[Medline]
25. Joel PB, Smith J, Sturgill TW, Fisher TL, Blenis J, Lannigan DA 1998 pp90rsk1 regulates estrogen receptor-mediated transcription through phosphorylation of Ser-167. Mol Cell Biol 18:1978–1984[Abstract/Free Full Text]
26. Rogatsky I, Trowbridge JM, Garabedian MJ 1999 Potentiation of human estrogen receptor transcriptional activation through phosphorylation of serines 104 and 106 by the cyclin A-CDK2 complex. J Biol Chem 274:22296–22302[Abstract/Free Full Text]
27. Chen D, Riedl T, Washbrook E, Pace PE, Coombes RC, Egly JM, Ali S 2000 Activation of estrogen receptor by S118 phosphorylation involves a ligand-dependent interaction with TFIIH and participation of CDK7. Mol Cell 6:127–137[CrossRef][Medline]
28. Campbell RA, Bhat-Nakshatri P, Patel NM, Constantinidou D, Ali S, Nakshatri H 2001 Phosphatidylinositol 3-kinase/AKT-mediated activation of estrogen receptor : a new model for anti-estrogen resistance. J Biol Chem 276:9817–9824[Abstract/Free Full Text]
29. Lannigan DA 2003 Estrogen receptor phosphorylation. Steroids 68:1–9[CrossRef][Medline]
30. Henttu PM, Kalkhoven E, Parker MG 1997 AF-2 activity and recruitment of steroid receptor coactivator 1 to the estrogen receptor depend on a lysine residue conserved in nuclear receptors. Mol Cell Biol 17:1832–1839[Abstract/Free Full Text]
31. Ramaswamy S, Nakamura N, Vazquez F, Batt DB, Perera S, Roberts TM, Sellers WR 1999 Regulation of G1 progression by the PTEN tumor suppressor protein is linked to inhibition of the phosphatidylinositol 3-kinase/Akt pathway. Proc Natl Acad Sci USA 96:2110–2115[Abstract/Free Full Text]
32. Millward TA, Zolnierowicz S, Hemmings BA 1999 Regulation of protein kinase cascades by protein phosphatase 2A. Trends Biochem Sci 24:186–191 (Review)[CrossRef][Medline]
33. Sato S, Fujita N, Tsuruo T 2000 Modulation of Akt kinase activity by binding to Hsp90. Proc Natl Acad Sci USA 97:10832–10837[Abstract/Free Full Text]
34. Pankov R, Cukierman E, Clark K, Matsumoto K, Hahn C, Poulin B, Yamada KM 2003 Specific ß1 integrin site selectively regulates Akt/protein kinase B signaling via local activation of protein phosphatase 2A. J Biol Chem 278:18671–18681[Abstract/Free Full Text]
35. Goodarzi AA, Jonnalagadda JC, Douglas P, Young D, Ye R, Moorhead GB, Lees-Miller SP, Khanna KK 2004 Autophosphorylation of ataxia-telangiectasia mutated is regulated by protein phosphatase 2A. EMBO J 23:4451–4461[CrossRef][Medline]
36. Altiok S, Batt D, Altiok N, Papautsky A, Downward J, Roberts TM, Avraham H 1999 Heregulin induces phosphorylation of BRCA1 through phosphatidylinositol 3-Kinase/AKT in breast cancer cells. J Biol Chem 274:32274–32278[Abstract/Free Full Text]
37. Miralem T, Avraham HK 2003 Extracellular matrix enhances heregulin-dependent BRCA1 phosphorylation and suppresses BRCA1 expression through its C terminus. Mol Cell Biol 23:579–593[Abstract/Free Full Text]
38. Simoncini T, Hafezi-Moghadam A, Brazil DP, Ley K, Chin WW, Liao JK 2000 Interaction of oestrogen receptor with the regulatory subunit of phosphatidylinositol-3-OH kinase. Nature 407:538–541[CrossRef][Medline]
39. Xu J, Fan S, Rosen EM 2005 Regulation of the estrogen-inducible gene expression profile by the breast cancer susceptibility gene BRCA1. Endocrinology 146:2031–2047[Abstract/Free Full Text]
40. Wang C, Fan S, Li Z, Fu M, Rao M, Ma Y, Lisanti MP, Albanese C, Katzenellenbogen BS, Kushner PJ, Weber B, Rosen EM, Pestell RG 2005 Cyclin D1 antagonizes BRCA1 repression of estrogen receptor activity. Cancer Res 65:6557–6567[Abstract/Free Full Text]
41. Chen D, Washbrook E, Sarwar N, Bates GJ, Pace PE, Thirunuvakkarasu V, Taylor J, Epstein RJ, Fuller-Pace FV, Egly JM, Coombes RC, Ali S 2002 Phosphorylation of human estrogen receptor at serine 118 by two distinct signal transduction pathways revealed by phosphorylation-specific antisera. Oncogene 21:4921–4931[CrossRef][Medline]
42. Feng W, Webb P, Nguyen P, Liu X, Li J, Karin M, Kushner PJ 2001 Potentiation of estrogen receptor activation function 1 (AF-1) by Src/JNK through a serine 118-independent pathway. Mol Endocrinol 15:32–45[Abstract/Free Full Text]
43. Rayala SK, Talukder AH, Balasenthil S, Tharakan R, Barnes CJ, Wang RA, Aldaz M, Khan S, Kumar R 2005 P21-activated kinase 1 regulation of estrogen receptor- activation involves serine 305 activation linked with serine 118 phosphorylation. Cancer Res 66:1694–1701[CrossRef]
44. Valley CC, Metivier R, Solodin NM, Fowler AM, Mashek MT, Hill L, Alarid ET 2005 Differential regulation of estrogen-inducible proteolysis and transcription by the estrogen receptor N terminus. Mol Cell Biol 25:5417–5428[Abstract/Free Full Text]
45. Martin MB, Franke TF, Stoica GE, Chambon P, Katzenellenbogen BS, Stoica BA, McLemore MS, Olivo SE, Stoica A 2000 A role for Akt in mediating the estrogenic functions of epidermal growth factor and insulin-like growth factor I. Endocrinology 141:4503–4511[Abstract/Free Full Text]
46. Le Goff P, Montano MM, Schodin DJ, Katzenellenbogen BS 1994 Phosphorylation of the human estrogen receptor. Identification of hormone-regulated sites and examination of their influence on transcriptional activity. J Biol Chem 269:4458–4466[Abstract/Free Full Text]
47. Fan S, Wang JA, Yuan RQ, Ma YX, Meng Q, Erdos MR, Brody LC, Goldberg ID, Rosen EM 1998 BRCA1 as a potential human prostate tumor suppressor: modulation of proliferation, damage responses and expression of cell regulatory proteins. Oncogene 16:3069–3082[CrossRef][Medline]
48. Cong LN, Chen H, Li Y, Zhou L, McGibbon MA, Taylor SI, Quon MJ 1997 Physiological role of Akt in insulin-stimulated translocation of GLUT4 in transfected rat adipose cells. Mol Endocrinol 11:1881–1890[Abstract/Free Full Text]
49. Zhong H, Chiles K, Feldser D, Laughner E, Hanrahan C, Georgescu MM, Simons JW, Semenza GL 2000 Modulation of hypoxia-inducible factor 1 expression by the epidermal growth factor/phosphatidylinositol 3-kinase/PTEN/AKT/FRAP pathway in human prostate cancer cells: implications for tumor angiogenesis and therapeutics. Cancer Res 60:1541–1545[Abstract/Free Full Text]
50. Xiong J, Fan S, Meng Q, Schramm L, Wang C, Bouzahza B, Zhou J, Zafonte B, Goldberg ID, Haddad BR, Pestell RG, Rosen EM 2003 BRCA1 inhibition of telomerase activity in cultured cells. Mol Cell Biol 23:8668–8690[Abstract/Free Full Text]
51. Alley MC, Scudiero DA, Monks A, Hursey ML, Czerwinski MJ, Fine DL, Abbott BJ, Mayo JG, Shoemaker RH, Boyd MR 1988 Feasibility of drug screening with panels of human tumor cell lines using a microculture tetrazolium assay. Cancer Res 48:589–601[Abstract/Free Full Text]

