Upregulation of IRS1 Enhances IGF1 Response in Y537S and D538G ESR1 Mutant Breast Cancer Cells

Upregulation of IRS1 Enhances IGF1 Response in Y537S and D538G ESR1 Mutant Breast Cancer Cells Abstract Increased evidence suggests that somatic mutations in the ligand-binding domain of estrogen receptor [ER (ERα/ESR1)] are critical mediators of endocrine-resistant breast cancer progression. Insulinlike growth factor-1 (IGF1) is an essential regulator of breast development and tumorigenesis and also has a role in endocrine resistance. A recent study showed enhanced crosstalk between IGF1 and ERα in ESR1 mutant cells, but detailed mechanisms are incompletely understood. Using genome-edited MCF-7 and T47D cell lines harboring Y537S and D538G ESR1 mutations, we characterized altered IGF1 signaling. RNA sequencing revealed upregulation of multiple genes in the IGF1 pathway, including insulin receptor substrate-1 (IRS1), consistent in both Y537S and D538G ESR1 mutant cell line models. Higher IRS1 expression was confirmed by quantitative reverse transcription polymerase chain reaction and immunoblotting. ESR1 mutant cells also showed increased levels of IGF-regulated genes, reflected by activation of an IGF signature. IGF1 showed increased sensitivity and potency in growth stimulation of ESR1 mutant cells. Analysis of downstream signaling revealed the phosphoinositide 3-kinase (PI3K)–Akt axis as a major pathway mediating the enhanced IGF1 response in ESR1 mutant cells. Decreasing IRS1 expression by small interfering RNA diminished the increased sensitivity to IGF1. Combination treatment with inhibitors against IGF1 receptor (IGF1R; OSI-906) and ER (fulvestrant) showed synergistic growth inhibition in ESR1 mutant cells, particularly at lower effective concentrations. Our study supports a critical role of enhanced IGF1 signaling in ESR1 mutant cell lines, pointing toward a potential for cotargeting IGF1R and ERα in endocrine-resistant breast tumors with mutant ESR1. Estrogen deprivation has become a prevalent therapeutic strategy for >70% of breast cancers that harbor estrogen receptor (ER)α expression (1). Despite improved patient outcomes, acquired endocrine resistance develops in a subset of ERα-positive breast cancers via various molecular mechanisms (2, 3). Recent studies have documented a group of recurrent hotspot mutations in ESR1, Y537S and D538G being the most frequent, that cluster in the ligand-binding domain and overlap with activation function 2 (4). Molecular modeling indicates that the hotspot mutations at positions Y537 and D538 alter the position of helix 12 to favor an agonist conformation of the receptor (5). Recent mechanistic studies performed by us and others, employing either transfected or genome-edited ESR1 mutant cells, show that mutant ESR1 has ligand-independent activity and diminished sensitivity toward antiestrogen drugs (4). Clinical studies have documented rare ESR1 mutations in primary breast cancers, but increased frequency in metastatic lesions and circulation-free DNA, suggesting a potential role of acquired ESR1 mutations in facilitating metastasis (4). Given studies suggesting poor outcomes in patients harboring breast cancers with ESR1 mutations (4), identification of new targets and design of novel therapeutic strategies have gained urgency. Multiple recent preclinical studies have uncovered promising therapeutic targets in breast cancer cells harboring ESR1 mutations. Harrod et al. (6) highlighted the potential utility of a CDK7 inhibitor to block growth in MCF-7 cells with CRISPR-edited Y537S. Mao et al. (7) found increased unfolded protein response in CRIPSR-edited Y537S and D538G ESR1 mutant cells. Recent findings from Gelsomino et al. (8) identified enhanced crosstalk between mutant ERα and insulinlike growth factor-1 receptor (IGF1R), proposing a role in tamoxifen resistance, indicating a potential for combination therapy by cotargeting ERα and mammalian target of rapamycin (mTOR) in ESR1 mutant tumors. Our recent transcriptomic analysis of genome-edited MCF-7 and T47D Y537S and D538G ESR1 cell lines revealed mutation site- and context-dependent gene expression changes compared with wild-type (WT) ESR1 (9). Similar to the findings by Gelsomino et al. (8), RNA-sequencing (RNA-seq) analysis revealed altered expression of insulinlike growth factor-1 (IGF1) pathway members in both ESR1 mutant cell lines. In this study, we performed a preclinical study to characterize the mechanisms underlying the augmented IGF1 response in ESR1 mutant cells and evaluated the strategy of cotargeting ER and IGF1R for future therapeutic development. Materials and Methods Cell culture T47D and MCF-7 cells were obtained from American Type Culture Collection. Both cell lines were authenticated at the University of Arizona Genetics Core. T47D and MCF-7 cells were maintained in RPMI 1640 plus 10% fetal bovine serum and Dulbecco’s modified Eagle medium plus 10% fetal bovine serum, respectively. Tamoxifen-resistant (TamR) and long-term estrogen deprivation (LTED) MCF-7 and ZR75-1 cell lines were gifts from Dr. Rachel Schiff (Baylor College of Medicine, Houston, TX). For hormone treatment experiments, cells were deprived of steroid hormones by placement in phenol red–free Iscove’s modified Eagle medium with 10% and 5% charcoal stripped serum (CSS) for T47D and MCF-7, respectively. CSS was purchased from Gibco (catalog no. 12676; Waltham, MA). 