Role of RBP2-Induced ER and IGF1R-ErbB Signaling in Tamoxifen Resistance in Breast Cancer

Role of RBP2-Induced ER and IGF1R-ErbB Signaling in Tamoxifen Resistance in Breast Cancer Abstract Background Despite the benefit of endocrine therapy, acquired resistance during or after treatment still remains a major challenge in estrogen receptor (ER)–positive breast cancer. We investigated the potential role of histone demethylase retinoblastoma-binding protein 2 (RBP2) in endocrine therapy resistance of breast cancer. Methods Survival of breast cancer patients according to RBP2 expression was analyzed in three different breast cancer cohorts including METABRIC (n = 1980) and KM plotter (n = 1764). RBP2-mediated tamoxifen resistance was confirmed by in vitro sulforhodamine B (SRB) colorimetric, colony-forming assays, and in vivo xenograft models (n = 8 per group). RNA-seq analysis and receptor tyrosine kinase assay were performed to identify the tamoxifen resistance mechanism by RBP2. All statistical tests were two-sided. Results RBP2 was associated with poor prognosis to tamoxifen therapy in ER-positive breast cancer (P = .04 in HYU cohort, P = .02 in KM plotter, P = .007 in METABRIC, log-rank test). Furthermore, RBP2 expression was elevated in patients with tamoxifen-resistant breast cancer (P = .04, chi-square test). Knockdown of RBP2 conferred tamoxifen sensitivity, whereas overexpression of RBP2 induced tamoxifen resistance in vitro and in vivo (MCF7 xenograft: tamoxifen-treated control, mean [SD] tumor volume = 70.8 [27.9] mm3, vs tamoxifen-treated RBP2, mean [SD] tumor volume = 387.9 [85.1] mm3, P < .001). Mechanistically, RBP2 cooperated with ER co-activators and corepressors and regulated several tamoxifen resistance–associated genes, including NRIP1, CCND1, and IGFBP4 and IGFBP5. Furthermore, epigenetic silencing of IGFBP4/5 by RBP2-ER-NRIP1-HDAC1 complex led to insulin-like growth factor–1 receptor (IGF1R) activation. RBP2 also increased IGF1R-ErbB crosstalk and subsequent PI3K-AKT activation via demethylase activity–independent ErbB protein stabilization. Combinational treatment with tamoxifen and PI3K inhibitor could overcome RBP2-mediated tamoxifen resistance (RBP2-overexpressing cells: % cell viability [SD], tamoxifen = 89.0 [3.8]%, vs tamoxifen with BKM120 = 41.3 [5.6]%, P < .001). Conclusions RBP2 activates ER-IGF1R-ErbB signaling cascade in multiple ways to induce tamoxifen resistance, suggesting that RBP2 is a potential therapeutic target for ER-driven cancer. Retinoblastoma-binding protein 2 (RBP2, also known as KDM5A) is a demethylase toward histone H3 lysine 4 di- and tri-methylation (H3K4me2/3) (1–3). RBP2 acts as a gene silencer via H3K4 demethylation (4–8), but it is also involved in transcriptional activation (4,9,10). RBP2 plays an important role in cancer development. RBP2 is amplified/overexpressed in various human tumors including breast and lung cancers (10–13) and induces tumorigenesis, epithelial-to-mesenchymal transition, metastasis, angiogenesis, and drug tolerance (10–18). It also interacts with estrogen receptor (ER) (9); however, its functional and clinical relevance in ER-driven cancer remains unclear. ER is the most important biomarker and therapeutic target in human breast cancer (19). There are several distinct pathways leading to ER activation, such as estrogen-dependent/-independent, genomic, and nongenomic pathways (20,21). For targeting ER-positive (ER+) breast cancer, ER signaling is inhibited via various routes, such as blocking estrogen biosynthesis, suppressing ER expression, or competing with estrogen (20). Among endocrine agents, the anti-estrogen tamoxifen is the most widely used for ER+ breast cancer treatment (22). Despite the benefits, many patients undergo endocrine resistance via several mechanisms, including altered expression or activity of ER and the activation of receptor tyrosine kinases (RTKs) or other bypass signaling pathways (23–27). In particular, ER+ breast cancer with acquired resistance to tamoxifen can use both estrogen and tamoxifen as the growth stimulus (28–30), yet the mechanism for the agonist action of tamoxifen has not been fully established. Furthermore, long-term treatment with anti-estrogens develops new mechanisms of resistance that are distinct from resistance mechanisms that occur by short-term treatment (30,31), implying the complexity of endocrine therapy resistance. In this study, we explored the potential oncogenic role of RBP2 in ER-positive breast cancer to identify the novel molecular mechanism of tamoxifen resistance in breast cancer. Methods Patients and Surgical Specimens Patients with invasive ductal carcinoma (n = 200) who successfully underwent surgery at Hanyang University Hospital (Seoul, Republic of Korea) between 2000 and 2009 were enrolled under the approval of the Institutional Review Boards of Hanyang University. Informed consent was provided by all patients enrolled in the study. Histopathological and clinical data were obtained from pathology reports and medical records. Tissue microarray construction was assessed as described previously (32). For survival analysis, breast cancer patients were stratified by the RBP2 level, defined by immunohistochemistry (Supplementary Methods, available online). Survival Analysis Using Public Data Using the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) (33) and Kaplan-Meier plotter (KM plotter, http://kmplot.com), the association between RBP2 expression level and disease-free survival (DFS) in human breast cancer was analyzed. Patients were stratified according to RBP2 mRNA expression (METABRIC, median; KM plotter, lower quartile). Further details are available in the Supplementary Methods (available online). Sulforhodamine B Colorimetric Assay The in vitro toxicology assay kit (Sulforhodamine B based, TOX 6) was purchased from Sigma-Aldrich (St Louis, MO). Cells (5 × 103 cells/ well) were cultured in phenol red-free DMEM containing 10% fetal bovine serum in a 96-well plate and supplemented with 4-hyrdoxytamoxifen for five days. The cells were fixed in trichloroacetic acid (TCA) for one hour at 4ºC and stained with 0.4% sulforhodamine B for 30 minutes. The stained cells were destained with 1% acetic acid and dissolved in 10 mM Tris for OD determination at 490 nm. Immunoblotting and Co-immunoprecipitation Assays Cells were lysed in radioimmunoprecipitation assay buffer supplemented with protease and phosphatase inhibitors. Immunoblotting and co-immunoprecipiation (IP) assays were then performed using appropriate antibodies as described previously (32). Further details are available in the Supplementary Methods (available online). Chromatin Immunoprecipitation Assay Chormatin immunopreceipitation (ChIP) was performed using a ChIP assay kit according to the manufacturer’s instructions (Upstate Biotechnology, Lake Placid, NY). Enrichment for the ChIP signal was detected by quantitative real-time polymerase chain reaction (qPCR) and analyzed based on input and normal IgG signals. Further details are available in the Supplementary Methods (available online). Phospho-RTK Array The Human Phospho-RTK array Kit was purchased from R&D systems (Minneapolis, MN). Diluted cell lysates were incubated with human phosphokinase array membranes, and bound phosphoproteins were analyzed according to the manufacturer’s instructions. Each membrane contained kinase-specific antibodies spotted in duplicate. Orthotopic Xenograft All mouse experiments were approved by the Hanyang University Animal Care and Use Committee (Seoul, Republic of Korea). Five-week-old female NOD/SCID mice were purchased from KOATECH (Pyeongtaek, Republic of Korea). MCF7 and BT474 cells (4 × 106 and 2 × 106 cells, respectively) mixed with Matrigel (BD bioscience, San Jose, CA) were injected into the mammary fat pads of mice implanted with E2 pellets (0.72 mg/pellet; 60-day release) before one week. After one week, the mice received a subcutaneous injection of tamoxifen pellet (5 mg/pellet; 60-day release), as described previously (34). Tumor growth was monitored twice per week, and the tumor volume was calculated as 1/2 × (long diameter) × (short diameter)2. Statistics The statistical significance of differences between two groups was analyzed with the unpaired Student’s t test using SPSS (version 12.0; SPSS, Inc., Chicago, IL). For multiple group comparisons and repeated measures, analysis of variance (ANOVA) and repeated-measures ANOVA (RM ANOVA), followed by post hoc least significant difference (LSD) test, were used. All P values were two-sided. P values of less than .05 were considered statistically significant. Results Effect of RBP2 on Tamoxifen Resistance in Breast Cancer To elucidate the role of RBP2 in ER+ breast cancer, we first examined the clinical relevance of RBP2 in human primary breast cancers. In our cohort, a high expression level of RBP2 protein was associated with worse DFS in all cases and ER+ breast cancer (n = 200 and 141, log-rank P = .05 in both cases) (Figure 1A;Supplementary Figure 1A, available online), as well as in patients who received tamoxifen therapy (n = 85, log-rank P = .04) (Figure 1A). Analysis of breast cancer patient survival according to RBP2 mRNA expression using METABRIC (33) and KM plotter (35) also showed similar results (ER+ cases, P = .05 in KM plotter and P = .001 in METABRIC; hormonal therapy+ cases, P = .02 in KM plotter and P = .007 in METABRIC) (Figure 1A;Supplementary Figure 1A, available online). Indeed, RBP2 expression was elevated in tamoxifen-resistant ER+ breast tumors compared with tamoxifen-sensitive breast tumors (P = .04, χ2 test; RBP2-positive rates = 91.7% and 61.6% in tumor-recurred and disease-free groups, respectively) (Figure 1B). The high RBP2 expression was also associated with higher stage and grade in ER+ breast cancer (P = .02 and .04, respectively, χ2 test) (Supplementary Tables 1 and 2, available online). RBP2 expression was not associated with resistance to other endocrine agents, such as aromatase inhibitors or fulvestrant (Supplementary Figure 1, B–E, available online). Figure 1. View largeDownload slide Association between retinoblastoma-binding protein 2 (RBP2) expression and tamoxifen resistance in estrogen receptor–positive (ER+) breast cancer. A) Analysis of the effect of the RBP2 expression level on disease-free survival (DFS) of ER+ breast cancer patients (upper panel) and ER+ breast cancer patients with hormonal therapy (lower panel) in the indicated cohorts. The Hanyang University (HYU) cohort, based on the RBP2 protein levels measured by immunohistochemistry; Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) and Kaplan-Meier (KM) plotter, based on the RBP2 mRNA levels. The data were assessed using the Kaplan-Meier method with the log-rank test. B) Representative images of immunohistochemical analysis of RBP2 expression levels in tamoxifen-sensitive (#1 and #3) and -resistant (#2 and #4) breast tumor tissues (left panel). Scale bars = 100 µm. A boxplot representing RBP2 expression levels in the indicated groups (upper right panel). The boxplot shows interquartile range (IQR), 95% confidence interval (CI), and outlier points. The P value was calculated using the two-sided Student’s t test. A table shows the association between RBP2 expression and tumor recurrence after tamoxifen therapy (lower right panel). Statistical significance was assessed using the χ2 test. All statistical tests were two-sided. CI = confidence interval; DFS = disease-free survival; ER = estrogen receptor; HYU = Hanyang University; IHC = immunohistochemistry; IQR = interquartile range; KM plotter = Kaplan-Meier plotter; METABRIC = Molecular Taxonomy of Breast Cancer International Consortium; RBP2 = retinoblastoma-binding protein 2. Figure 1. View largeDownload slide Association between retinoblastoma-binding protein 2 (RBP2) expression and tamoxifen resistance in estrogen receptor–positive (ER+) breast cancer. A) Analysis of the effect of the RBP2 expression level on disease-free survival (DFS) of ER+ breast cancer patients (upper panel) and ER+ breast cancer patients with hormonal therapy (lower panel) in the indicated cohorts. The Hanyang University (HYU) cohort, based on the RBP2 protein levels measured by immunohistochemistry; Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) and Kaplan-Meier (KM) plotter, based on the RBP2 mRNA levels. The data were assessed using the Kaplan-Meier method with the log-rank test. B) Representative images of immunohistochemical analysis of RBP2 expression levels in tamoxifen-sensitive (#1 and #3) and -resistant (#2 and #4) breast tumor tissues (left panel). Scale bars = 100 µm. A boxplot representing RBP2 expression levels in the indicated groups (upper right panel). The boxplot shows interquartile range (IQR), 95% confidence interval (CI), and outlier points. The P value was calculated using the two-sided Student’s t test. A table shows the association between RBP2 expression and tumor recurrence after tamoxifen therapy (lower right panel). Statistical significance was assessed using the χ2 test. All statistical tests were two-sided. CI = confidence interval; DFS = disease-free survival; ER = estrogen receptor; HYU = Hanyang University; IHC = immunohistochemistry; IQR = interquartile range; KM plotter = Kaplan-Meier plotter; METABRIC = Molecular Taxonomy of Breast Cancer International Consortium; RBP2 = retinoblastoma-binding protein 2. Consistent with the clinical evidence, RBP2 overexpression induced tamoxifen resistance in MCF7 cells with tamoxifen sensitivity and low RBP2 expression (P < .001, RM ANOVA) (Figure 2A;Supplementary Figure 2, available online). The catalytic inactive RBP2 H483A mutant could not completely rescue tamoxifen sensitivity in these cells. In tamoxifen-resistant MCF7 cells (MCF7-TamR), endogenous RBP2 expression was upregulated and the knockdown of its expression restored the tamoxifen sensitivity (P < .001, RM ANOVA). Similarly, shRNA-mediated RBP2 knockdown increased tamoxifen sensitivity in T47D and BT474 cells. Mice bearing orthotopic xenografts of RBP2-overexpressing MCF7 tumors (n = 8 per group) exhibited accelerated tumor growth regardless of tamoxifen treatment (mean [SD] tumor volume, tamoxifen-treated control = 70.8 [27.9] mm3, vs tamoxifen-treated RBP2 = 387.9 [85.1] mm3, P < .001) (Figure 2B), whereas the mice injected with RBP2-knockdown BT474 cells (n = 8 per group) showed a restored sensitivity to tamoxifen (tamoxifen-treated groups; shCON, mean [SD] tumor volume = 169.5 [25.3] mm3, vs shRBP2, mean [SD] tumor volume = 45.0 [7.6] mm3, P < .001) (Figure 2, B and C). Therefore, these data indicated that RBP2 mediates tamoxifen resistance in ER+ breast cancer. Figure 2. View largeDownload slide In vitro and in vivo effects of retinoblastoma-binding protein 2 (RBP2) on tamoxifen response in estrogen receptor-positive (ER+) breast