NURSA Molecule Pages Link:

Nuclear Receptors:   ERα

Coregulators:   PTEN  |  BRCA1

Ligands:   17β-Estradiol

This article has been cited by other articles:

				Y. Ma, S. Fan, C. Hu, Q. Meng, S. A. Fuqua, R. G. Pestell, Y. A. Tomita, and E. M. Rosen BRCA1 Regulates Acetylation and Ubiquitination of Estrogen Receptor-{alpha} Mol. Endocrinol., January 1, 2010; 24(1): 76 - 90. [Abstract] [Full Text] [PDF]

				P. Katiyar, Y. Ma, A. Riegel, S. Fan, and E. M. Rosen Mechanism of BRCA1-Mediated Inhibition of Progesterone Receptor Transcriptional Activity Mol. Endocrinol., August 1, 2009; 23(8): 1135 - 1146. [Abstract] [Full Text] [PDF]

				T. Xiang, A. Ohashi, Y. Huang, T. K. Pandita, T. Ludwig, S. N. Powell, and Q. Yang Negative Regulation of AKT Activation by BRCA1 Cancer Res., December 15, 2008; 68(24): 10040 - 10044. [Abstract] [Full Text] [PDF]

				V. C. Jordan Estrogen Receptors in BRCA1-Mutant Breast Cancer: Now You See Them, Now You Don't J Natl Cancer Inst, November 21, 2007; 99(22): 1655 - 1657. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Submit a related Letter to the Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager NURSA Molecule Pages Link Reprints, Permissions and Rights Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Ma, Y. Articles by Rosen, E. M. Search for Related Content PubMed PubMed Citation Articles by Ma, Y. Articles by Rosen, E. M.