17β-Estradiol (E2) and fulvestrant were obtained from Sigma-Aldrich (catalog no. E8875; St. Louis, MO) and Tocris Bioscience (catalog no. 1047; Bristol, United Kingdom), respectively. IGF1 was purchased from GroPep Bioreagents (catalog no. AM001; Tbarton, South Australia, Australia). IGF-II (catalog no. 110-12), vascular endothelial growth factor (VEGF; catalog no. 100-20), platelet-derived growth factor (PDGF)–BB (catalog no. 100-14B), and fibroblast growth factor-basic (bFGF; catalog no. 100-18B) were from PeproTech (Rocky Hill, NJ). OSI-906 (catalog no. S1091), wortmannin (catalog no. S2758), BX-795 (catalog no. S1274), and BMS-754807 (catalog no. S1124) were obtained from Selleck Chemicals (Houston, TX). U0126 (catalog no. 1144) and rapamycin (catalog no. 1292) were purchased from Tocris Bioscience. Immunoblotting After 3 days in CSS, 120,000 (MCF-7) and 90,000 cells (T47D) were plated in six-well plates. For IGF1 stimulation, cells were starved in serum-free Iscove’s modified Eagle medium for 24 hours and then treated with IGF1 for 15 minutes. For small interfering RNA (siRNA) knockdown experiments, cells were reverse transfected with siRNA-targeting insulin receptor substrate-1 (IRS1; catalog no. L-003015; Dharmacon, Lafayette, CO) and/or IGF1R (catalog no. L-003012; Dharmacon) for 24 hours and subjected to IGF1 stimulation. The cells were lysed with RIPA buffer, sonicated, 60 μg of protein separated using a sodium dodecyl sulfate polyacrylamide gel electrophoresis gel, and then transferred onto polyvinylidene difluoride membrane. Membranes were immunostained using the following antibodies: phosphorylated (p-)IGF1R-Y1135 (catalog no. 3918), IGF1R (catalog no.9750), p-IRS1–S320 (catalog no. 2384), p-AKT–T308 (catalog no. 13038), p-AKT–S473 (catalog no. 4060), AKT (catalog no. 2920), p-ERK1/2–T202/T204 (catalog no. 4377), ERK (catalog no. 9102), and IR (catalog no. 3025), all purchased from Cell Signaling Technology (Danvers, MA). IRS1 (catalog no. sc-7200) and β-actin (catalog no. A3853) antibodies were obtained from Santa Cruz Biotechnology (Dallas, TX) and Sigma-Aldrich, respectively. Quantifications were performed using ImageJ software (National Institutes of Health). Immunoblots were repeated twice unless otherwise stated. RNA analysis using quantitative reverse transcription polymerase chain reaction Posthormone-deprived cells were seeded into six-well plates and treated with 1 nM E2, 1 μM ICI 182,780, or their combination. RNA samples were extracted after 24 hours and complementary DNA synthesized using an iScript kit (catalog no. 1708890; Bio-Rad, Hercules, CA). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) reactions were performed with SYBR Green Supermix (catalog no. 1726275; Bio-Rad), the ΔΔ threshold cycle method was used to analyze relative messenger RNA fold changes, and RPLP0 levels were measured as the internal control. Primers sequences are as follows: RPLP0 (forward), 5′-TAAACCCTGCGTGGCAATC-3′; RPLP0 (reverse), 5′-TTGTCTGCTCCCACAATGAAA-3′; ESR1 (forward), 5′-GAGTATGATCCTACCAGACCCTTC-3′; ESR1 (reverse), 5′-CCTGATCATGGAGGGTCAAATC-3′; IRS1 (forward), 5′-TCTGCTCAGCGTTGGTG-3′; and IRS1 (reverse), 5′-GTGCATGCTCTTGGGTTTG-3′. Growth assays As previously described (8), individual MCF-7 or T47D CRISPR-edited clones were evenly pooled after 3 days of hormone deprivation in CSS and plated into 96-well plates using 2500 cells/well (MCF-7) or 4000 cells/well (T47D). After 24 hours, the cells were treated with various concentrations of growth factors or inhibitors, as well as vehicle control. For time-course growth assay, the cells were harvested after 0, 2, 4, 6, and 9 days and quantified with the FluoReporter kit (catalog no. F2962; Life Technologies, Carlsbad, CA) following the manufacturer’s protocol, and half maximal inhibitory concentrations or half maximal effective concentrations (EC50s) were calculated using the Prism statistical package (GraphPad Software, La Jolla, CA). For analysis of drug synergy, the combination index values were calculated using the CalcuSync package (Biosoft, Great Shelford, United Kingdom). All experiments were performed with six biological replicates. Calculation of IGF activation score The IGF activation score was calculated for each cell line as described previously (10). Briefly, the activation score was calculated as the Pearson correlation between the IGF gene signature pattern (by using 1 and −1 for up and down, respectively) and the gene expression values of each cell line. The gene expression values were the log2 (transcripts per million + 1) of Salmon (v.0.6.0) mapped reads to Ensembl gene-level identification numbers from Bahreini et al. (9). The microarray probes from the IGF upregulated and downregulated genes were mapped to Ensembl identification numbers using the hgu133a2.db annotation package [hgu133a2.db: Affymetrix Human Genome U133A 2.0 Array annotation data (chip hgu133a2); Affymetrix, Santa Clara, CA]. This was performed using R version 3.2.2, and the R code is provided as a supplemental text file. Statistical analysis Each experiment was performed with corresponding biological and technical replicates and repeated, as indicated. Multiple statistical tests were used to assess the statistical significance depending on the design of the experiments, and the P value was calculated accordingly (*P < 0.05; **P < 0.01). Results Y537S and D538G mutant