cancer. A) Cell viability was measured by sulforhodamine B (SRB) colorimetric assay in the indicated cell lines (CON, empty vector; shCON, shRNA control; RBP2, RBP2 wild-type overexpression; RBP2H483A, catalytically inactive RBP2 mutant overexpression) with different doses of 4-hydroxytamoxifen (4-OH Tam) treatment for five days. Data are mean ± SD (n = 3). P values were calculated using repeated measures analysis of variance (RM ANOVA) with post hoc least significant difference (LSD) test. B) The effect of RBP2 on the in vivo tamoxifen response of breast cancer. Nonobese diabetic/severe combined immunodeficiency mice were orthotopically injected with RBP2-overexpressing MCF7 cells (upper panel) or knockdown BT474 cells (lower panel) following implantation with 17-β estradiol (E2) pellets and administration of tamoxifen (Tam) for four to five weeks. The growth curve of each group was analyzed by measuring the tumor size twice per week (n = 8 per group; mean (SD); RM ANOVA with post hoc LSD test). C) Immunohistochemical analysis of RBP2 and Ki67 expression levels in the xenograft tumors from the indicated groups of mice. Scale bars = 100 μm. Mean (SD) of three mice (P values by ANOVA with post hoc LSD test). All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; E2 = 17-β estradiol; ER = estrogen receptor; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; TamR = tamoxifen-resistant cell lines. Figure 2. View largeDownload slide In vitro and in vivo effects of retinoblastoma-binding protein 2 (RBP2) on tamoxifen response in estrogen receptor-positive (ER+) breast cancer. A) Cell viability was measured by sulforhodamine B (SRB) colorimetric assay in the indicated cell lines (CON, empty vector; shCON, shRNA control; RBP2, RBP2 wild-type overexpression; RBP2H483A, catalytically inactive RBP2 mutant overexpression) with different doses of 4-hydroxytamoxifen (4-OH Tam) treatment for five days. Data are mean ± SD (n = 3). P values were calculated using repeated measures analysis of variance (RM ANOVA) with post hoc least significant difference (LSD) test. B) The effect of RBP2 on the in vivo tamoxifen response of breast cancer. Nonobese diabetic/severe combined immunodeficiency mice were orthotopically injected with RBP2-overexpressing MCF7 cells (upper panel) or knockdown BT474 cells (lower panel) following implantation with 17-β estradiol (E2) pellets and administration of tamoxifen (Tam) for four to five weeks. The growth curve of each group was analyzed by measuring the tumor size twice per week (n = 8 per group; mean (SD); RM ANOVA with post hoc LSD test). C) Immunohistochemical analysis of RBP2 and Ki67 expression levels in the xenograft tumors from the indicated groups of mice. Scale bars = 100 μm. Mean (SD) of three mice (P values by ANOVA with post hoc LSD test). All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; E2 = 17-β estradiol; ER = estrogen receptor; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; TamR = tamoxifen-resistant cell lines. Functional Role of RBP2 in ER-Dependent Transcriptional Activation We next investigated the molecular mechanism underlying tamoxifen resistance by RBP2 based on RNA sequencing (RNA-seq). In RBP2-overexpressing MCF7 cells, approximately 60% of the RBP2 target genes overlapped with ER-bound and/or tamoxifen resistance–associated genes (Figure 3A;Supplementary Table 3, available online). Gene set enrichment analysis (GSEA) also showed that RBP2-target genes were linked to the estradiol response and tamoxifen resistance pathways (false discovery rate–adjusted P < .001) (Figure 3B;Supplementary Figure 3A and Supplementary Table 4, available online). These gene sets included several well-known ER-target genes, such as NRIP1, CCND1, IGFBP4, and IGFBP5, which are known to be associated with tamoxifen resistance or to have clinical significance in ER+ breast cancer (36–39). The altered expression levels of these genes were validated by quantitative reverse-transcription PCR (qRT-PCR) (Figure 3C). Figure 3. View largeDownload slide Regulation of estrogen receptor (ER)–dependent transcription in response to E2 and tamoxifen by retinoblastoma-binding protein 2 (RBP2). A) Heatmap displaying differentially expressed genes (≤1.5-fold) between control (CON) and RB2-overexpressing (RBP2) MCF7 cells based on RNA-seq results among common genes from the Venn diagram (left panel). The Venn diagram presents the number of common genes from RNA-seq results as compared with the ER ChIP-seq results obtained using MCF7 cells (ER ChIP-seq, GSE32222 (60), and the gene expression array in tamoxifen-resistant cell lines (TamR array, GSE14986 (61), respectively (right panel). B) Gene set enrichment analysis (GSEA) of differential gene expression profile between control and RBP2-overexpressing groups obtained from the RNA-seq analysis. False discovery rate–adjusted P value (q-value) was calculated to determine the statistical significance. C) The mRNA levels of the indicated genes were validated in RBP2-overexpressing MCF7 cells by quantitative real-time polymerase chain reaction (qRT-PCR) analysis. The data represent the mean (SD) of triplicate measurements. P values were calculated using the two-sided Student’s t test. D) The mRNA levels of NRIP1 and CCND1 in control (CON) or RBP2-overexpressing (RBP2) MCF7 cells following treatment with E2 and/or 4-OH Tam for 24 hours were analyzed by qRT-PCR. Mean (SD) (n = 3). P values by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. E) Lysates from the indicated cells in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours were subjected to co-immunoprecipitation using the indicated antibodies. Normal rabbit/mouse immunoglobulin G was used for negative control. F) Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the NRIP1 promoter regions in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours. The data represent the mean (SD) of triplicate measurements. Symbols indicate statistical significance. *P < .05 compared with CON; †††P < .001 compared with vehicle; ‡‡P < .01, ‡‡‡P < .001 compared with E2 (ANOVA with post hoc LSD test). G) Schematic illustration of the E2 agonistic effect of tamoxifen by RBP2. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ChIP = chromatin immunoprecipitation; E2 = 17-β estradiol; ER = estrogen receptor; ES = enrichment score; H3ac = histone H3 acetylation; IgG = immunoglobulin G; IP = immunoprecipitation; RBP2 = retinoblastoma-binding protein 2; seq = sequencing; TamR = tamoxifen-resistant cell lines. Figure 3. View largeDownload slide Regulation of estrogen receptor (ER)–dependent transcription in response to E2 and tamoxifen by retinoblastoma-binding protein 2 (RBP2). A) Heatmap displaying differentially expressed genes (≤1.5-fold) between control (CON) and RB2-overexpressing (RBP2) MCF7 cells based on RNA-seq results among common genes from the Venn diagram (left panel). The Venn diagram presents the number of common genes from RNA-seq results as compared with the ER ChIP-seq results obtained using MCF7 cells (ER ChIP-seq, GSE32222 (60), and the gene expression array in tamoxifen-resistant cell lines (TamR array, GSE14986 (61), respectively (right panel). B) Gene set enrichment analysis (GSEA) of differential gene expression profile between control and RBP2-overexpressing groups obtained from the RNA-seq analysis. False discovery rate–adjusted P value (q-value) was calculated to determine the statistical significance. C) The mRNA levels of the indicated genes were validated in RBP2-overexpressing MCF7 cells by quantitative real-time polymerase chain reaction (qRT-PCR) analysis. The data represent the mean (SD) of triplicate measurements. P values were calculated using the two-sided Student’s t test. D) The mRNA levels of NRIP1 and CCND1 in control (CON) or RBP2-overexpressing (RBP2) MCF7 cells following treatment with E2 and/or 4-OH Tam for 24 hours were analyzed by qRT-PCR. Mean (SD) (n = 3). P values by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. E) Lysates from the indicated cells in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours were subjected to co-immunoprecipitation using the indicated antibodies. Normal rabbit/mouse immunoglobulin G was used for negative control. F) Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the NRIP1 promoter regions in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours. The data represent the mean (SD) of triplicate measurements. Symbols indicate statistical significance. *P < .05 compared with CON; †††P < .001 compared with vehicle; ‡‡P < .01, ‡‡‡P < .001 compared with E2 (ANOVA with post hoc LSD test). G) Schematic illustration of the E2 agonistic effect of tamoxifen by RBP2. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ChIP = chromatin immunoprecipitation; E2 = 17-β estradiol; ER = estrogen receptor; ES = enrichment score; H3ac = histone H3 acetylation; IgG = immunoglobulin G; IP = immunoprecipitation; RBP2 = retinoblastoma-binding protein 2; seq = sequencing; TamR = tamoxifen-resistant cell lines. We further verified the RBP2 effect on ER-dependent transcription. In RBP2-overexpressing MCF7 cells, estrogen response element (ERE) luciferase activity was stimulated in response to 17β-estradiol (E2) and/or 4-hydroxytamoxifen treatment (P < .001 for vehicle vs E2, vehicle vs 4-hydroxytamoxifen) (Supplementary Figure 3B, available online). The expression levels of tamoxifen resistance–associated ER-target genes, including NRIP1 and CCND1, were subsequently upregulated under these conditions (Figure 3D;Supplementary Figure 3C, available online). Similarly, RBP2 knockdown recovered the inhibitory effect of tamoxifen on ER-dependent transcription in MCF7-TamR and BT474 cells (ERE-luc: P < .001 for E2 vs 4-hydroxytamoxifen, in both cells) (Figure 3D;Supplementary Figure 3D, available online). This E2 agonist action of tamoxifen by RBP2 was retained in RBP2 H483A-mutant-expressing cells (Supplementary Figure 3B, available online), implying that RBP enhanced ER-dependent transcription regardless of its demethylase activity. In MCF7 cells, RBP2 overexpression increased the RBP2 interaction with ER and histone acetyltransferase (HAT) p300 in response to E2 and/or 4-hydroxytamoxifen treatment (Figure 3E). Furthermore, RBP2-ER-p300 directly bound to the promoter regions of NRIP1 and CCND1, which contain RBP2-binding motifs ‘CCGCCC’ (40) and EREs, and induced histone H3 acetylation (H3ac) in this region (all P < .001) (Figure 3F;Supplementary Figure 3E, available online). These findings suggested that RBP2 switches tamoxifen function from an E2 antagonist to an E2 agonist by sustaining RBP2-ER-p300 complex in the regions of ER-dependent transcriptional activation (Figure 3G). RBP2-ER-NRIP1 Complex–Mediated Gene Silencing of IGFBP4/5 Because RBP2 was also involved in the transcriptional repression of ER target genes, including IGFBP4 and IGFBP5 (Figure 3, A and C; Supplementary Figure 3C and Supplementary Table 3, available online), we next investigated the molecular mechanism by which RBP2 suppresses ER-dependent transcription. Nuclear receptor interacting protein 1 (NRIP1) has been known as a major ER corepressor (41) and also was involved in histone deacetylase (HDAC)-mediated gene silencing (42,43), raising the possibility of cooperation between RBP2 and NRIP1 in the regulation of ER-dependent transcriptional repression. Indeed, the increase in NRIP1 expression in response to RBP2 facilitated the interaction between RBP2, ER, NRIP1, and HDAC1 (Figure 4A). Unlike NRIP1 and Cyclin D1, regulation of IGFBP4 and IGFBP5 by RBP2 was dependent on its demethylase activity (control vs RBP2, P < .001 for IGFBP4, P = .006 for IGFBP5; RBP2 vs RBP2H483A, P = .001 for both genes) (Figure 4B;Supplementary Figure 4, available online). Consistently, RBP2-ER-NRIP1-HDAC1 complex was recruited to the promoter regions of IGFBP4 and IGFBP5 containing both RBP2- and ER-binding sites, thus repressing H3K4me2/me3 and H3ac (all P < .001) (Figure 4C). The RBP2 H483A mutant recovered H3K4me2/3 and H3ac in their promoter regions (Figure 4D). Figure 4. View largeDownload slide Cooperation of retinoblastoma-binding protein 2 (RBP2) with estrogen receptor (ER) and nuclear receptor interacting protein 1 (NRIP1) in the epigenetic regulation of IGFBP4 and IGFBP5. A) Cell lysates from control or RBP2-overexpressing MCF7 cells were immunoprecipitated with anti-RBP2, NRIP1, and HDAC1 antibodies and subjected to immunoblotting using the indicated antibodies. B) Analysis of IGFBP4 and IGFBP5 expression in response to RBP2 expression. Indicated stable cell lines were maintained in DMEM containing 10% fetal bovine serum for 48 hours and subjected to immunoblotting (left panel) and quantitative real-time polymerase chain reaction (qRT-PCR; right panel). Mean (SD) (n = 3). P values were calculated by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test (for MCF7 cells) or two-sided Student’s t test (for BT474 cells). C) Schematic illustration of the promoter and transcription start site (TSS) of IGFBP4 and IGFBP5 gene locus containing the indicated binding motifs (left panel). Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis showing the fold enrichment of the indicated proteins and histone modifications at the IGFBP4 and IGFBP5 promoter regions in the indicated MCF7 stable cell lines (right panel). Data are presented as the mean (SD) (n = 3). P values by two-sided Student’s t test. D) ChIP–qPCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the IGFBP4 and IGFBP5 promoter region. Mean (SD) (n = 3). P values by ANOVA with post hoc LSD test. E) Effect of RBP2 on insulin-like growth factor-1 receptor (IGF1R) signaling. The total level and phosphorylated level of the indicated proteins were analyzed by immunoblotting. F) Schematic model of the regulation of IGFBP4 and IGFBP5 gene transcription by RBP2. All statistical tests were two-sided. bp = base pair; ChIP = chromatin immunoprecipitation; ERE = estrogen response element; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; NRIP1 = nuclear receptor interacting protein 1; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; TSS = transcription start site. Figure 4. View largeDownload slide Cooperation of retinoblastoma-binding protein 2 (RBP2) with estrogen receptor (ER) and nuclear receptor interacting protein 1 (NRIP1) in the epigenetic regulation of IGFBP4 and IGFBP5. A) Cell lysates from control or RBP2-overexpressing MCF7 cells were immunoprecipitated with anti-RBP2, NRIP1, and HDAC1 antibodies and subjected to immunoblotting using the indicated antibodies. B) Analysis of IGFBP4 and IGFBP5 expression in response to RBP2 expression. Indicated stable cell lines were maintained in DMEM containing 10% fetal bovine serum for 48 hours and subjected to immunoblotting (left panel) and quantitative real-time polymerase chain reaction (qRT-PCR; right panel). Mean (SD) (n = 3). P values were calculated by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test (for MCF7 cells) or two-sided Student’s t test (for BT474 cells). C) Schematic illustration of the promoter and transcription start site (TSS) of IGFBP4 and IGFBP5 gene locus containing the indicated binding motifs (left panel). Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis showing the fold enrichment of the indicated proteins and histone modifications at the IGFBP4 and IGFBP5 promoter regions in the indicated MCF7 stable cell lines (right panel). Data are presented as the mean (SD) (n = 3). P values by two-sided Student’s t test. D) ChIP–qPCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the IGFBP4 and IGFBP5 promoter region. Mean (SD) (n = 3). P values by ANOVA with post hoc LSD test. E) Effect of RBP2 on insulin-like growth factor-1 receptor (IGF1R) signaling. The total level and phosphorylated level of the indicated proteins were analyzed by immunoblotting. F) Schematic model of the regulation of IGFBP4 and IGFBP5 gene transcription by RBP2. All statistical tests were two-sided. bp = base pair; ChIP = chromatin immunoprecipitation; ERE = estrogen response element; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; NRIP1 = nuclear receptor interacting protein 1; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; TSS = transcription start site. Insulin-like growth factor binding proteins (IGFBPs) block insulin-like growth factor-1 receptor (IGF1R) activation by disrupting ligand binding to IGF1R (44). Indeed, the reduced IGFBP4 and IGFBP5 expression in response to RBP2 increased IGF1R phosphorylation and activated the downstream PI3K/AKT pathway (Figure 4E). Interestingly, the enhanced IGF1R activity was completely inhibited, whereas AKT was incompletely dephosphorylated by expression of the RBP2 H483A, suggesting that RBP2-mediated AKT activation partially depends on IGF1R signaling and RBP2 catalytic activity. Together, these findings indicated that RBP2 activates IGF1R signaling via the suppression of IGFBP4 and IGFBP5 transcription by collaborating with ER-NRIP1-HDAC1 in a histone demethylation–dependent manner (Figure 4F). RBP2 Effect on IGF1R/EGFR/HER2 Signaling in Mediating Tamoxifen Resistance To further define the mechanism responsible for the regulatory effect of RBP2 on the PI3K/AKT pathway, changes in the activities of multiple receptor tyrosine kinases (RTKs) following RBP2 overexpression were analyzed. Not only the IGF1R but also ErbB family receptors were remarkably activated by RBP2 overexpression in MCF7 cells (Figure 5, A and B; Supplementary Figure 5, available online). Similar results were shown in RBP2-knockdown cells (Figure 5B). Moreover, RBP2 increased the expression of EGFR and HER2 at the posttranscriptional level in a demethylase activity–independent manner (Figure 5B;Supplementary Figure 6A, available online). RBP2 wild-type, but not RBP2 H483A, induces prolonged protein stability of EGFR and HER2 (Figure 5C;Supplementary Figure 5B, available online). Likewise, RBP2 knockdown enhanced the ubiquitin/proteasome-dependent degradation of EGFR and HER2 (Figure 5, C–E; Supplementary Figure 6B, available online). These data suggested that RBP2 activates ErbB signaling through the stabilization of EGFR and HER2 proteins in a demethylase activity–independent manner. Figure 5. View largeDownload slide Effect of retinoblastoma-binding protein 2 (RBP2) on insulin-like growth factor-1 receptor (IGF1R) and epidermal growth factor receptor (EGFR)/human epidermal growth factor 2 (HER2) signaling pathway. A) Activation of phosphokinase-related proteins in RBP2-overexpressing cells were analyzed and represented by a bar graph using a human phospho-RTK array. B) Immunoblots showing activated and total expression levels of EGFR, HER2, and IGF1R in the indicated stable cell lines. C) Cell lysates from the indicated cells treated with 100 µg/mL cycloheximide for the indicate times were subjected to immunoblotting. D) Cell lysates from the indicated cells without or with MG132 (40 µM) for six hours were analyzed to determine the proteasomal degradation of EGFR and HER2. E) Ubiquitin levels of endogenous EGFR and HER2 proteins in control and RBP2-knockdown BT474 cells using the ubiquitin assay. F) Cell lysates from the indicated cells were immunoprecipitated with the indicated antibodies, and each protein was detected by co-immunoprecipitation. G) Phosphorylated AKT levels were measured by immunoblotting in RBP2-overexpressing MCF7 cells treated with 5 μM 4-OH Tam, 1 μM BKM120, and combinational treatment for three days. H) Colony formation assay (left panel) and sulforhodamine B (SRB) assay (right panel) in the indicated cells treated with 5 μM 4-OH Tam, 0.5 μM BKM120, or BKM120 with 4-OH Tam (4-OH Tam + BKM120) for 15 days and five days, respectively. Data are presented as the mean (SD) of triplicate measurements. P values were calculated using analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ANOVA = analysis of variance; CHX = cycloheximide; EGFR = epidermal growth factor receptor; HER2 = human epidermal growth factor 2; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; Phospho-RTK = phosphorylated receptor tyrosine kinase; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; ub = ubiquitin. Figure 5. View largeDownload slide Effect of retinoblastoma-binding protein 2 (RBP2) on insulin-like growth factor-1 receptor (IGF1R) and epidermal growth factor receptor (EGFR)/human epidermal growth factor 2 (HER2) signaling pathway. A) Activation of phosphokinase-related proteins in RBP2-overexpressing cells were analyzed and represented by a bar graph using a human phospho-RTK array. B) Immunoblots showing activated and total expression levels of EGFR, HER2, and IGF1R in the indicated stable cell lines. C) Cell lysates from the indicated cells treated with 100 µg/mL cycloheximide for the indicate times were subjected to immunoblotting. D) Cell lysates from the indicated cells without or with MG132 (40 µM) for six hours were analyzed to determine the proteasomal degradation of EGFR and HER2. E) Ubiquitin levels of endogenous EGFR and HER2 proteins in control and RBP2-knockdown BT474 cells using the ubiquitin assay. F) Cell lysates from the indicated cells were immunoprecipitated with the indicated antibodies, and each protein was detected by co-immunoprecipitation. G) Phosphorylated AKT levels were measured by immunoblotting in RBP2-overexpressing MCF7 cells treated with 5 μM 4-OH Tam, 1 μM BKM120, and combinational treatment for three days. H) Colony formation assay (left panel) and sulforhodamine B (SRB) assay (right panel) in the indicated cells treated with 5 μM 4-OH Tam, 0.5 μM BKM120, or BKM120 with 4-OH Tam (4-OH Tam + BKM120) for 15 days and five days, respectively. Data are presented as the mean (SD) of triplicate measurements. P values were calculated using analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ANOVA = analysis of variance; CHX = cycloheximide; EGFR = epidermal growth factor receptor; HER2 = human epidermal growth factor 2; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; Phospho-RTK = phosphorylated receptor tyrosine kinase; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; ub = ubiquitin. Previous studies have demonstrated that crosstalk between the IGF1R and the ErbB family activates downstream signaling pathways (45,46). In BT474 cells, which have elevated levels of the IGF1R and ErbB family, the interaction between IGF1R, EGFR, and HER2 was inhibited by RBP2 knockdown (Figure 5F), raising the possibility that RBP2 may activate PI3K/AKT signaling by facilitating the crosstalk between IGF1R and EGFR/HER2. Hyperactivation of the PI3K/AKT pathway has been considered a possible mechanism leading to endocrine therapy resistance in ER+ breast cancer (47). RBP2-overexpressing MCF7 cells were sensitized to the PI3K inhibitor BKM120, but not tamoxifen (% cell viability [SD], BKM120 = 67.1 [2.7]%, tamoxifen = 89.0 [3.8]%, P < .001) (Figure 5H). Moreover, tamoxifen resistance by RBP2 was recovered by combination treatment with tamoxifen and BKM120 (% cell viability [SD], tamoxifen + BKM120 = 41.3 [5.6]%, P < .001). Collectively, these findings implied that RBP2-induced tamoxifen resistance is due to activation of the PI3K/AKT pathway via the IGF1R and ErbB signaling. Discussion Here, we demonstrated the functional role of RBP2 in mediating ER activity and tamoxifen resistance in breast cancer. RBP2 regulated both ER-dependent transcriptional activation and repression in collaboration with ER-p300 and ER-NRIP1-HDAC1, respectively (Figure 6A). The decreased expression of IGFBP4/5 by RBP2-ER-NRIP1-HDAC1 resulted in IGF1R activation. RBP also promoted the crosstalk between IGF1R and ErbB receptors by enhancing the EGFR and HER2 protein stability, resulting in activation of the PI3K/AKT pathway and tamoxifen resistance (Figure 6B). This evidence suggests RBP2 as a crucial molecular target for tamoxifen-resistant ER+ breast cancer. Figure 6. View largeDownload slide A proposed model for the regulation of tamoxifen resistance by retinoblastoma-binding protein 2 (RBP2) in breast cancer. A) Schematic model of the regulation of estrogen receptor (ER) target genes by RBP2 with ER and ER coregulators in histone demethylation–independent (left panel) and –dependent (right panel) manners. B) Proposed model for the regulation of IGF1R-ErbB-PI3K-AKT signaling by RBP2 to induce tamoxifen resistance in ER+ breast cancer. ER = estrogen receptor; H3ac = histone H3 acetylation; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; P = phosphorylation; RBP2 = retinoblastoma-binding protein 2. Figure 6. View largeDownload slide A proposed model for the regulation of tamoxifen resistance by retinoblastoma-binding protein 2 (RBP2) in breast cancer. A) Schematic model of the regulation of estrogen receptor (ER) target genes by RBP2 with ER and ER coregulators in histone demethylation–independent (left panel) and –dependent (right panel) manners. B) Proposed model for the regulation of IGF1R-ErbB-PI3K-AKT signaling by RBP2 to induce tamoxifen resistance in ER+ breast cancer. ER = estrogen receptor; H3ac = histone H3 acetylation; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; P = phosphorylation; RBP2 = retinoblastoma-binding protein 2. Recent studies have suggested an oncogenic role of RBP2 in breast cancer. RBP2 knockout in MMTV-neu mice abrogated tumorigenesis and metastatic potential (14). In human breast tumor, amplified and overexpressed RBP2 was associated with increased cell proliferation and resistance to erlotinib (11). Combinational treatment with an RBP2 inhibitor and herceptin or gefitinib repressed colony formation in HER2+ breast cancer cells (18). To our knowledge, this study provides the first evidence of oncogenic RBP2 function in hormone-dependent breast cancer as a key regulator of ER and RTK signaling. Due to the innate and acquired resistance to endocrine therapy via the activation of ER, RTKs, and other bypass tracks, various studies have proposed combination treatment with endocrine therapy and targeted drugs for RTK signaling, such as ErbB and IGF1R inhibitors, and PI3K-AKT inhibitors (47–51). We also observed that combination treatment with tamoxifen and PI3K inhibitor BKM120 was effective for RBP2-overexpressing ER+ breast cancer cells with tamoxifen resistance. These findings suggest that ER+ breast cancers with high expression of RBP2 may be required for combined treatment with tamoxifen and other targeted drugs, such as a PI3K/AKT blocker, related to RBP2-dependent pathways. Collectively, RBP2 might serve as a new prognostic marker and therapeutic target in ER+ breast cancer. Notably, RBP2 activates not only E2-bound ER but also tamoxifen-bound ER to allow ER-dependent transcriptional regulation, leading to tamoxifen resistance. Abnormal activity of ER coregulators contributes to the development of breast cancer and the induction of endocrine resistance. For example, the upregulation of co-activators, including NCOA1/p300 complex and AIB1, is associated with the E2 agonist function of tamoxifen via ER activation and ErbB family overexpression (52–55). Consistently, high levels of corepressors, such as NRIP1, FOXA1, and DNMT1, induce tamoxifen resistance (38,56,57). Here, we verified that RBP2 induces consistent ER and p300 binding to the NRIP1 and CCND1 promoter regions, thus upregulating their expression in both E2- and tamoxifen treatment conditions, suggesting that RBP2 may block the antagonist function of tamoxifen-bound ER via sustaining the interaction with ER co-activators in the transcriptionally active ER-bound regions. In addition, NRIP1 upregulation appears to change the RBP2-containing transcriptional active complex to a transcriptional repressive complex. Although NRIP1 was initially identified as a co-activator of ER (58), a recent study demonstrated that NRIP1 expression is promptly induced in response to estrogen and subsequently mediates ER-dependent transcriptional repression (41). Based on our findings regarding RBP2/ER/NRIP1/HDAC1-mediated epigenetic repression of IGFBP4/5, increased NRIP1 by RBP2 may be crucial for a formation of RBP2/ER-containing transcriptional repressive complex. Collectively, RBP2 acts as both a co-activator and a corepressor of ER and alters tamoxifen-bound ER activity to regulate tamoxifen resistance–associated gene transcription. Accumulating studies have identified the demethylase activity–dependent oncogenic function of RBP2 (10,12,15). For example, RBP2 removes H3K4 methylation from p27 promoters, thus inducing tumorigenesis and metastasis (12). The reduction in global levels of H3K4 methylation that requires RBP2 is associated with gefitinib resistance in lung cancer (15). A recent study reported that the PI3K/AKT pathway inhibits RBP2-mediated H3K4 demethylation by phosphorylating RBP2 protein in breast cancer progression (59). However, another study demonstrated that RBP2 promotes breast cancer progression and metastasis via the histone demethylation–independent regulation of genes related to invasion, although the molecular mechanism remains unclear (14). We suggest that RBP2 have both demethylation-dependent and -independent functions in oncogenic processes. In the regulation of ER-dependent transcription, RBP2/ER/NRIP1/HDAC1 repressive complex was dependent on a demethylase activity, while RBP2 cooperated with ER and p300 in a histone demethylase activity–independent manner. Thus, the association between RBP2, ER, and ER-dependent coregulators might be very important to determine RBP2-mediated transcriptional activation or repression and its dependence on catalytic activity. Furthermore, RBP2 stabilized EGFR and HER2 proteins independently of its demethylase activity. Together, our findings suggest that RBP2 is closely linked to multiple transcriptional- and post-translational regulatory mechanisms to promote breast progression in a demethylation-dependent or demethylation-independent manner. A limitation of this study is that there is still no way to test the pharmacological effect of RBP2 inhibition on tamoxifen-resistant breast cancer. Currently available RBP2 inhibitors are not specific to RBP2 and only target the demethylase activity of RBP2, even though both histone demethylation–dependent and –independent function of RBP2 are crucial for tamoxifen resistance. The development of novel selective inhibitors for a broad spectrum of RBP2 function and further preclinical and clinical studies using the inhibitors will be needed for therapeutically targeting RBP2 in tamoxifen-resistant breast cancer. In conclusion, we demonstrated that RBP2 is crucial for tamoxifen resistance by altering ER-dependent transcription with ER coregulators and enhancing the IGF1R/EGFR/HER2 signaling pathway in various transcriptional and post-transcriptional regulatory mechanisms. Thus, RBP2 may be a prospective prognostic marker and therapeutic target in breast cancer, particularly in tamoxifen-resistant ER+ breast cancer. Funding This work was supported by National Research Foundation of Korea (NRF) grants funded by the Korean government (No. 2015R1A2A1A10052578). Notes The study funders had no role in the design of the study; the collection, analysis, or interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication. All authors declare no potential conflicts of interests. References 1 Defeo-Jones D, Huang PS, Jones REet al.  , Cloning of cDNAs for cellular proteins that bind to the retinoblastoma gene product. Nature . 1991; 352 6332: 251– 254. Google Scholar CrossRef Search ADS PubMed  2 Christensen J, Agger K, Cloos PAet al.  , RBP2 belongs to a family of demethylases, specific for tri-and dimethylated lysine 4 on histone 3. Cell.  2007; 128 6: 1063– 1076. Google Scholar CrossRef Search ADS PubMed  3 Klose RJ, Yan Q, Tothova Zet al.  , The retinoblastoma binding protein RBP2 is an H3K4 demethylase. Cell.  2007; 128 5: 889– 900. Google Scholar CrossRef Search ADS PubMed  4 Benevolenskaya EV, Murray HL, Branton Pet al.  , Binding of pRB to the PHD protein RBP2 promotes cellular differentiation. Mol Cell.  2005; 18 6: 623– 635. Google Scholar CrossRef Search ADS PubMed  5 Pasini D, Hansen KH, Christensen Jet al.  , Coordinated regulation of transcriptional repression by the RBP2 H3K4 demethylase and Polycomb-Repressive Complex 2. Genes Dev.  2008; 22 10: 1345– 1355. Google Scholar CrossRef Search ADS PubMed  6 Chaturvedi CP, Somasundaram B, Singh Ket al.  , Maintenance of gene silencing by the coordinate action of the H3K9 methyltransferase G9a/KMT1C and the H3K4 demethylase Jarid1a/KDM5A. Proc Natl Acad Sci U S A.  2012; 109 46: 18845– 18850. Google Scholar CrossRef Search ADS PubMed  7 Liefke R, Oswald F, Alvarado Cet al.  , Histone demethylase KDM5A is an integral part of the core Notch-RBP-J repressor complex. Genes Dev.  2010; 24 6: 590– 601. Google Scholar CrossRef Search ADS PubMed  8 Ge Z, Li W, Wang Net al.  , Chromatin remodeling: Recruitment of histone demethylase RBP2 by Mad1 for transcriptional repression of a Myc target gene, telomerase reverse transcriptase. FASEB J.  2010; 24 2: 579– 586. Google Scholar CrossRef Search ADS PubMed  9 Chan SW, Hong W. Retinoblastoma-binding protein 2 (Rbp2) potentiates nuclear hormone receptor-mediated transcription. J Biol Chem.  2001; 276 30: 28402– 28412. Google Scholar CrossRef Search ADS PubMed  10 Li L, Wang L, Song Pet al.  , Critical role of histone demethylase RBP2 in human gastric cancer angiogenesis. Mol Cancer.  2014; 13: 81. Google Scholar CrossRef Search ADS PubMed  11 Hou J, Wu J, Dombkowski Aet al.  , Genomic amplification and a role in drug-resistance for the KDM5A histone demethylase in breast cancer. Am J Transl Res.  2012; 4 3: 247– 256. Google Scholar PubMed  12 Teng YC, Lee CF, Li YSet al.  , Histone demethylase RBP2 promotes lung tumorigenesis and cancer metastasis. Cancer Res.  2013; 73 15: 4711– 4721. Google Scholar CrossRef Search ADS PubMed  13 Wang S, Wang Y, Wu Het al.  , RBP2 induces epithelial-mesenchymal transition in non-small cell lung cancer. PLoS One.  2013; 8 12: e84735. Google Scholar CrossRef Search ADS PubMed  14 Cao J, Liu Z, Cheung WKet al.  , Histone demethylase RBP2 is critical for breast cancer progression and metastasis. Cell Rep.  2014; 6 5: 868– 877. Google Scholar CrossRef Search ADS PubMed  15 Sharma SV, Lee DY, Li Bet al.  , A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell.  2010; 141 1: 69– 80. Google Scholar CrossRef Search ADS PubMed  16 Qi L, Zhu F, Li SHet al.  , Retinoblastoma binding protein 2 (RBP2) promotes HIF-1alpha-VEGF-induced angiogenesis of non-small cell lung cancer via the Akt pathway. PLoS One.  2014; 9 8: e106032. Google Scholar CrossRef Search ADS PubMed  17 Lin W, Cao J, Liu Jet al.  , Loss of the retinoblastoma binding protein 2 (RBP2) histone demethylase suppresses tumorigenesis in mice lacking Rb1 or Men1. Proc Natl Acad Sci U S A.  2011; 108 33: 13379– 13386. Google Scholar CrossRef Search ADS PubMed  18 Gale M, Sayegh J, Cao Jet al.  , Screen-identified selective inhibitor of lysine demethylase 5A blocks cancer cell growth and drug resistance. Oncotarget . 2016; in press. 19 Jensen EV, Jordan VC. The estrogen receptor: A model for molecular medicine. Clin Cancer Res.  2003; 9 6: 1980– 1989. Google Scholar PubMed  20 Ali S, Coombes RC. Endocrine-responsive breast cancer and strategies for combating resistance. Nat Rev Cancer.  2002; 2 2: 101– 112. Google Scholar CrossRef Search ADS PubMed  21 Green KA, Carroll JS. Oestrogen-receptor-mediated transcription and the influence of co-factors and chromatin state. Nat Rev Cancer.  2007; 7 9: 713– 722. Google Scholar CrossRef Search ADS PubMed  22 Jordan VC. Tamoxifen: A most unlikely pioneering medicine. Nat Rev Drug Discov.  2003; 2 3: 205– 213. Google Scholar CrossRef Search ADS PubMed  23 Shou J, Massarweh S, Osborne CKet al.  , Mechanisms of tamoxifen resistance: Increased estrogen receptor-HER2/neu cross-talk in ER/HER2-positive breast cancer. J Natl Cancer Inst.  2004; 96 12: 926– 935. Google Scholar CrossRef Search ADS PubMed  24 Jin K, Park S, Teo WWet al.  , HOXB7 is an ERalpha cofactor in the activation of HER2 and multiple ER Target genes leading to endocrine resistance. Cancer Discov.  2015; 5 9: 944– 959. Google Scholar CrossRef Search ADS PubMed  25 Campbell RA, Bhat-Nakshatri P, Patel NMet al.  , Phosphatidylinositol 3-kinase/AKT-mediated activation of estrogen receptor alpha: A new model for anti-estrogen resistance. J Biol Chem.  2001; 276 13: 9817– 9824. Google Scholar CrossRef Search ADS PubMed  26 Fan P, Agboke FA, Cunliffe HEet al.  , A molecular model for the mechanism of acquired tamoxifen resistance in breast cancer. Eur J Cancer.  2014; 50 16: 2866– 2876. Google Scholar CrossRef Search ADS PubMed  27 Osborne CK, Schiff R. Mechanisms of endocrine resistance in breast cancer. Annu Rev Med.  2011; 62: 233– 247. Google Scholar CrossRef Search ADS PubMed  28 Gottardis MM, Jordan VC. Development of tamoxifen-stimulated growth of MCF-7 tumors in athymic mice after long-term antiestrogen administration. Cancer Res.  1988; 48 18: 5183– 5187. Google Scholar PubMed  29 Gottardis MM, Wagner RJ, Borden ECet al.  , Differential ability of antiestrogens to stimulate breast cancer cell (MCF-7) growth in vivo and in vitro. Cancer Res.  1989; 49 17: 4765– 4769. Google Scholar PubMed  30 Yao K, Lee ES, Bentrem DJet al.  , Antitumor action of physiological estradiol on tamoxifen-stimulated breast tumors grown in athymic mice. Clin Cancer Res . 2000; 6 5: 2028– 2036. Google Scholar PubMed  31 Jordan VC. Selective estrogen receptor modulation: Concept and consequences in cancer. Cancer Cell.  2004; 5 3: 207– 213. Google Scholar CrossRef Search ADS PubMed  32 Cho MH, Park JH, Choi HJet al.  , DOT1L cooperates with the c-Myc-p300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression. Nat Commun.  2015; 6: 7821. Google Scholar CrossRef Search ADS PubMed  33 Curtis C, Shah SP, Chin SFet al.  , The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature.  2012; 486 7403: 346– 352. Google Scholar PubMed  34 Lee JY, Won HY, Park JHet al.  , MEL-18 loss mediates estrogen receptor-alpha downregulation and hormone independence. J Clin Invest.  2015; 125 5: 1801– 1814. Google Scholar CrossRef Search ADS PubMed  35 Gyorffy B, Lanczky A, Eklund ACet al.  , An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients. Breast Cancer Res Treat.  2010; 123 3: 725– 731. Google Scholar CrossRef Search ADS PubMed  36 Ahn BY, Elwi AN, Lee Bet al.  , Genetic screen identifies insulin-like growth factor binding protein 5 as a modulator of tamoxifen resistance in breast cancer. Cancer Res.  2010; 70 8: 3013– 3019. Google Scholar CrossRef Search ADS PubMed  37 Walker G, MacLeod K, Williams ARet al.  , Insulin-like growth factor binding proteins IGFBP3, IGFBP4, and IGFBP5 predict endocrine responsiveness in patients with ovarian cancer. Clin Cancer Res.  2007; 13 5: 1438– 1444. Google Scholar CrossRef Search ADS PubMed  38 Chan CM, Lykkesfeldt AE, Parker MGet al.  , Expression of nuclear receptor interacting proteins TIF-1, SUG-1, receptor interacting protein 140, and corepressor SMRT in tamoxifen-resistant breast cancer. Clin Cancer Res . 1999; 5 11: 3460– 3467. Google Scholar PubMed  39 Brown LA, Johnson K, Leung Set al.  , Co-amplification of CCND1 and EMSY is associated with an adverse outcome in ER-positive tamoxifen-treated breast cancers. Breast Cancer Res Treat.  2010; 121 2: 347– 354. Google Scholar CrossRef Search ADS PubMed  40 Tu S, Teng YC, Yuan Cet al.  , The ARID domain of the H3K4 demethylase RBP2 binds to a DNA CCGCCC motif. Nat Struct Mol Biol.  2008; 15 4: 419– 421. Google Scholar CrossRef Search ADS PubMed  41 Carroll JS, Meyer CA, Song Jet al.  , Genome-wide analysis of estrogen receptor binding sites. Nat Genet.  2006; 38 11: 1289– 1297. Google Scholar CrossRef Search ADS PubMed  42 Wei LN, Hu X, Chandra Det al.  , Receptor-interacting protein 140 directly recruits histone deacetylases for gene silencing. J Biol Chem.  2000; 275 52: 40782– 40787. Google Scholar CrossRef Search ADS PubMed  43 Castet A, Boulahtouf A, Versini Get al.  , Multiple domains of the Receptor-Interacting Protein 140 contribute to transcription inhibition. Nucleic Acids Res.  2004; 32 6: 1957– 1966. Google Scholar CrossRef Search ADS PubMed  44 Baxter RC. IGF binding proteins in cancer: Mechanistic and clinical insights. Nat Rev Cancer.  2014; 14 5: 329– 341. Google Scholar CrossRef Search ADS PubMed  45 Gallardo A, Lerma E, Escuin Det al.  , Increased signalling of EGFR and IGF1R, and deregulation of PTEN/PI3K/Akt pathway are related with trastuzumab resistance in HER2 breast carcinomas. Br J Cancer.  2012; 106 8: 1367– 1373. Google Scholar CrossRef Search ADS PubMed  46 Huang X, Gao L, Wang Set al.  , Heterotrimerization of the growth factor receptors erbB2, erbB3, and insulin-like growth factor-i receptor in breast cancer cells resistant to herceptin. Cancer Res.  2010; 70 3: 1204– 1214. Google Scholar CrossRef Search ADS PubMed  47 Miller TW, Balko JM, Arteaga CL. Phosphatidylinositol 3-kinase and antiestrogen resistance in breast cancer. J Clin Oncol.  2011; 29 33: 4452– 4461. Google Scholar CrossRef Search ADS PubMed  48 Hudis CA. Trastuzumab—mechanism of action and use in clinical practice. N Engl J Med.  2007; 357 1: 39– 51. Google Scholar CrossRef Search ADS PubMed  49 Johnston SR. Enhancing endocrine therapy for hormone receptor-positive advanced breast cancer: Cotargeting signaling pathways. J Natl Cancer Inst.  2015; 107 10:djv212. 50 Robertson JF, Ferrero JM, Bourgeois Het al.  , Ganitumab with either exemestane or fulvestrant for postmenopausal women with advanced, hormone-receptor-positive breast cancer: A randomised, controlled, double-blind, phase 2 trial. Lancet Oncol.  2013; 14 3: 228– 235. Google Scholar CrossRef Search ADS PubMed  51 Bachelot T, Bourgier C, Cropet Cet al.  , Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: A GINECO study. J Clin Oncol.  2012; 30 22: 2718– 2724. Google Scholar CrossRef Search ADS PubMed  52 Zwart W, Theodorou V, Kok Met al.  , Oestrogen receptor-co-factor-chromatin specificity in the transcriptional regulation of breast cancer. EMBO J.  2011; 30 23: 4764– 4776. Google Scholar CrossRef Search ADS PubMed  53 Osborne CK, Bardou V, Hopp TAet al.  , Role of the estrogen receptor coactivator AIB1 (SRC-3) and HER-2/neu in tamoxifen resistance in breast cancer. J Natl Cancer Inst.  2003; 95 5: 353– 361. Google Scholar CrossRef Search ADS PubMed  54 Hurtado A, Holmes KA, Geistlinger TRet al.  , Regulation of ERBB2 by oestrogen receptor-PAX2 determines response to tamoxifen. Nature.  2008; 456 7222: 663– 666. Google Scholar CrossRef Search ADS PubMed  55 Wu RC, Qin J, Yi Pet al.  , Selective phosphorylations of the SRC-3/AIB1 coactivator integrate genomic reponses to multiple cellular signaling pathways. Mol Cell.  2004; 15 6: 937– 949. Google Scholar CrossRef Search ADS PubMed  56 Fu X, Jeselsohn R, Pereira Ret al.  , FOXA1 overexpression mediates endocrine resistance by altering the ER transcriptome and IL-8 expression in ER-positive breast cancer. Proc Natl Acad Sci U S A.  2016; 113 43: E6600– E6609. Google Scholar CrossRef Search ADS PubMed  57 Phuong NT, Kim SK, Lim SCet al.  , Role of PTEN promoter methylation in tamoxifen-resistant breast cancer cells. Breast Cancer Res Treat.  2011; 130 1: 73– 83. Google Scholar CrossRef Search ADS PubMed  58 Cavailles V, Dauvois S, L'Horset Fet al.  , Nuclear factor RIP140 modulates transcriptional activation by the estrogen receptor. EMBO J.  1995; 14 15: 3741– 3751. Google Scholar PubMed  59 Spangle JM, Dreijerink KM, Groner ACet al.  , PI3K/AKT signaling regulates H3K4 methylation in breast cancer. Cell Rep.  2016; 15 12: 2692– 2704. Google Scholar CrossRef Search ADS PubMed  60 Ross-Innes CS, Stark R, Teschendorff AEet al.  , Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature.  2012; 481 7381: 389– 393. Google Scholar CrossRef Search ADS PubMed  61 Coser KR, Wittner BS, Rosenthal NFet al.  , Antiestrogen-resistant subclones of MCF-7 human breast cancer cells are derived from a common monoclonal drug-resistant progenitor. Proc Natl Acad Sci U S A.  2009; 106 34: 14536– 14541. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI: Journal of the National Cancer Institute Oxford University Press

Role of RBP2-Induced ER and IGF1R-ErbB Signaling in Tamoxifen Resistance in Breast Cancer

Loading next page...
 
/lp/ou_press/role-of-rbp2-induced-er-and-igf1r-erbb-signaling-in-tamoxifen-DWLT9HfevW
Publisher
Oxford University Press
Copyright
© The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com.
ISSN
0027-8874
eISSN
1460-2105
D.O.I.
10.1093/jnci/djx207
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Despite the benefit of endocrine therapy, acquired resistance during or after treatment still remains a major challenge in estrogen receptor (ER)–positive breast cancer. We investigated the potential role of histone demethylase retinoblastoma-binding protein 2 (RBP2) in endocrine therapy resistance of breast cancer. Methods Survival of breast cancer patients according to RBP2 expression was analyzed in three different breast cancer cohorts including METABRIC (n = 1980) and KM plotter (n = 1764). RBP2-mediated tamoxifen resistance was confirmed by in vitro sulforhodamine B (SRB) colorimetric, colony-forming assays, and in vivo xenograft models (n = 8 per group). RNA-seq analysis and receptor tyrosine kinase assay were performed to identify the tamoxifen resistance mechanism by RBP2. All statistical tests were two-sided. Results RBP2 was associated with poor prognosis to tamoxifen therapy in ER-positive breast cancer (P = .04 in HYU cohort, P = .02 in KM plotter, P = .007 in METABRIC, log-rank test). Furthermore, RBP2 expression was elevated in patients with tamoxifen-resistant breast cancer (P = .04, chi-square test). Knockdown of RBP2 conferred tamoxifen sensitivity, whereas overexpression of RBP2 induced tamoxifen resistance in vitro and in vivo (MCF7 xenograft: tamoxifen-treated control, mean [SD] tumor volume = 70.8 [27.9] mm3, vs tamoxifen-treated RBP2, mean [SD] tumor volume = 387.9 [85.1] mm3, P < .001). Mechanistically, RBP2 cooperated with ER co-activators and corepressors and regulated several tamoxifen resistance–associated genes, including NRIP1, CCND1, and IGFBP4 and IGFBP5. Furthermore, epigenetic silencing of IGFBP4/5 by RBP2-ER-NRIP1-HDAC1 complex led to insulin-like growth factor–1 receptor (IGF1R) activation. RBP2 also increased IGF1R-ErbB crosstalk and subsequent PI3K-AKT activation via demethylase activity–independent ErbB protein stabilization. Combinational treatment with tamoxifen and PI3K inhibitor could overcome RBP2-mediated tamoxifen resistance (RBP2-overexpressing cells: % cell viability [SD], tamoxifen = 89.0 [3.8]%, vs tamoxifen with BKM120 = 41.3 [5.6]%, P < .001). Conclusions RBP2 activates ER-IGF1R-ErbB signaling cascade in multiple ways to induce tamoxifen resistance, suggesting that RBP2 is a potential therapeutic target for ER-driven cancer. Retinoblastoma-binding protein 2 (RBP2, also known as KDM5A) is a demethylase toward histone H3 lysine 4 di- and tri-methylation (H3K4me2/3) (1–3). RBP2 acts as a gene silencer via H3K4 demethylation (4–8), but it is also involved in transcriptional activation (4,9,10). RBP2 plays an important role in cancer development. RBP2 is amplified/overexpressed in various human tumors including breast and lung cancers (10–13) and induces tumorigenesis, epithelial-to-mesenchymal transition, metastasis, angiogenesis, and drug tolerance (10–18). It also interacts with estrogen receptor (ER) (9); however, its functional and clinical relevance in ER-driven cancer remains unclear. ER is the most important biomarker and therapeutic target in human breast cancer (19). There are several distinct pathways leading to ER activation, such as estrogen-dependent/-independent, genomic, and nongenomic pathways (20,21). For targeting ER-positive (ER+) breast cancer, ER signaling is inhibited via various routes, such as blocking estrogen biosynthesis, suppressing ER expression, or competing with estrogen (20). Among endocrine agents, the anti-estrogen tamoxifen is the most widely used for ER+ breast cancer treatment (22). Despite the benefits, many patients undergo endocrine resistance via several mechanisms, including altered expression or activity of ER and the activation of receptor tyrosine kinases (RTKs) or other bypass signaling pathways (23–27). In particular, ER+ breast cancer with acquired resistance to tamoxifen can use both estrogen and tamoxifen as the growth stimulus (28–30), yet the mechanism for the agonist action of tamoxifen has not been fully established. Furthermore, long-term treatment with anti-estrogens develops new mechanisms of resistance that are distinct from resistance mechanisms that occur by short-term treatment (30,31), implying the complexity of endocrine therapy resistance. In this study, we explored the potential oncogenic role of RBP2 in ER-positive breast cancer to identify the novel molecular mechanism of tamoxifen resistance in breast cancer. Methods Patients and Surgical Specimens Patients with invasive ductal carcinoma (n = 200) who successfully underwent surgery at Hanyang University Hospital (Seoul, Republic of Korea) between 2000 and 2009 were enrolled under the approval of the Institutional Review Boards of Hanyang University. Informed consent was provided by all patients enrolled in the study. Histopathological and clinical data were obtained from pathology reports and medical records. Tissue microarray construction was assessed as described previously (32). For survival analysis, breast cancer patients were stratified by the RBP2 level, defined by immunohistochemistry (Supplementary Methods, available online). Survival Analysis Using Public Data Using the Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) (33) and Kaplan-Meier plotter (KM plotter, http://kmplot.com), the association between RBP2 expression level and disease-free survival (DFS) in human breast cancer was analyzed. Patients were stratified according to RBP2 mRNA expression (METABRIC, median; KM plotter, lower quartile). Further details are available in the Supplementary Methods (available online). Sulforhodamine B Colorimetric Assay The in vitro toxicology assay kit (Sulforhodamine B based, TOX 6) was purchased from Sigma-Aldrich (St Louis, MO). Cells (5 × 103 cells/ well) were cultured in phenol red-free DMEM containing 10% fetal bovine serum in a 96-well plate and supplemented with 4-hyrdoxytamoxifen for five days. The cells were fixed in trichloroacetic acid (TCA) for one hour at 4ºC and stained with 0.4% sulforhodamine B for 30 minutes. The stained cells were destained with 1% acetic acid and dissolved in 10 mM Tris for OD determination at 490 nm. Immunoblotting and Co-immunoprecipitation Assays Cells were lysed in radioimmunoprecipitation assay buffer supplemented with protease and phosphatase inhibitors. Immunoblotting and co-immunoprecipiation (IP) assays were then performed using appropriate antibodies as described previously (32). Further details are available in the Supplementary Methods (available online). Chromatin Immunoprecipitation Assay Chormatin immunopreceipitation (ChIP) was performed using a ChIP assay kit according to the manufacturer’s instructions (Upstate Biotechnology, Lake Placid, NY). Enrichment for the ChIP signal was detected by quantitative real-time polymerase chain reaction (qPCR) and analyzed based on input and normal IgG signals. Further details are available in the Supplementary Methods (available online). Phospho-RTK Array The Human Phospho-RTK array Kit was purchased from R&D systems (Minneapolis, MN). Diluted cell lysates were incubated with human phosphokinase array membranes, and bound phosphoproteins were analyzed according to the manufacturer’s instructions. Each membrane contained kinase-specific antibodies spotted in duplicate. Orthotopic Xenograft All mouse experiments were approved by the Hanyang University Animal Care and Use Committee (Seoul, Republic of Korea). Five-week-old female NOD/SCID mice were purchased from KOATECH (Pyeongtaek, Republic of Korea). MCF7 and BT474 cells (4 × 106 and 2 × 106 cells, respectively) mixed with Matrigel (BD bioscience, San Jose, CA) were injected into the mammary fat pads of mice implanted with E2 pellets (0.72 mg/pellet; 60-day release) before one week. After one week, the mice received a subcutaneous injection of tamoxifen pellet (5 mg/pellet; 60-day release), as described previously (34). Tumor growth was monitored twice per week, and the tumor volume was calculated as 1/2 × (long diameter) × (short diameter)2. Statistics The statistical significance of differences between two groups was analyzed with the unpaired Student’s t test using SPSS (version 12.0; SPSS, Inc., Chicago, IL). For multiple group comparisons and repeated measures, analysis of variance (ANOVA) and repeated-measures ANOVA (RM ANOVA), followed by post hoc least significant difference (LSD) test, were used. All P values were two-sided. P values of less than .05 were considered statistically significant. Results Effect of RBP2 on Tamoxifen Resistance in Breast Cancer To elucidate the role of RBP2 in ER+ breast cancer, we first examined the clinical relevance of RBP2 in human primary breast cancers. In our cohort, a high expression level of RBP2 protein was associated with worse DFS in all cases and ER+ breast cancer (n = 200 and 141, log-rank P = .05 in both cases) (Figure 1A;Supplementary Figure 1A, available online), as well as in patients who received tamoxifen therapy (n = 85, log-rank P = .04) (Figure 1A). Analysis of breast cancer patient survival according to RBP2 mRNA expression using METABRIC (33) and KM plotter (35) also showed similar results (ER+ cases, P = .05 in KM plotter and P = .001 in METABRIC; hormonal therapy+ cases, P = .02 in KM plotter and P = .007 in METABRIC) (Figure 1A;Supplementary Figure 1A, available online). Indeed, RBP2 expression was elevated in tamoxifen-resistant ER+ breast tumors compared with tamoxifen-sensitive breast tumors (P = .04, χ2 test; RBP2-positive rates = 91.7% and 61.6% in tumor-recurred and disease-free groups, respectively) (Figure 1B). The high RBP2 expression was also associated with higher stage and grade in ER+ breast cancer (P = .02 and .04, respectively, χ2 test) (Supplementary Tables 1 and 2, available online). RBP2 expression was not associated with resistance to other endocrine agents, such as aromatase inhibitors or fulvestrant (Supplementary Figure 1, B–E, available online). Figure 1. View largeDownload slide Association between retinoblastoma-binding protein 2 (RBP2) expression and tamoxifen resistance in estrogen receptor–positive (ER+) breast cancer. A) Analysis of the effect of the RBP2 expression level on disease-free survival (DFS) of ER+ breast cancer patients (upper panel) and ER+ breast cancer patients with hormonal therapy (lower panel) in the indicated cohorts. The Hanyang University (HYU) cohort, based on the RBP2 protein levels measured by immunohistochemistry; Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) and Kaplan-Meier (KM) plotter, based on the RBP2 mRNA levels. The data were assessed using the Kaplan-Meier method with the log-rank test. B) Representative images of immunohistochemical analysis of RBP2 expression levels in tamoxifen-sensitive (#1 and #3) and -resistant (#2 and #4) breast tumor tissues (left panel). Scale bars = 100 µm. A boxplot representing RBP2 expression levels in the indicated groups (upper right panel). The boxplot shows interquartile range (IQR), 95% confidence interval (CI), and outlier points. The P value was calculated using the two-sided Student’s t test. A table shows the association between RBP2 expression and tumor recurrence after tamoxifen therapy (lower right panel). Statistical significance was assessed using the χ2 test. All statistical tests were two-sided. CI = confidence interval; DFS = disease-free survival; ER = estrogen receptor; HYU = Hanyang University; IHC = immunohistochemistry; IQR = interquartile range; KM plotter = Kaplan-Meier plotter; METABRIC = Molecular Taxonomy of Breast Cancer International Consortium; RBP2 = retinoblastoma-binding protein 2. Figure 1. View largeDownload slide Association between retinoblastoma-binding protein 2 (RBP2) expression and tamoxifen resistance in estrogen receptor–positive (ER+) breast cancer. A) Analysis of the effect of the RBP2 expression level on disease-free survival (DFS) of ER+ breast cancer patients (upper panel) and ER+ breast cancer patients with hormonal therapy (lower panel) in the indicated cohorts. The Hanyang University (HYU) cohort, based on the RBP2 protein levels measured by immunohistochemistry; Molecular Taxonomy of Breast Cancer International Consortium (METABRIC) and Kaplan-Meier (KM) plotter, based on the RBP2 mRNA levels. The data were assessed using the Kaplan-Meier method with the log-rank test. B) Representative images of immunohistochemical analysis of RBP2 expression levels in tamoxifen-sensitive (#1 and #3) and -resistant (#2 and #4) breast tumor tissues (left panel). Scale bars = 100 µm. A boxplot representing RBP2 expression levels in the indicated groups (upper right panel). The boxplot shows interquartile range (IQR), 95% confidence interval (CI), and outlier points. The P value was calculated using the two-sided Student’s t test. A table shows the association between RBP2 expression and tumor recurrence after tamoxifen therapy (lower right panel). Statistical significance was assessed using the χ2 test. All statistical tests were two-sided. CI = confidence interval; DFS = disease-free survival; ER = estrogen receptor; HYU = Hanyang University; IHC = immunohistochemistry; IQR = interquartile range; KM plotter = Kaplan-Meier plotter; METABRIC = Molecular Taxonomy of Breast Cancer International Consortium; RBP2 = retinoblastoma-binding protein 2. Consistent with the clinical evidence, RBP2 overexpression induced tamoxifen resistance in MCF7 cells with tamoxifen sensitivity and low RBP2 expression (P < .001, RM ANOVA) (Figure 2A;Supplementary Figure 2, available online). The catalytic inactive RBP2 H483A mutant could not completely rescue tamoxifen sensitivity in these cells. In tamoxifen-resistant MCF7 cells (MCF7-TamR), endogenous RBP2 expression was upregulated and the knockdown of its expression restored the tamoxifen sensitivity (P < .001, RM ANOVA). Similarly, shRNA-mediated RBP2 knockdown increased tamoxifen sensitivity in T47D and BT474 cells. Mice bearing orthotopic xenografts of RBP2-overexpressing MCF7 tumors (n = 8 per group) exhibited accelerated tumor growth regardless of tamoxifen treatment (mean [SD] tumor volume, tamoxifen-treated control = 70.8 [27.9] mm3, vs tamoxifen-treated RBP2 = 387.9 [85.1] mm3, P < .001) (Figure 2B), whereas the mice injected with RBP2-knockdown BT474 cells (n = 8 per group) showed a restored sensitivity to tamoxifen (tamoxifen-treated groups; shCON, mean [SD] tumor volume = 169.5 [25.3] mm3, vs shRBP2, mean [SD] tumor volume = 45.0 [7.6] mm3, P < .001) (Figure 2, B and C). Therefore, these data indicated that RBP2 mediates tamoxifen resistance in ER+ breast cancer. Figure 2. View largeDownload slide In vitro and in vivo effects of retinoblastoma-binding protein 2 (RBP2) on tamoxifen response in estrogen receptor-positive (ER+) breast cancer. A) Cell viability was measured by sulforhodamine B (SRB) colorimetric assay in the indicated cell lines (CON, empty vector; shCON, shRNA