cells exhibit upregulation of IRS1 We have previously shown a notable similarity between genes for which expression is altered by E2 or IGF1 (10). In this study, we investigated levels of these same IGF1-stimulated genes in ESR1 mutant cells by performing a correlation analysis between the RNA-seq data sets of ESR1 mutant cells and the previously reported IGF1 signature panel of 976 transcripts (814 genes) (10). Replicates from each group of RNA-seq results were compared with the IGF1 signature individually, and Pearson correlation coefficient values were calculated to indicate the similarities (Fig. 1A). In the absence of E2, gene expression in both MCF-7 and T47D ESR1 mutant cells showed a significantly increased similarity to the IGF gene signature pattern compared with WT groups. Consistent with previous results (10), E2 stimulation led to a higher similarity to the IGF gene signature in all cell line models. We then directly examined genes in the IGF1 signaling pathway (n = 39; Supplemental Table 1) that showed a fold change >1.5 (P < 0.005) in ESR1 mutant cells (Fig. 1B). We found that IGF2, IGFBP6, IRS1, IGFBP4, and INSR were consistently altered in both ESR1 mutant cell line models compared with WT cells. IRS1, which encodes an essential adaptor in the IGF1 pathway, was the most consistently upregulated gene in both mutations and cell lines in the entire panel. IGF1R levels were also induced in MCF-7 ESR1 mutant cells. Interestingly, the INSR gene exhibited decreased levels in both cell lines (Supplemental Fig. 1A and 1B), perhaps reflecting a secondary negative-feedback regulation following increased IGF1R/IRS1 signaling. Moreover, the transcript reads from RNA-seq suggest no difference in IR isoform ratios (IR-A/IR-B) between different mutants of each cell line (Supplemental Fig. 1C). Upregulation of IRS1 was validated using immunoblotting (Fig. 1C) and qRT-PCR (Fig. 1D). Increased protein levels of IRS1 were observed in both ESR1 mutant cell lines under hormone-deprived conditions, whereas higher IGF1R expression was only seen in MCF-7 mutant cells, a result that correlated well with the RNA-seq data. The qRT-PCR results also demonstrated higher basal level of IRS1 in ESR1 mutant cell lines, with the exception of T47D-D538G clones in which the increase was not noticeable. Similar data were observed when measuring IRS1 messenger RNA in individual clones that constitute the pool (Supplemental Fig. 2A). qRT-PCR results indicated that IRS1 levels were induced by E2 and inhibited by fulvestrant treatment, which is consistent with previous studies characterizing IRS1 as an ER-induced gene (11, 12). Decreasing ESR1 expression with siRNA attenuated the higher IRS1 levels in ESR1 mutant cells, confirming the dependency of the IRS1 upregulation on the genome-edited mutant ESR1 (Supplemental Fig. 2B). Figure 1. View largeDownload slide MCF-7 and T47D cells with CRISPR-edited ESR1 Y537S and D538G mutations exhibit upregulation of IGF signaling components and downstream activity. (A) RNA-seq results from all four replicates of each group were aligned and compared with the IGF gene signature panel. Pearson values from vehicle groups were compared between WT and two mutants with one-way analysis of variance (ANOVA) . (B) The heat maps from our previous RNA-seq shows normalized log2 fold change of IGF1 pathway related genes differentially regulated in mutants vs WT (fold change >1.5; P < 0.005). (C) Posthormone-deprived cells were pooled, and protein samples were extracted and subjected for immunoblotting with β-actin as the internal control. (D) Hormone-deprived MCF-7 and T47D cells were treated with vehicle, 1 nM of E2, or combination of 1 nM E2 and 1 μM fulvestrant for 24 hours. RNA was isolated, and qRT-PCR was performed (one-way ANOVA for comparison of basal level). Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Figure 1. View largeDownload slide MCF-7 and T47D cells with CRISPR-edited ESR1 Y537S and D538G mutations exhibit upregulation of IGF signaling components and downstream activity. (A) RNA-seq results from all four replicates of each group were aligned and compared with the IGF gene signature panel. Pearson values from vehicle groups were compared between WT and two mutants with one-way analysis of variance (ANOVA) . (B) The heat maps from our previous RNA-seq shows normalized log2 fold change of IGF1 pathway related genes differentially regulated in mutants vs WT (fold change >1.5; P < 0.005). (C) Posthormone-deprived cells were pooled, and protein samples were extracted and subjected for immunoblotting with β-actin as the internal control. (D) Hormone-deprived MCF-7 and T47D cells were treated with vehicle, 1 nM of E2, or combination of 1 nM E2 and 1 μM fulvestrant for 24 hours. RNA was isolated, and qRT-PCR was performed (one-way ANOVA for comparison of basal level). Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Y537S and D538G ESR1 mutant cells display enhanced proliferation in response to IGF1 Given the alteration in IGF signaling components, including the constitutive expression of IRS1, we hypothesized that ESR1 mutant cells would show an elevated response to IGF1. We therefore performed time-course proliferation assays with or without IGF1 after hormone deprivation and serum-free starvation (Fig. 2A). A hallmark of ESR1 mutant cells is the greater estrogen-independent growth in CSS (9). In the complete absence of serum, however, the ligand-independent growth of T47D mutant cells was not discerned (Fig. 2A), whereas MCF-7 mutant cells still maintained a stronger proliferation rate compared with WT controls, suggesting that the ligand-independent growth of T47D mutant cells depends on additional factors in CSS. When exposed to IGF1, T47D-Y537S cells exhibited a significantly (P < 0.0001) and approximately twofold stronger IGF1 response compared with WT control cells, whereas D538G mutant cells showed only a weak IGF1-induced proliferation that did not reach significance (Fig. 2A, bottom panels). In MCF-7, both mutations resulted in marked IGF1-mediated induction of proliferation, with D538G mutant cells showing a stronger response (Fig. 2A, top panels). Figure 2. View largeDownload slide ESR1 mutant cells display enhanced IGF1-stimulated growth response. (A) MCF-7 and T47D cells were hormone deprived for 3 days, followed by incubation in serum-free medium for 24 hours. Cells were then seeded into 96-well plates and treated with or without 100 ng/mL of IGF1 for up to 8 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Growth was normalized to day 0 and presented as fold change (FC). Two-way analysis of variance (ANOVA) was performed to compare the time-course growth rates between WT and mutant cells. Representative data shown from two independent experiments. (B) EC50 values were determined by IGF1 dose-response growth curves following the identical procedure as in (A) with IGF1 concentrations as indicated and 20 pM of E2. Mean EC50s are shown in the table with three (T47D) or two (MCF-7) independent experiments. One-way ANOVA was performed to compare the EC50s of WT and mutant cells. (C) Hormone-deprived and serum-free starved MCF-7 and T47D cells were treated with 100 ng/mL IGF1, 100 ng/mL IGF-II, 10 ng/mL VEGF, 10 ng/mL fibroblast growth factor-basic (bFGF), 10 ng/mL PDGF-BB, or 50 ng/mL EGF for 6 (MCF-7) or 7 (T47D) days. FCs induced by growth factors were obtained by normalizing to the average of vehicle controls. One-way ANOVA was applied to compare the growth factor effects between WT and mutant cells. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Figure 2. View largeDownload slide ESR1 mutant cells display enhanced IGF1-stimulated growth response. (A) MCF-7 and T47D cells were hormone deprived for 3 days, followed by incubation in serum-free medium for 24 hours. Cells were then seeded into 96-well plates and treated with or without 100 ng/mL of IGF1 for up to 8 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Growth was normalized to day 0 and presented as fold change (FC). Two-way analysis of variance (ANOVA) was performed to compare the time-course growth rates between WT and mutant cells. Representative data shown from two independent experiments. (B) EC50 values were determined by IGF1 dose-response growth curves following the identical procedure as in (A) with IGF1 concentrations as indicated and 20 pM of E2. Mean EC50s are shown in the table with three (T47D) or two (MCF-7) independent experiments. One-way ANOVA was performed to compare the EC50s of WT and mutant cells. (C) Hormone-deprived and serum-free starved MCF-7 and T47D cells were treated with 100 ng/mL IGF1, 100 ng/mL IGF-II, 10 ng/mL VEGF, 10 ng/mL fibroblast growth factor-basic (bFGF), 10 ng/mL PDGF-BB, or 50 ng/mL EGF for 6 (MCF-7) or 7 (T47D) days. FCs induced by growth factors were obtained by normalizing to the average of vehicle controls. One-way ANOVA was applied to compare the growth factor effects between WT and mutant cells. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Dose-response studies were performed to evaluate the sensitivity and potency of IGF1 (Fig. 2B). Both mutations resulted in increased potency of IGF1-stimulated proliferation, particularly at concentrations >1 ng/mL of IGF1 (Fig. 2B, top panels), and this was observed in both cell line models. Examining sensitivity, EC50 calculated from the dose response curves suggested no difference between T47D WT and mutant cells, whereas MCF-7 Y537S and D538G mutants both showed a threefold to fourfold decrease in EC50 (Fig. 2B, bottom panel). We next determined whether the enhanced IGF1 response was a unique phenotype in ESR1 mutant cells or whether it is present in other endocrine-resistant models. We therefore tested IGF1 response in TamR breast cancer cells and in LTED models, mimicking resistance to aromatase inhibitors (Supplemental Fig. 3). These analyses showed that there was no noteworthy difference in IGF1 response between parental and the endocrine-resistant TamR and LTED models (Supplemental Fig. 3A and 3B). Furthermore, the dose-response curves in MCF-7 cells indicated that both TamR and LTED cells exhibited reduced sensitivity toward IGF1 exposure: a 4-fold and 18-fold increase in EC50 values was detected in the MCF-7–LTED and TamR models, respectively (Supplemental Fig. 3C). Together, these results suggest that the enhanced IGF1 response might be a unique property in endocrine-resistant cells with ESR1 mutations. We next determined the effect of other growth factors known to play a role in breast cancer progression to determine whether the IGF1 effects are unique or whether the ESR1 mutations sensitize the cells to activation of other growth factor receptors. Therefore, we measured growth response of ESR1 WT and mutant cells to IGF-II, VEGF, PDGF, fibroblast growth factor-basic (bFGF), and epidermal growth factor (EGF) (Fig. 2C). Although we did observe increased response of the Y537S mutant cells to IGF-II, bFGF, PDGF, and EGF, consistent increase in both cell lines for both mutations was only observed for IGF1. Phosphoinositide 3-kinase–Akt axis mediates the enhanced IGF1 response in ESR1 mutant cells We next investigated which downstream signaling pathways mediate the enhanced IGF1 response in ESR1 mutant cells (Fig. 3). Cells were stimulated with increasing concentrations of IGF1 for 15 minutes, and the phosphorylation status of IGF1R (Y1135) and canonical IGF1 downstream substrates Akt (T308 and S473) and ERK1/2 (T202/T204) were measured (Fig. 3). Consistent with the growth response data, greater IGF1R phosphorylation at Y1135 was observed in both MCF-7 (Fig. 3A and 3B) and T47D (Fig. 3C and 3D) and Y537S and D538G mutant cells, particularly at higher concentrations of IGF1 stimulation (80 or 100 ng/mL). The two phosphorylation sites (T308 and S473) of Akt are phosphorylated by different kinases and involved in distinct biological functions (13, 14). We noted that only pT308-Akt was increased in ESR1 mutant cells, whereas phosphorylation at pS473-Akt displayed higher basal levels but weaker or no IGF1-induced stimulation. In contrast, p-ERK1/2 (T202/T204) showed a cell line–specific induction, with greater induction in T47D but not in MCF-7 ESR1 mutants compared with WT. Figure 3. View largeDownload slide PI3K-Akt axis mediates the enhanced IGF1 response in ESR1 mutant cells. (A and C) MCF-7 and T47D cells were hormone deprived for 3 days and starved in serum-free medium for 24 hours. Cells were then stimulated with various concentrations of IGF1 for 15 minutes. Proteins were isolated, and immunoblots were performed with antibodies as indicated, with β-actin as loading control. (B and D) The intensities of bands were quantified by ImageJ (National Institutes of Health), and relative phosphorylation levels were calculated by correcting phosphorylation levels to total protein and then normalizing to vehicle controls. Representative data shown from two independent experiments. Figure 3. View largeDownload slide PI3K-Akt axis mediates the enhanced IGF1 response in ESR1 mutant cells. (A and C) MCF-7 and T47D cells were hormone deprived for 3 days and starved in serum-free medium for 24 hours. Cells were then stimulated with various concentrations of IGF1 for 15 minutes. Proteins were isolated, and immunoblots were performed with antibodies as indicated, with β-actin as loading control. (B and D) The intensities of bands were quantified by ImageJ (National Institutes of Health), and relative phosphorylation levels were calculated by correcting phosphorylation levels to total protein and then normalizing to vehicle controls. Representative data shown from two independent experiments. A parallel set of time-dependent IGF1 stimulation experiments were performed to compare the rates of IGF1 downstream substrate phosphorylation response in the various cell lines (Supplemental Fig. 4). In T47D WT cells, p-IGF1R (Y1135) was maximal after 45 minutes and declined at 60 minutes, whereas in both Y537S and D538G mutant cells p-IGF1R increased rapidly with a maximal response within only 10 minutes (Supplemental Fig. 4B). In MCF-7 cells, similar time-dependent effects were seen, but differences in p-IGF1R were found in ESR1 mutant cells compared with WT cells (Supplemental Fig. 4A). Consistently, pT308-Akt not pS473 site was differentially affected in cells with both types of ESR1 mutations. pT308-Akt reached the peak intensity within 5 and 10 minutes of IGF1 stimulation in T47D and MCF-7 mutant cells, whereas 10 and 30 minutes were required for the corresponding WT cells. Consistently, phosphorylation of S6 kinase also exhibited a stronger response in both ESR1 mutant cells, with a higher basal level in T47D cell line. We also discerned a more rapid ERK response in MCF-7-Y537S mutant cell line but not in other mutant cells. Given the observed effects on IGF sensitivity and potency, we next used a number of inhibitors to test the dependency of the observed increased IGF1 signaling on the phosphoinositide 3-kinase (PI3K)–Akt axis (Supplemental Fig. 4C). Specifically, we used wortmannin, BX-579, UO126, and rapamycin to inhibit PI3K, PDK1, mitogen-activated protein kinase kinase (MEK), and mTOR, respectively. In addition, as a control, we inhibited IGF1R with OSI-906 that, as expected, completely abrogated IGF1-induced growth. Effects of inhibition of PI3K, mTOR, and MEK were cell line dependent, with mTOR inhibitors blocking IGF1-induced growth in ESR1 mutant T47D cells, but not MCF-7 cells. In contrast, inhibition of PI3K and MEK was effective in ESR1 mutant MCF-7, but not in T47D cells. The only inhibitor that consistently resulted in inhibition of IGF1-induced growth in both cell line models was BX-579, an inhibitor of PDK1, a kinase that specifically catalyzes T308-Akt phosphorylation. In addition, to further confirm this signaling enhancement is a unique mechanism for ESR1 mutation, we examined the status of IGF1 downstream regulators in TamR and LTED models. TamR and LTED cell models showed decreased IGF1-induced phosphorylation of IGF1R (Y1135), IRS1 (S320), and Akt (T308 and S473) compared with parental controls (Supplemental Fig. 5A), which is consistent with previous studies using these