control; RBP2, RBP2 wild-type overexpression; RBP2H483A, catalytically inactive RBP2 mutant overexpression) with different doses of 4-hydroxytamoxifen (4-OH Tam) treatment for five days. Data are mean ± SD (n = 3). P values were calculated using repeated measures analysis of variance (RM ANOVA) with post hoc least significant difference (LSD) test. B) The effect of RBP2 on the in vivo tamoxifen response of breast cancer. Nonobese diabetic/severe combined immunodeficiency mice were orthotopically injected with RBP2-overexpressing MCF7 cells (upper panel) or knockdown BT474 cells (lower panel) following implantation with 17-β estradiol (E2) pellets and administration of tamoxifen (Tam) for four to five weeks. The growth curve of each group was analyzed by measuring the tumor size twice per week (n = 8 per group; mean (SD); RM ANOVA with post hoc LSD test). C) Immunohistochemical analysis of RBP2 and Ki67 expression levels in the xenograft tumors from the indicated groups of mice. Scale bars = 100 μm. Mean (SD) of three mice (P values by ANOVA with post hoc LSD test). All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; E2 = 17-β estradiol; ER = estrogen receptor; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; TamR = tamoxifen-resistant cell lines. Figure 2. View largeDownload slide In vitro and in vivo effects of retinoblastoma-binding protein 2 (RBP2) on tamoxifen response in estrogen receptor-positive (ER+) breast cancer. A) Cell viability was measured by sulforhodamine B (SRB) colorimetric assay in the indicated cell lines (CON, empty vector; shCON, shRNA control; RBP2, RBP2 wild-type overexpression; RBP2H483A, catalytically inactive RBP2 mutant overexpression) with different doses of 4-hydroxytamoxifen (4-OH Tam) treatment for five days. Data are mean ± SD (n = 3). P values were calculated using repeated measures analysis of variance (RM ANOVA) with post hoc least significant difference (LSD) test. B) The effect of RBP2 on the in vivo tamoxifen response of breast cancer. Nonobese diabetic/severe combined immunodeficiency mice were orthotopically injected with RBP2-overexpressing MCF7 cells (upper panel) or knockdown BT474 cells (lower panel) following implantation with 17-β estradiol (E2) pellets and administration of tamoxifen (Tam) for four to five weeks. The growth curve of each group was analyzed by measuring the tumor size twice per week (n = 8 per group; mean (SD); RM ANOVA with post hoc LSD test). C) Immunohistochemical analysis of RBP2 and Ki67 expression levels in the xenograft tumors from the indicated groups of mice. Scale bars = 100 μm. Mean (SD) of three mice (P values by ANOVA with post hoc LSD test). All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; E2 = 17-β estradiol; ER = estrogen receptor; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; TamR = tamoxifen-resistant cell lines. Functional Role of RBP2 in ER-Dependent Transcriptional Activation We next investigated the molecular mechanism underlying tamoxifen resistance by RBP2 based on RNA sequencing (RNA-seq). In RBP2-overexpressing MCF7 cells, approximately 60% of the RBP2 target genes overlapped with ER-bound and/or tamoxifen resistance–associated genes (Figure 3A;Supplementary Table 3, available online). Gene set enrichment analysis (GSEA) also showed that RBP2-target genes were linked to the estradiol response and tamoxifen resistance pathways (false discovery rate–adjusted P < .001) (Figure 3B;Supplementary Figure 3A and Supplementary Table 4, available online). These gene sets included several well-known ER-target genes, such as NRIP1, CCND1, IGFBP4, and IGFBP5, which are known to be associated with tamoxifen resistance or to have clinical significance in ER+ breast cancer (36–39). The altered expression levels of these genes were validated by quantitative reverse-transcription PCR (qRT-PCR) (Figure 3C). Figure 3. View largeDownload slide Regulation of estrogen receptor (ER)–dependent transcription in response to E2 and tamoxifen by retinoblastoma-binding protein 2 (RBP2). A) Heatmap displaying differentially expressed genes (≤1.5-fold) between control (CON) and RB2-overexpressing (RBP2) MCF7 cells based on RNA-seq results among common genes from the Venn diagram (left panel). The Venn diagram presents the number of common genes from RNA-seq results as compared with the ER ChIP-seq results obtained using MCF7 cells (ER ChIP-seq, GSE32222 (60), and the gene expression array in tamoxifen-resistant cell lines (TamR array, GSE14986 (61), respectively (right panel). B) Gene set enrichment analysis (GSEA) of differential gene expression profile between control and RBP2-overexpressing groups obtained from the RNA-seq analysis. False discovery rate–adjusted P value (q-value) was calculated to determine the statistical significance. C) The mRNA levels of the indicated genes were validated in RBP2-overexpressing MCF7 cells by quantitative real-time polymerase chain reaction (qRT-PCR) analysis. The data represent the mean (SD) of triplicate measurements. P values were calculated using the two-sided Student’s t test. D) The mRNA levels of NRIP1 and CCND1 in control (CON) or RBP2-overexpressing (RBP2) MCF7 cells following treatment with E2 and/or 4-OH Tam for 24 hours were analyzed by qRT-PCR. Mean (SD) (n = 3). P values by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. E) Lysates from the indicated cells in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours were subjected to co-immunoprecipitation using the indicated antibodies. Normal rabbit/mouse immunoglobulin G was used for negative control. F) Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the NRIP1 promoter regions in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours. The data represent the mean (SD) of triplicate measurements. Symbols indicate statistical significance. *P < .05 compared with CON; †††P < .001 compared with vehicle; ‡‡P < .01, ‡‡‡P < .001 compared with E2 (ANOVA with post hoc LSD test). G) Schematic illustration of the E2 agonistic effect of tamoxifen by RBP2. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ChIP = chromatin immunoprecipitation; E2 = 17-β estradiol; ER = estrogen receptor; ES = enrichment score; H3ac = histone H3 acetylation; IgG = immunoglobulin G; IP = immunoprecipitation; RBP2 = retinoblastoma-binding protein 2; seq = sequencing; TamR = tamoxifen-resistant cell lines. Figure 3. View largeDownload slide Regulation of estrogen receptor (ER)–dependent transcription in response to E2 and tamoxifen by retinoblastoma-binding protein 2 (RBP2). A) Heatmap displaying differentially expressed genes (≤1.5-fold) between control (CON) and RB2-overexpressing (RBP2) MCF7 cells based on RNA-seq results among common genes from the Venn diagram (left panel). The Venn diagram presents the number of common genes from RNA-seq results as compared with the ER ChIP-seq results obtained using MCF7 cells (ER ChIP-seq, GSE32222 (60), and the gene expression array in tamoxifen-resistant cell lines (TamR array, GSE14986 (61), respectively (right panel). B) Gene set enrichment analysis (GSEA) of differential gene expression profile between control and RBP2-overexpressing groups obtained from the RNA-seq analysis. False discovery rate–adjusted P value (q-value) was calculated to determine the statistical significance. C) The mRNA levels of the indicated genes were validated in RBP2-overexpressing MCF7 cells by quantitative real-time polymerase chain reaction (qRT-PCR) analysis. The data represent the mean (SD) of triplicate measurements. P values were calculated using the two-sided Student’s t test. D) The mRNA levels of NRIP1 and CCND1 in control (CON) or RBP2-overexpressing (RBP2) MCF7 cells following treatment with E2 and/or 4-OH Tam for 24 hours were analyzed by qRT-PCR. Mean (SD) (n = 3). P values by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. E) Lysates from the indicated cells in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours were subjected to co-immunoprecipitation using the indicated antibodies. Normal rabbit/mouse immunoglobulin G was used for negative control. F) Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the NRIP1 promoter regions in the absence and presence of 10 nM E2 and/or 5 μM 4-OH Tam for 24 hours. The data represent the mean (SD) of triplicate measurements. Symbols indicate statistical significance. *P < .05 compared with CON; †††P < .001 compared with vehicle; ‡‡P < .01, ‡‡‡P < .001 compared with E2 (ANOVA with post hoc LSD test). G) Schematic illustration of the E2 agonistic effect of tamoxifen by RBP2. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ChIP = chromatin immunoprecipitation; E2 = 17-β estradiol; ER = estrogen receptor; ES = enrichment score; H3ac = histone H3 acetylation; IgG = immunoglobulin G; IP = immunoprecipitation; RBP2 = retinoblastoma-binding protein 2; seq = sequencing; TamR = tamoxifen-resistant cell lines. We further verified the RBP2 effect on ER-dependent transcription. In RBP2-overexpressing MCF7 cells, estrogen response element (ERE) luciferase activity was stimulated in response to 17β-estradiol (E2) and/or 4-hydroxytamoxifen treatment (P < .001 for vehicle vs E2, vehicle vs 4-hydroxytamoxifen) (Supplementary Figure 3B, available online). The expression levels of tamoxifen resistance–associated ER-target genes, including NRIP1 and CCND1, were subsequently upregulated under these conditions (Figure 3D;Supplementary Figure 3C, available online). Similarly, RBP2 knockdown recovered the inhibitory effect of tamoxifen on ER-dependent transcription in MCF7-TamR and BT474 cells (ERE-luc: P < .001 for E2 vs 4-hydroxytamoxifen, in both cells) (Figure 3D;Supplementary Figure 3D, available online). This E2 agonist action of tamoxifen by RBP2 was retained in RBP2 H483A-mutant-expressing cells (Supplementary Figure 3B, available online), implying that RBP enhanced ER-dependent transcription regardless of its demethylase activity. In MCF7 cells, RBP2 overexpression increased the RBP2 interaction with ER and histone acetyltransferase (HAT) p300 in response to E2 and/or 4-hydroxytamoxifen treatment (Figure 3E). Furthermore, RBP2-ER-p300 directly bound to the promoter regions of NRIP1 and CCND1, which contain RBP2-binding motifs ‘CCGCCC’ (40) and EREs, and induced histone H3 acetylation (H3ac) in this region (all P < .001) (Figure 3F;Supplementary Figure 3E, available online). These findings suggested that RBP2 switches tamoxifen function from an E2 antagonist to an E2 agonist by sustaining RBP2-ER-p300 complex in the regions of ER-dependent transcriptional activation (Figure 3G). RBP2-ER-NRIP1 Complex–Mediated Gene Silencing of IGFBP4/5 Because RBP2 was also involved in the transcriptional repression of ER target genes, including IGFBP4 and IGFBP5 (Figure 3, A and C; Supplementary Figure 3C and Supplementary Table 3, available online), we next investigated the molecular mechanism by which RBP2 suppresses ER-dependent transcription. Nuclear receptor interacting protein 1 (NRIP1) has been known as a major ER corepressor (41) and also was involved in histone deacetylase (HDAC)-mediated gene silencing (42,43), raising the possibility of cooperation between RBP2 and NRIP1 in the regulation of ER-dependent transcriptional repression. Indeed, the increase in NRIP1 expression in response to RBP2 facilitated the interaction between RBP2, ER, NRIP1, and HDAC1 (Figure 4A). Unlike NRIP1 and Cyclin D1, regulation of IGFBP4 and IGFBP5 by RBP2 was dependent on its demethylase activity (control vs RBP2, P < .001 for IGFBP4, P = .006 for IGFBP5; RBP2 vs RBP2H483A, P = .001 for both genes) (Figure 4B;Supplementary Figure 4, available online). Consistently, RBP2-ER-NRIP1-HDAC1 complex was recruited to the promoter regions of IGFBP4 and IGFBP5 containing both RBP2- and ER-binding sites, thus repressing H3K4me2/me3 and H3ac (all P < .001) (Figure 4C). The RBP2 H483A mutant recovered H3K4me2/3 and H3ac in their promoter regions (Figure 4D). Figure 4. View largeDownload slide Cooperation of retinoblastoma-binding protein 2 (RBP2) with estrogen receptor (ER) and nuclear receptor interacting protein 1 (NRIP1) in the epigenetic regulation of IGFBP4 and IGFBP5. A) Cell lysates from control or RBP2-overexpressing MCF7 cells were immunoprecipitated with anti-RBP2, NRIP1, and HDAC1 antibodies and subjected to immunoblotting using the indicated antibodies. B) Analysis of IGFBP4 and IGFBP5 expression in response to RBP2 expression. Indicated stable cell lines were maintained in DMEM containing 10% fetal bovine serum for 48 hours and subjected to immunoblotting (left panel) and quantitative real-time polymerase chain reaction (qRT-PCR; right panel). Mean (SD) (n = 3). P values were calculated by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test (for MCF7 cells) or two-sided Student’s t test (for BT474 cells). C) Schematic illustration of the promoter and transcription start site (TSS) of IGFBP4 and IGFBP5 gene locus containing the indicated binding motifs (left panel). Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis showing the fold enrichment of the indicated proteins and histone modifications at the IGFBP4 and IGFBP5 promoter regions in the indicated MCF7 stable cell lines (right panel). Data are presented as the mean (SD) (n = 3). P values by two-sided Student’s t test. D) ChIP–qPCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the IGFBP4 and IGFBP5 promoter region. Mean (SD) (n = 3). P values by ANOVA with post hoc LSD test. E) Effect of RBP2 on insulin-like growth factor-1 receptor (IGF1R) signaling. The total level and phosphorylated level of the indicated proteins were analyzed by immunoblotting. F) Schematic model of the regulation of IGFBP4 and IGFBP5 gene transcription by RBP2. All statistical tests were two-sided. bp = base pair; ChIP = chromatin immunoprecipitation; ERE = estrogen response element; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; NRIP1 = nuclear receptor interacting protein 1; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; TSS = transcription start site. Figure 4. View largeDownload slide Cooperation of retinoblastoma-binding protein 2 (RBP2) with estrogen receptor (ER) and nuclear receptor interacting protein 1 (NRIP1) in the epigenetic regulation of IGFBP4 and IGFBP5. A) Cell lysates from control or RBP2-overexpressing MCF7 cells were immunoprecipitated with anti-RBP2, NRIP1, and HDAC1 antibodies and subjected to immunoblotting using the indicated antibodies. B) Analysis of IGFBP4 and IGFBP5 expression in response to RBP2 expression. Indicated stable cell lines were maintained in DMEM containing 10% fetal bovine serum for 48 hours and subjected to immunoblotting (left panel) and quantitative real-time polymerase chain reaction (qRT-PCR; right panel). Mean (SD) (n = 3). P values were calculated by analysis of variance (ANOVA) with post hoc least significant difference (LSD) test (for MCF7 cells) or two-sided Student’s t test (for BT474 cells). C) Schematic illustration of the promoter and transcription start site (TSS) of IGFBP4 and IGFBP5 gene locus containing the indicated binding motifs (left panel). Chromatin immunoprecipitation (ChIP)–quantitative reverse-transcription PCR analysis showing the fold enrichment of the indicated proteins and histone modifications at the IGFBP4 and IGFBP5 promoter regions in the indicated MCF7 stable cell lines (right panel). Data are presented as the mean (SD) (n = 3). P values by two-sided Student’s t test. D) ChIP–qPCR analysis presenting the fold enrichment of the indicated proteins or histone marks at the IGFBP4 and IGFBP5 promoter region. Mean (SD) (n = 3). P values by ANOVA with post hoc LSD test. E) Effect of RBP2 on insulin-like growth factor-1 receptor (IGF1R) signaling. The total level and phosphorylated level of the indicated proteins were analyzed by immunoblotting. F) Schematic model of the regulation of IGFBP4 and IGFBP5 gene transcription by RBP2. All statistical tests were two-sided. bp = base pair; ChIP = chromatin immunoprecipitation; ERE = estrogen response element; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; NRIP1 = nuclear receptor interacting protein 1; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; TSS = transcription start site. Insulin-like growth factor binding proteins (IGFBPs) block insulin-like growth factor-1 receptor (IGF1R) activation by disrupting ligand binding to IGF1R (44). Indeed, the reduced IGFBP4 and IGFBP5 expression in response to RBP2 increased IGF1R phosphorylation and activated the downstream PI3K/AKT pathway (Figure 4E). Interestingly, the enhanced IGF1R activity was completely inhibited, whereas AKT was incompletely dephosphorylated by expression of the RBP2 H483A, suggesting that RBP2-mediated AKT activation partially depends on IGF1R signaling and RBP2 catalytic activity. Together, these findings indicated that RBP2 activates IGF1R signaling via the suppression of IGFBP4 and IGFBP5 transcription by collaborating with ER-NRIP1-HDAC1 in a histone demethylation–dependent manner (Figure 4F). RBP2 Effect on IGF1R/EGFR/HER2 Signaling in Mediating Tamoxifen Resistance To further define the mechanism responsible for the regulatory effect of RBP2 on the PI3K/AKT pathway, changes in the activities of multiple receptor tyrosine kinases (RTKs) following RBP2 overexpression were analyzed. Not only the IGF1R but also ErbB family receptors were remarkably activated by RBP2 overexpression in MCF7 cells (Figure 5, A and B; Supplementary Figure 5, available online). Similar results were shown in RBP2-knockdown cells (Figure 5B). Moreover, RBP2 increased the expression of EGFR and HER2 at the posttranscriptional level in a demethylase activity–independent manner (Figure 5B;Supplementary Figure 6A, available online). RBP2 wild-type, but not RBP2 H483A, induces prolonged protein stability of EGFR and HER2 (Figure 5C;Supplementary Figure 5B, available online). Likewise, RBP2 knockdown enhanced the ubiquitin/proteasome-dependent degradation of EGFR and HER2 (Figure 5, C–E; Supplementary Figure 6B, available online). These data suggested that RBP2 activates ErbB signaling through the stabilization of EGFR and HER2 proteins in a demethylase activity–independent manner. Figure 5. View largeDownload slide Effect of retinoblastoma-binding protein 2 (RBP2) on insulin-like growth factor-1 receptor (IGF1R) and epidermal growth factor receptor (EGFR)/human epidermal growth factor 2 (HER2) signaling pathway. A) Activation of phosphokinase-related proteins in RBP2-overexpressing cells were analyzed and represented by a bar graph using a human phospho-RTK array. B) Immunoblots showing activated and total expression levels of EGFR, HER2, and IGF1R in the indicated stable cell lines. C) Cell lysates from the indicated cells treated with 100 µg/mL cycloheximide for the indicate times were subjected to immunoblotting. D) Cell lysates from the indicated cells without or with MG132 (40 µM) for six hours were analyzed to determine the proteasomal degradation of EGFR and HER2. E) Ubiquitin levels of endogenous EGFR and HER2 proteins in control and RBP2-knockdown BT474 cells using the ubiquitin assay. F) Cell lysates from the indicated cells were immunoprecipitated with the indicated antibodies, and each protein was detected by co-immunoprecipitation. G) Phosphorylated AKT levels were measured by immunoblotting in RBP2-overexpressing MCF7 cells treated with 5 μM 4-OH Tam, 1 μM BKM120, and combinational treatment for three days. H) Colony formation assay (left panel) and sulforhodamine B (SRB) assay (right panel) in the indicated cells treated with 5 μM 4-OH Tam, 0.5 μM BKM120, or BKM120 with 4-OH Tam (4-OH Tam + BKM120) for 15 days and five days, respectively. Data are presented as the mean (SD) of triplicate measurements. P values were calculated using analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ANOVA = analysis of variance; CHX = cycloheximide; EGFR = epidermal growth factor receptor; HER2 = human epidermal growth factor 2; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; Phospho-RTK = phosphorylated receptor tyrosine kinase; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; ub = ubiquitin. Figure 5. View largeDownload slide Effect of retinoblastoma-binding protein 2 (RBP2) on insulin-like growth factor-1 receptor (IGF1R) and epidermal growth factor receptor (EGFR)/human epidermal growth factor 2 (HER2) signaling pathway. A) Activation of phosphokinase-related proteins in RBP2-overexpressing cells were analyzed and represented by a bar graph using a human phospho-RTK array. B) Immunoblots showing activated and total expression levels of EGFR, HER2, and IGF1R in the indicated stable cell lines. C) Cell lysates from the indicated cells treated with 100 µg/mL cycloheximide for the indicate times were subjected to immunoblotting. D) Cell lysates from the indicated cells without or with MG132 (40 µM) for six hours were analyzed to determine the proteasomal degradation of EGFR and HER2. E) Ubiquitin levels of endogenous EGFR and HER2 proteins in control and RBP2-knockdown BT474 cells using the ubiquitin assay. F) Cell lysates from the indicated cells were immunoprecipitated with the indicated antibodies, and each protein was detected by co-immunoprecipitation. G) Phosphorylated AKT levels were measured by immunoblotting in RBP2-overexpressing MCF7 cells treated with 5 μM 4-OH Tam, 1 μM BKM120, and combinational treatment for three days. H) Colony formation assay (left panel) and sulforhodamine B (SRB) assay (right panel) in the indicated cells treated with 5 μM 4-OH Tam, 0.5 μM BKM120, or BKM120 with 4-OH Tam (4-OH Tam + BKM120) for 15 days and five days, respectively. Data are presented as the mean (SD) of triplicate measurements. P values were calculated using analysis of variance (ANOVA) with post hoc least significant difference (LSD) test. All statistical tests were two-sided. 4-OH Tam = 4-hydroxytamoxifen; ANOVA = analysis of variance; CHX = cycloheximide; EGFR = epidermal growth factor receptor; HER2 = human epidermal growth factor 2; IGF1R = insulin-like growth factor-1 receptor; IgG = immunoglobulin G; IP = immunoprecipitation; Phospho-RTK = phosphorylated receptor tyrosine kinase; RBP2 = retinoblastoma-binding protein 2; sh = short hairpin RNA; SRB = sulforhodamine B; Tam = tamoxifen; ub = ubiquitin. Previous studies have demonstrated that crosstalk between the IGF1R and the ErbB family activates downstream signaling pathways (45,46). In BT474 cells, which have elevated levels of the IGF1R and ErbB family, the interaction between IGF1R, EGFR, and HER2 was inhibited by RBP2 knockdown (Figure 5F), raising the possibility that RBP2 may activate PI3K/AKT signaling by facilitating the crosstalk between IGF1R and EGFR/HER2. Hyperactivation of the PI3K/AKT pathway has been considered a possible mechanism leading to endocrine therapy resistance in ER+ breast cancer (47). RBP2-overexpressing MCF7 cells were sensitized to the PI3K inhibitor BKM120, but not tamoxifen (% cell viability [SD], BKM120 = 67.1 [2.7]%, tamoxifen = 89.0 [3.8]%, P < .001) (Figure 5H). Moreover, tamoxifen resistance by RBP2 was recovered by combination treatment with tamoxifen and BKM120 (% cell viability [SD], tamoxifen + BKM120 = 41.3 [5.6]%, P < .001). Collectively, these findings implied that RBP2-induced tamoxifen resistance is due to activation of the PI3K/AKT pathway via the IGF1R and ErbB signaling. Discussion Here, we demonstrated the functional role of RBP2 in mediating ER activity and tamoxifen resistance in breast cancer. RBP2 regulated both ER-dependent transcriptional activation and repression in collaboration with ER-p300 and ER-NRIP1-HDAC1, respectively (Figure 6A). The decreased expression of IGFBP4/5 by RBP2-ER-NRIP1-HDAC1 resulted in IGF1R activation. RBP also promoted the crosstalk between IGF1R and ErbB receptors by enhancing the EGFR and HER2 protein stability, resulting in activation of the PI3K/AKT pathway and tamoxifen resistance (Figure 6B). This evidence suggests RBP2 as a crucial molecular target for tamoxifen-resistant ER+ breast cancer. Figure 6. View largeDownload slide A proposed model for the regulation of tamoxifen resistance by retinoblastoma-binding protein 2 (RBP2) in breast cancer. A) Schematic model of the regulation of estrogen receptor (ER) target genes by RBP2 with ER and ER coregulators in histone demethylation–independent (left panel) and –dependent (right panel) manners. B) Proposed model for the regulation of IGF1R-ErbB-PI3K-AKT signaling by RBP2 to induce tamoxifen resistance in ER+ breast cancer. ER = estrogen receptor; H3ac = histone H3 acetylation; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; P = phosphorylation; RBP2 = retinoblastoma-binding protein 2. Figure 6. View largeDownload slide A proposed model for the regulation of tamoxifen resistance by retinoblastoma-binding protein 2 (RBP2) in breast cancer. A) Schematic model of the regulation of estrogen receptor (ER) target genes by RBP2 with ER and ER coregulators in histone demethylation–independent (left panel) and –dependent (right panel) manners. B) Proposed model for the regulation of IGF1R-ErbB-PI3K-AKT signaling by RBP2 to induce tamoxifen resistance in ER+ breast cancer. ER = estrogen receptor; H3ac = histone H3 acetylation; H3K4me2 = histone H3 lysine 4 di-methylation; H3K4me3 = histone H3 lysine 4 tri-methylation; P = phosphorylation; RBP2 = retinoblastoma-binding protein 2. Recent studies have suggested an oncogenic role of RBP2 in breast cancer. RBP2 knockout in MMTV-neu mice abrogated tumorigenesis and metastatic potential (14). In human breast tumor, amplified and overexpressed RBP2 was associated with increased cell proliferation and resistance to erlotinib (11). Combinational treatment with an RBP2 inhibitor and herceptin or gefitinib repressed colony formation in HER2+ breast cancer cells (18). To our knowledge, this study provides the first evidence of oncogenic RBP2 function in hormone-dependent breast cancer as a key regulator of ER and RTK signaling. Due to the innate and acquired resistance to endocrine therapy via the activation of ER, RTKs, and other bypass tracks, various studies have proposed combination treatment with endocrine therapy and targeted drugs for RTK signaling, such as ErbB and IGF1R inhibitors, and PI3K-AKT inhibitors (47–51). We also observed that combination treatment with tamoxifen and PI3K inhibitor BKM120 was effective for RBP2-overexpressing ER+ breast cancer cells with tamoxifen resistance. These findings suggest that ER+ breast cancers with high expression of RBP2 may be required for combined treatment with tamoxifen and other targeted drugs, such as a PI3K/AKT blocker, related to RBP2-dependent pathways. Collectively, RBP2 might serve as a new prognostic marker and therapeutic target in ER+ breast cancer. Notably, RBP2 activates not only E2-bound ER but also tamoxifen-bound ER to allow ER-dependent transcriptional regulation, leading to tamoxifen resistance. Abnormal activity of ER coregulators contributes to the development of breast cancer and the induction of endocrine resistance. For example, the upregulation of co-activators, including NCOA1/p300 complex and AIB1, is associated with the E2 agonist function of tamoxifen via ER activation and ErbB family overexpression (52–55). Consistently, high levels of corepressors, such as NRIP1, FOXA1, and DNMT1, induce tamoxifen resistance (38,56,57). Here, we verified that RBP2 induces consistent ER and p300 binding to the NRIP1 and CCND1 promoter regions, thus upregulating their expression in both E2- and tamoxifen treatment conditions, suggesting that RBP2 may block the antagonist function of tamoxifen-bound ER via sustaining the interaction with ER co-activators in the transcriptionally active ER-bound regions. In addition, NRIP1 upregulation appears to change the RBP2-containing transcriptional active complex to a transcriptional repressive complex. Although NRIP1 was initially identified as a co-activator of ER (58), a recent study demonstrated that NRIP1 expression is promptly induced in response to estrogen and subsequently mediates ER-dependent transcriptional repression (41). Based on our findings regarding RBP2/ER/NRIP1/HDAC1-mediated epigenetic repression of IGFBP4/5, increased NRIP1 by RBP2 may be crucial for a formation of RBP2/ER-containing transcriptional repressive complex. Collectively, RBP2 acts as both a co-activator and a corepressor of ER and alters tamoxifen-bound ER activity to regulate tamoxifen resistance–associated gene transcription. Accumulating studies have identified the demethylase activity–dependent oncogenic function of RBP2 (10,12,15). For example, RBP2 removes H3K4 methylation from p27 promoters, thus inducing tumorigenesis and metastasis (12). The reduction in global levels of H3K4 methylation that requires RBP2 is associated with gefitinib resistance in lung cancer (15). A recent study reported that the PI3K/AKT pathway inhibits RBP2-mediated H3K4 demethylation by phosphorylating RBP2 protein in breast cancer progression (59). However, another study demonstrated that RBP2 promotes breast cancer progression and metastasis via the histone demethylation–independent regulation of genes related to invasion, although the molecular mechanism remains unclear (14). We suggest that RBP2 have both demethylation-dependent and -independent functions in oncogenic processes. In the regulation of ER-dependent transcription, RBP2/ER/NRIP1/HDAC1 repressive complex was dependent on a demethylase activity, while RBP2 cooperated with ER and p300 in a histone demethylase activity–independent manner. Thus, the association between RBP2, ER, and ER-dependent coregulators might be very important to determine RBP2-mediated transcriptional activation or repression and its dependence on catalytic activity. Furthermore, RBP2 stabilized EGFR and HER2 proteins independently of its demethylase activity. Together, our findings suggest that RBP2 is closely linked to multiple transcriptional- and post-translational regulatory mechanisms to promote breast progression in a demethylation-dependent or demethylation-independent manner. A limitation of this study is that there is still no way to test the pharmacological effect of RBP2 inhibition on tamoxifen-resistant breast cancer. Currently available RBP2 inhibitors are not specific to RBP2 and only target the demethylase activity of RBP2, even though both histone demethylation–dependent and –independent function of RBP2 are crucial for tamoxifen resistance. The development of novel selective inhibitors for a broad spectrum of RBP2 function and further preclinical and clinical studies using the inhibitors will be needed for therapeutically targeting RBP2 in tamoxifen-resistant breast cancer. In conclusion, we demonstrated that RBP2 is crucial for tamoxifen resistance by altering ER-dependent transcription with ER coregulators and enhancing the IGF1R/EGFR/HER2 signaling pathway in various transcriptional and post-transcriptional regulatory mechanisms. Thus, RBP2 may be a prospective prognostic marker and therapeutic target in breast cancer, particularly in tamoxifen-resistant ER+ breast cancer. Funding This work was supported by National Research Foundation of Korea (NRF) grants funded by the Korean government (No. 2015R1A2A1A10052578). Notes The study funders had no role in the design of the study; the collection, analysis, or interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication. All authors declare no potential conflicts of interests. References 1 Defeo-Jones D, Huang PS, Jones REet al.  , Cloning of cDNAs for cellular proteins that bind to the retinoblastoma gene product. Nature . 1991; 352 6332: 251– 254. Google Scholar CrossRef Search ADS PubMed  2 Christensen J, Agger K, Cloos PAet al.  , RBP2 belongs to a family of demethylases, specific for tri-and dimethylated lysine 4 on histone 3. Cell.  2007; 128 6: 1063– 1076. Google Scholar CrossRef Search ADS PubMed  3 Klose RJ, Yan Q, Tothova Zet al.  , The retinoblastoma binding protein RBP2 is an H3K4 demethylase. Cell.  2007; 128 5: 889– 900. Google Scholar CrossRef Search ADS PubMed  4 Benevolenskaya EV, Murray HL, Branton Pet al.  , Binding of pRB to the PHD protein RBP2 promotes cellular differentiation. Mol Cell.  2005; 18 6: 623– 635. Google Scholar CrossRef Search ADS PubMed  5 Pasini D, Hansen KH, Christensen Jet al.  , Coordinated regulation of transcriptional repression by the RBP2 H3K4 demethylase and Polycomb-Repressive Complex 2. Genes Dev.  2008; 22 10: 1345– 1355. Google Scholar CrossRef Search ADS PubMed  6 Chaturvedi CP, Somasundaram B, Singh Ket al.  , Maintenance of gene silencing by the coordinate action of the H3K9 methyltransferase G9a/KMT1C and the H3K4 demethylase Jarid1a/KDM5A. Proc Natl Acad Sci U S A.  2012; 109 46: 18845– 18850. Google Scholar CrossRef Search ADS PubMed  7 Liefke R, Oswald F, Alvarado Cet al.  , Histone demethylase KDM5A is an integral part of the core Notch-RBP-J repressor complex. Genes Dev.  2010; 24 6: 590– 601. Google Scholar CrossRef Search ADS PubMed  8 Ge Z, Li W, Wang Net al.  , Chromatin remodeling: Recruitment of histone demethylase RBP2 by Mad1 for transcriptional repression of a Myc target gene, telomerase reverse transcriptase. FASEB J.  2010; 24 2: 579– 586. Google Scholar CrossRef Search ADS PubMed  9 Chan SW, Hong W. Retinoblastoma-binding protein 2 (Rbp2) potentiates nuclear hormone receptor-mediated transcription. J Biol Chem.  2001; 276 30: 28402– 28412. Google Scholar CrossRef Search ADS PubMed  10 Li L, Wang L, Song Pet al.  , Critical role of histone demethylase RBP2 in human gastric cancer angiogenesis. Mol Cancer.  2014; 13: 81. Google Scholar CrossRef Search ADS PubMed  11 Hou J, Wu J, Dombkowski Aet al.  , Genomic amplification and a role in drug-resistance for the KDM5A histone demethylase in breast cancer. Am J Transl Res.  2012; 4 3: 247– 256. Google Scholar PubMed  12 Teng YC, Lee CF, Li YSet al.  , Histone demethylase RBP2 promotes lung tumorigenesis and cancer metastasis. Cancer Res.  2013; 73 15: 4711– 4721. Google Scholar CrossRef Search ADS PubMed  13 Wang S, Wang Y, Wu Het al.  , RBP2 induces epithelial-mesenchymal transition in non-small cell lung cancer. PLoS One.  2013; 8 12: e84735. Google Scholar CrossRef Search ADS PubMed  14 Cao J, Liu Z, Cheung WKet al.  , Histone demethylase RBP2 is critical for breast cancer progression and metastasis. Cell Rep.  2014; 6 5: 868– 877. Google Scholar CrossRef Search ADS PubMed  15 Sharma SV, Lee DY, Li Bet al.  , A chromatin-mediated reversible drug-tolerant state in cancer cell subpopulations. Cell.  2010; 141 1: 69– 80. Google Scholar CrossRef Search ADS PubMed  16 Qi L, Zhu F, Li SHet al.  , Retinoblastoma binding protein 2 (RBP2) promotes HIF-1alpha-VEGF-induced angiogenesis of non-small cell lung cancer via the Akt pathway. PLoS One.  2014; 9 8: e106032. Google Scholar CrossRef Search ADS PubMed  17 Lin W, Cao J, Liu Jet al.  , Loss of the retinoblastoma binding protein 2 (RBP2) histone demethylase suppresses tumorigenesis in mice lacking Rb1 or Men1. Proc Natl Acad Sci U S A.  2011; 108 33: 13379– 13386. Google Scholar CrossRef Search ADS PubMed  18 Gale M, Sayegh J, Cao Jet al.  , Screen-identified selective inhibitor of lysine demethylase 5A blocks cancer cell growth and drug resistance. Oncotarget . 2016; in press. 19 Jensen EV, Jordan VC. The estrogen receptor: A model for molecular medicine. Clin Cancer Res.  2003; 9 6: 1980– 1989. Google Scholar PubMed  20 Ali S, Coombes RC. Endocrine-responsive breast cancer and strategies for combating resistance. Nat Rev Cancer.  2002; 2 2: 101– 112. Google Scholar CrossRef Search ADS PubMed  21 Green KA, Carroll JS. Oestrogen-receptor-mediated transcription and the influence of co-factors and chromatin state. Nat Rev Cancer.  2007; 7 9: 713– 722. Google Scholar CrossRef Search ADS PubMed  22 Jordan VC. Tamoxifen: A most unlikely pioneering medicine. Nat Rev Drug Discov.  2003; 2 3: 205– 213. Google Scholar CrossRef Search ADS PubMed  23 Shou J, Massarweh S, Osborne CKet al.  , Mechanisms of tamoxifen resistance: Increased estrogen receptor-HER2/neu cross-talk in ER/HER2-positive breast cancer. J Natl Cancer Inst.  2004; 96 12: 926– 935. Google Scholar CrossRef Search ADS PubMed  24 Jin K, Park S, Teo WWet al.  , HOXB7 is an ERalpha cofactor in the activation of HER2 and multiple ER Target genes leading to endocrine resistance. Cancer Discov.  2015; 5 9: 944– 959. Google Scholar CrossRef Search ADS PubMed  25 Campbell RA, Bhat-Nakshatri P, Patel NMet al.  , Phosphatidylinositol 3-kinase/AKT-mediated activation of estrogen receptor alpha: A new model for anti-estrogen resistance. J Biol Chem.  2001; 276 13: 9817– 9824. Google Scholar CrossRef Search ADS PubMed  26 Fan P, Agboke FA, Cunliffe HEet al.  , A molecular model for the mechanism of acquired tamoxifen resistance in breast cancer. Eur J Cancer.  2014; 50 16: 2866– 2876. Google Scholar CrossRef Search ADS PubMed  27 Osborne CK, Schiff R. Mechanisms of endocrine resistance in breast cancer. Annu Rev Med.  2011; 62: 233– 247. Google Scholar CrossRef Search ADS PubMed  28 Gottardis MM, Jordan VC. Development of tamoxifen-stimulated growth of MCF-7 tumors in athymic mice after long-term antiestrogen administration. Cancer Res.  1988; 48 18: 5183– 5187. Google Scholar PubMed  29 Gottardis MM, Wagner RJ, Borden ECet al.  , Differential ability of antiestrogens to stimulate breast cancer cell (MCF-7) growth in vivo and in vitro. Cancer Res.  1989; 49 17: 4765– 4769. Google Scholar PubMed  30 Yao K, Lee ES, Bentrem DJet al.  , Antitumor action of physiological estradiol on tamoxifen-stimulated breast tumors grown in athymic mice. Clin Cancer Res . 2000; 6 5: 2028– 2036. Google Scholar PubMed  31 Jordan VC. Selective estrogen receptor modulation: Concept and consequences in cancer. Cancer Cell.  2004; 5 3: 207– 213. Google Scholar CrossRef Search ADS PubMed  32 Cho MH, Park JH, Choi HJet al.  , DOT1L cooperates with the c-Myc-p300 complex to epigenetically derepress CDH1 transcription factors in breast cancer progression. Nat Commun.  2015; 6: 7821. Google Scholar CrossRef Search ADS PubMed  33 Curtis C, Shah SP, Chin SFet al.  , The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature.  2012; 486 7403: 346– 352. Google Scholar PubMed  34 Lee JY, Won HY, Park JHet al.  , MEL-18 loss mediates estrogen receptor-alpha downregulation and hormone independence. J Clin Invest.  2015; 125 5: 1801– 1814. Google Scholar CrossRef Search ADS PubMed  35 Gyorffy B, Lanczky A, Eklund ACet al.  , An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients. Breast Cancer Res Treat.  2010; 123 3: 725– 731. Google Scholar CrossRef Search ADS PubMed  36 Ahn BY, Elwi AN, Lee Bet al.  , Genetic screen identifies insulin-like growth factor binding protein 5 as a modulator of tamoxifen resistance in breast cancer. Cancer Res.  2010; 70 8: 3013– 3019. Google Scholar CrossRef Search ADS PubMed  37 Walker G, MacLeod K, Williams ARet al.  , Insulin-like growth factor binding proteins IGFBP3, IGFBP4, and IGFBP5 predict endocrine responsiveness in patients with ovarian cancer. Clin Cancer Res.  2007; 13 5: 1438– 1444. Google Scholar CrossRef Search ADS PubMed  38 Chan CM, Lykkesfeldt AE, Parker MGet al.  , Expression of nuclear receptor interacting proteins TIF-1, SUG-1, receptor interacting protein 140, and corepressor SMRT in tamoxifen-resistant breast cancer. Clin Cancer Res . 1999; 5 11: 3460– 3467. Google Scholar PubMed  39 Brown LA, Johnson K, Leung Set al.  , Co-amplification of CCND1 and EMSY is associated with an adverse outcome in ER-positive tamoxifen-treated breast cancers. Breast Cancer Res Treat.  2010; 121 2: 347– 354. Google Scholar CrossRef Search ADS PubMed  40 Tu S, Teng YC, Yuan Cet al.  , The ARID domain of the H3K4 demethylase RBP2 binds to a DNA CCGCCC motif. Nat Struct Mol Biol.  2008; 15 4: 419– 421. Google Scholar CrossRef Search ADS PubMed  41 Carroll JS, Meyer CA, Song Jet al.  , Genome-wide analysis of estrogen receptor binding sites. Nat Genet.  2006; 38 11: 1289– 1297. Google Scholar CrossRef Search ADS PubMed  42 Wei LN, Hu X, Chandra Det al.  , Receptor-interacting protein 140 directly recruits histone deacetylases for gene silencing. J Biol Chem.  2000; 275 52: 40782– 40787. Google Scholar CrossRef Search ADS PubMed  43 Castet A, Boulahtouf A, Versini Get al.  , Multiple domains of the Receptor-Interacting Protein 140 contribute to transcription inhibition. Nucleic Acids Res.  2004; 32 6: 1957– 1966. Google Scholar CrossRef Search ADS PubMed  44 Baxter RC. IGF binding proteins in cancer: Mechanistic and clinical insights. Nat Rev Cancer.  2014; 14 5: 329– 341. Google Scholar CrossRef Search ADS PubMed  45 Gallardo A, Lerma E, Escuin Det al.  , Increased signalling of EGFR and IGF1R, and deregulation of PTEN/PI3K/Akt pathway are related with trastuzumab resistance in HER2 breast carcinomas. Br J Cancer.  2012; 106 8: 1367– 1373. Google Scholar CrossRef Search ADS PubMed  46 Huang X, Gao L, Wang Set al.  , Heterotrimerization of the growth factor receptors erbB2, erbB3, and insulin-like growth factor-i receptor in breast cancer cells resistant to herceptin. Cancer Res.  2010; 70 3: 1204– 1214. Google Scholar CrossRef Search ADS PubMed  47 Miller TW, Balko JM, Arteaga CL. Phosphatidylinositol 3-kinase and antiestrogen resistance in breast cancer. J Clin Oncol.  2011; 29 33: 4452– 4461. Google Scholar CrossRef Search ADS PubMed  48 Hudis CA. Trastuzumab—mechanism of action and use in clinical practice. N Engl J Med.  2007; 357 1: 39– 51. Google Scholar CrossRef Search ADS PubMed  49 Johnston SR. Enhancing endocrine therapy for hormone receptor-positive advanced breast cancer: Cotargeting signaling pathways. J Natl Cancer Inst.  2015; 107 10:djv212. 50 Robertson JF, Ferrero JM, Bourgeois Het al.  , Ganitumab with either exemestane or fulvestrant for postmenopausal women with advanced, hormone-receptor-positive breast cancer: A randomised, controlled, double-blind, phase 2 trial. Lancet Oncol.  2013; 14 3: 228– 235. Google Scholar CrossRef Search ADS PubMed  51 Bachelot T, Bourgier C, Cropet Cet al.  , Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: A GINECO study. J Clin Oncol.  2012; 30 22: 2718– 2724. Google Scholar CrossRef Search ADS PubMed  52 Zwart W, Theodorou V, Kok Met al.  , Oestrogen receptor-co-factor-chromatin specificity in the transcriptional regulation of breast cancer. EMBO J.  2011; 30 23: 4764– 4776. Google Scholar CrossRef Search ADS PubMed  53 Osborne CK, Bardou V, Hopp TAet al.  , Role of the estrogen receptor coactivator AIB1 (SRC-3) and HER-2/neu in tamoxifen resistance in breast cancer. J Natl Cancer Inst.  2003; 95 5: 353– 361. Google Scholar CrossRef Search ADS PubMed  54 Hurtado A, Holmes KA, Geistlinger TRet al.  , Regulation of ERBB2 by oestrogen receptor-PAX2 determines response to tamoxifen. Nature.  2008; 456 7222: 663– 666. Google Scholar CrossRef Search ADS PubMed  55 Wu RC, Qin J, Yi Pet al.  , Selective phosphorylations of the SRC-3/AIB1 coactivator integrate genomic reponses to multiple cellular signaling pathways. Mol Cell.  2004; 15 6: 937– 949. Google Scholar CrossRef Search ADS PubMed  56 Fu X, Jeselsohn R, Pereira Ret al.  , FOXA1 overexpression mediates endocrine resistance by altering the ER transcriptome and IL-8 expression in ER-positive breast cancer. Proc Natl Acad Sci U S A.  2016; 113 43: E6600– E6609. Google Scholar CrossRef Search ADS PubMed  57 Phuong NT, Kim SK, Lim SCet al.  , Role of PTEN promoter methylation in tamoxifen-resistant breast cancer cells. Breast Cancer Res Treat.  2011; 130 1: 73– 83. Google Scholar CrossRef Search ADS PubMed  58 Cavailles V, Dauvois S, L'Horset Fet al.  , Nuclear factor RIP140 modulates transcriptional activation by the estrogen receptor. EMBO J.  1995; 14 15: 3741– 3751. Google Scholar PubMed  59 Spangle JM, Dreijerink KM, Groner ACet al.  , PI3K/AKT signaling regulates H3K4 methylation in breast cancer. Cell Rep.  2016; 15 12: 2692– 2704. Google Scholar CrossRef Search ADS PubMed  60 Ross-Innes CS, Stark R, Teschendorff AEet al.  , Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature.  2012; 481 7381: 389– 393. Google Scholar CrossRef Search ADS PubMed  61 Coser KR, Wittner BS, Rosenthal NFet al.  , Antiestrogen-resistant subclones of MCF-7 human breast cancer cells are derived from a common monoclonal drug-resistant progenitor. Proc Natl Acad Sci U S A.  2009; 106 34: 14536– 14541. Google Scholar CrossRef Search ADS PubMed  © The Author 2017. Published by Oxford University Press. All rights reserved. For Permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

JNCI: Journal of the National Cancer InstituteOxford University Press

Published: Oct 11, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off