cells and TamR breast cancers (15, 16). In contrast, an identical IGF1 stimulation triggered consistent and robust enhanced IGF1 downstream substrate phosphorylation in ESR1-mutant cells (Supplemental Fig. 5B). Knockdown of IRS1 attenuates the enhanced IGF1 response in ESR1 mutant cells Given the consistent and high upregulation of IRS1 in the ESR1 mutant cells, we next performed knockdown studies to test the functional significance of IRS1 in the increased IGF1 response. Growth curves demonstrated that loss of IRS1 reduced IGF1-induced proliferation in both WT and mutant cell lines, with stronger effects in the mutant cells (Fig. 4A). Figure 4. View largeDownload slide Knockdown of IRS1 attenuates the IGF1 response in ESR1 mutant cells. (A) IRS1 or scramble siRNAs were reverse transfected into hormone-deprived MCF-7 for 24 hours. Cells were then starved in serum-free medium and seeded into 96-well plates with 2500 (MCF-7) cells per well and treated with or without 100 ng/mL IGF1 for up to 6 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Cell numbers in scramble and IRS1 knockdown groups with IGF1 simulation were compared by Student t test. (B) MCF-7 and T47D cells were hormone deprived and transfected with IRS1, IGF1R, or scramble siRNA (siScramble). Cells were then starved in serum-free medium and treated with 100 ng/mL IGF1 for 15 minutes. Immunoblots were performed with antibodies as indicated. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. veh, vehicle. Figure 4. View largeDownload slide Knockdown of IRS1 attenuates the IGF1 response in ESR1 mutant cells. (A) IRS1 or scramble siRNAs were reverse transfected into hormone-deprived MCF-7 for 24 hours. Cells were then starved in serum-free medium and seeded into 96-well plates with 2500 (MCF-7) cells per well and treated with or without 100 ng/mL IGF1 for up to 6 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Cell numbers in scramble and IRS1 knockdown groups with IGF1 simulation were compared by Student t test. (B) MCF-7 and T47D cells were hormone deprived and transfected with IRS1, IGF1R, or scramble siRNA (siScramble). Cells were then starved in serum-free medium and treated with 100 ng/mL IGF1 for 15 minutes. Immunoblots were performed with antibodies as indicated. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. veh, vehicle. To investigate the corresponding signaling changes, we studied IGF1-induced phosphorylation of downstream targets in the absence and presence of IRS1 (Fig. 4B). The immunoblotting of scramble controls correlated strongly with the concentration and time-gradient stimulation results in Fig. 3. As a positive control, knockdown of IGF1R markedly reduced IGF1-induced activation of IRS1, Akt, and ERK1/2. Intriguingly, knockdown of IRS1 resulted in diminished IGF1R phosphorylation in all groups, indicating a yet-to-be-identified feedback mechanism. Consequently, the downstream activation of Akt (both S473 and T308) and ERK1/2 (T202/T204) were blocked, albeit the effect on p-ERK1/2 was weaker in MCF-7 normalized to scramble control groups. The enhanced phosphorylation of AKT (T308) was also diminished in MCF-7 cell line after IRS1 knockdown. Combination of anti-IGF1R (OSI-906) and anti-ER (fulvestrant) exhibits stronger synergism in ESR1 mutant cells Finally, we asked whether activation of the IGF1 pathway altered ESR1 mutant cells response to small-molecule inhibitors targeting IGF1R. First, we examined the effect of BMS-754807 and OSI-906 on two-dimensional growth of ESR1 mutant cells (Supplemental Fig. 6). MCF-7 cells were more sensitive to IGF1R inhibition compared with T47D. Both MCF-7 Y537S and D538G mutant cells exhibited greater half maximal inhibitory concentration toward both IGF1R inhibitors, with Y537S showing a stronger resistance. The similar resistance trend was also detected in response of T47D cells to BMS-754807; however, no obvious differences were detected with OSI-906. Combination therapies cotargeting IGF1R and ERα in breast cancer have been developed and evaluated (4). With respect to this therapeutic strategy, we examined the combined effect of OSI-906 and fulvestrant (Fig. 5A). Therefore, we treated WT and ESR1 mutant T47D and MCF-7 cells with increasing concentrations of fulvestrant (1 nM to 100 nM) and OSI-906 (0.1 μM to 10 μM). Effects on growth rates are shown in Fig. 5A, and combination index values are shown in Fig. 5B. These data show that under EC50s, OSI-906 and fulvestrant display synergistic effects [combination index (CI) <1] in both mutant cell line models in both MCF-7 and T47D, with the exception of T47D-Y537S cells that showed no effect (CI = 1). In contrast, antagonistic effects (CI > 1) were observed in both WT cell line models. At 75% and 90% effective dose, the drugs displayed synergistic activities in all mutant cell line models, with consistently stronger effects compared with WT cells. Figure 5. View largeDownload slide Combination treatment with IGF1R inhibitors and fulvestrant exhibits synergistic effects in ESR1 mutant cells. (A) Hormone-deprived MCF-7 and T47D clones were pooled and treated with the corresponding combined concentrations of OSI-906 and fulvestrant in the presence of 20 pM of E2. Cell numbers were quantified with FluoReporter kit (Life Technologies) after 6 (MCF-7) and 7 (T47D) days. Percentages of growth were calculated by normalizing to the averages of the vehicle control groups. The data were calculated by combining two independent experiments. (B) Combination index mean values at 50% effective dose (ED50), 75% effective dose (ED75), and 90% effective dose (ED90) were calculated based on two independent experiments. Figure 5. View largeDownload slide Combination treatment with IGF1R inhibitors and fulvestrant exhibits synergistic effects in ESR1 mutant cells. (A) Hormone-deprived MCF-7 and T47D clones were pooled and treated with the corresponding combined concentrations of OSI-906 and fulvestrant in the presence of 20 pM of E2. Cell numbers were quantified with FluoReporter kit (Life Technologies) after 6 (MCF-7) and 7 (T47D) days. Percentages of growth were calculated by normalizing to the averages of the vehicle control groups. The data were calculated by combining two independent experiments. (B) Combination index mean values at 50% effective dose (ED50), 75% effective dose (ED75), and 90% effective dose (ED90) were calculated based on two independent experiments. Discussion Recent data indicate that mutations in the ligand-binding domain of ESR1 confer constitutive ligand-independent gene expression and that this may lead to endocrine resistance and poor survival (17, 18). Using two breast cancer cell lines that are genome-edited with two of the most common ESR1 mutations, we have characterized augmented IGF1 response in these cell lines, which is partially attributed to the upregulation of IRS1. Elevated levels of IRS1 in ESR1 mutant cells increased IGF1 potency through the PI3K-Akt axis and triggered enhanced IGF1-induced growth (Fig. 6). Of note, employment of an IGF1R inhibitor (OSI-906) in combination with a selective ER downregulator (fulvestrant) showed synergistic growth inhibitory effects in ESR1 mutant cells, yet showed antagonism in WT MCF-7 cells. This study sheds light on a putative therapeutic strategy for ESR1 mutant breast cancers. Figure 6. View largeDownload slide Schematic view of enhanced IGF1 response in ESR1 mutant cells. In ESR1 WT cells, IGF1 activates IGF1R, which recruits either IRS1 or Shc to increase PI3K-Akt and/or mitogen-activated protein kinase (MAPK) pathways. With the acquisition of Y537S or D538G mutations, mutant ER binds to IRS1 promoter and increases IRS1 expression levels. Upregulation of IRS1 further amplifies the IGF1 response via the PI3K-Akt axis and thus enhances cell proliferation. Figure 6. View largeDownload slide Schematic view of enhanced IGF1 response in ESR1 mutant cells. In ESR1 WT cells, IGF1 activates IGF1R, which recruits either IRS1 or Shc to increase PI3K-Akt and/or mitogen-activated protein kinase (MAPK) pathways. With the acquisition of Y537S or D538G mutations, mutant ER binds to IRS1 promoter and increases IRS1 expression levels. Upregulation of IRS1 further amplifies the IGF1 response via the PI3K-Akt axis and thus enhances cell proliferation. Initial reports from Gelsomino et al. (8) identified enhanced IGF1R-ERα crosstalk in breast cancer cells transfected with mutant ESR1, with a focus on the higher basal IGF1R activation levels in ESR1 mutant cells. We find a similar increase in IGF1R-ERα crosstalk in genome-edited cells, corroborating the study of Gelsomino et al. (8); however, there are several important differences. Gelsomino et al. (8) found that IGF1R and IRS1 levels were unaltered in ESR1-transfected cells, despite these being estrogen-regulated genes. In contrast, we found elevated IGF1R and IRS1. One potential reason for these discrepancies is the difference in cell model construction, perhaps mainly the CRISPR vs transfection, with CRISPR heterozygous mutant clones more closely mimicking the genetic alteration in human breast cancers (19, 20). In addition, our findings were shown in two cell lines with two distinct high-frequent mutations allowing us to identify different context-dependent effects (e.g., greater IGF sensitization MCF-7 mutant ESR1 clones) (21, 22). This pattern matched the observation of IGF1R upregulation in MCF-7 mutant cells according to RNA-seq, which in part increases ligand activation and downstream signaling. According to our previous RNA-seq analysis, although estrogen-independent gene expression patterns were detected in all cell models, genes overlapping between Y537S and D538G mutant cells were only 20.1% and 43.5% of all of the constitutively expressed genes in T47D and MCF-7, respectively, suggesting mutation-specific ligand-independent genes and their consequential phenotypes. For instance, it is interesting to observe the enhanced bFGF response in T47D-Y537S and EGF response in Y537S mutant exclusively of both cell lines (Fig. 2C). It could be presumably explained by either the upregulation of signaling downstream kinases (PRKCA) or transcriptional factors (STAT5A and MYC), particularly under these signaling receptors. We also discerned higher basal phosphorylation levels in D538G but limited levels in Y537S mutants of MCF-7 cell line. This may also be explained by either the exclusively downregulation of TIMP3 (a matrix metalloproteinase inhibitor) or upregulation of GAB2 (a signaling adaptor) in the MCF7-D538G line, which potentially enhances the AKT basal levels through integrin or ER crosstalk. Intriguingly, we found constitutive phosphorylation of S473 on AKT in ESR1 mutant clones. Akt sites T308 and S473 sites are phosphorylated by PDK1 and mTORC2, respectively (23). Previous studies have identified site-specific regulatory mechanisms of Akt phosphorylation and downstream signaling (23). The Ser473 site phosphorylation is reported to be exclusively regulated by integrin-linked kinase, protein kinase Cα, and other receptor tyrosine kinase activation (4). It is possible that ESR1 mutant cells have these pathways activated, which increase basal pS473-Akt. Moreover, the mutant or cell line–specific upregulations of kinase components genes (DEPTOR) (24) or signaling adaptor genes (GAB2) (25) will potentially enhances the basal pS473-Akt. Additionally, our previous study (9) reveals that D538G mutation of T47D cell line exhibits relatively lower ligand-independent growth rate but stronger E2 response after hormone deprivation, which indicates a weaker constitutive ligand-independent activation gene pattern in this particular line, including the group IGF1 pathway genes shown in Fig. 1B. Consequently, the IGF1-induced proliferation was consistently weaker in T47D-D538G cell line, compared with other ESR1 mutation cell lines (Fig. 2A). Furthermore, as a ligand of insulin receptor (IR), IGF1 has the capacity to activate IR (26, 27). Interestingly, IR consistently decreased in all the mutant lines at both RNA (Supplemental Fig. 1A) and protein levels (Supplemental Fig. 1B), implicating that stronger growth response dominantly depends on IGF1R activation in ESR1 mutant cells. Similarly, the inhibitory effects of OSI-906 also mainly rely on the block of IGF1 activity rather than IR, even IR serves as one of the targets of OSI-906 (28). The equivalent IR isoform ratios (IR-A/IR-B) between different mutants of each cell line (Supplemental Fig. 1C) rules out the possibility that the altered IR isoform ratios leads to differentiated IGF-II response (Fig. 2C). We also examined the ER binding patterns on the IRS1 promoter in our cell models via chromatin immunoprecipitation quantitative polymerase chain reaction, with the primers designed based on the two reported putative binding sites (29) (Supplemental Fig. 7A). However, only MCF7-Y537S exhibits ligand-independent ER recruitment (Supplemental Fig. 7B and 7C). It is possible that the ER binding is present as a cell line–specific property or other potential ER-related indirect binding exists in T47D cell line. Finally, our data indicate that the enhanced IGF1 response is exclusive to ESR1 mutant cells and not found in TamR and LTED cells, indicating that it is not a general endocrine resistance mechanism. Consistent with this, Fagan et al. (16) reported the loss of IGF1R expression and IGF1 response in TamR cell models. These phenotypic distinctions further stress the need to dissect out the specific mechanisms of resistance and suggest that precise treatment of endocrine resistance may require the monitoring of ESR1 mutations, in which IGF1R inhibitors may have effect. To this end, further analysis examining gene profiles and identifying novel pathways, together with clinical archived materials, is warranted. Appendix. Antibody Table Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Monoclonal or Polyclonal  Dilution Used  RRID  p-IGF1R-Y1135    Phospho-IGF-I Receptor β (Tyr1135) (DA7A8) Rabbit mAb  Cell Signaling Technology, 3918  Monoclonal  1:200  AB_10548764  IGF1R    IGF-I Receptor β (D23H3) XP® Rabbit mAb  Cell Signaling Technology, 9750  Monoclonal  1:1000  AB_10950969  p-IRS1-S320    Phospho-IRS-1 (Ser302) Antibody  Cell Signaling Technology, 2384  Polyclonal  1:500  AB_330360  p-AKT-T308    Phospho-Akt (Thr308) (D25E6) XP® Rabbit mAb  Cell Signaling Technology, 13038  Monoclonal  1:1000  AB_2629447  p-AKT-S473    Phospho-Akt (Ser473) (D9E) XP® Rabbit mAb  Cell Signaling Technology, 4060  Monoclonal  1:1000  AB_2315049  AKT    Akt (pan) (40D4) Mouse mAb  Cell Signaling Technology, 2920  Monoclonal  1:1000  AB_1147620  p-ERK1/2-T202/T204    Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (197G2) Rabbit mAb  Cell Signaling Technology, 4377  Monoclonal  1:1000  AB_331775  ERK    p44/42 MAPK (Erk1/2) Antibody  Cell Signaling Technology, 9102  Polyclonal  1:1000  AB_330744  IRS1    IRS1  Santa Cruz Biotechnology, sc7200  Polyclonal  1:1000  AB_649615      Insulin Receptor β  Cell Signaling Technology, 3025  Monoclonal  1:1000  AB_2280448  β-actin    Monoclonal Anti-Actin antibody produced in mouse  Thermo Fisher Scientific, A3853  Monoclonal  1:10,000  AB_262137  Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Monoclonal or Polyclonal  Dilution Used  RRID  p-IGF1R-Y1135    Phospho-IGF-I Receptor β (Tyr1135) (DA7A8) Rabbit mAb  Cell Signaling Technology, 3918  Monoclonal  1:200  AB_10548764  IGF1R    IGF-I Receptor β (D23H3) XP® Rabbit mAb  Cell Signaling Technology, 9750  Monoclonal  1:1000  AB_10950969  p-IRS1-S320    Phospho-IRS-1 (Ser302) Antibody  Cell Signaling Technology, 2384  Polyclonal  1:500  AB_330360  p-AKT-T308    Phospho-Akt (Thr308) (D25E6) XP® Rabbit mAb  Cell Signaling Technology, 13038  Monoclonal  1:1000  AB_2629447  p-AKT-S473    Phospho-Akt (Ser473) (D9E) XP® Rabbit mAb  Cell Signaling Technology, 4060  Monoclonal  1:1000  AB_2315049  AKT    Akt (pan) (40D4) Mouse mAb  Cell Signaling Technology, 2920  Monoclonal  1:1000  AB_1147620  p-ERK1/2-T202/T204    Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (197G2) Rabbit mAb  Cell Signaling Technology, 4377  Monoclonal  1:1000  AB_331775  ERK    p44/42 MAPK (Erk1/2) Antibody  Cell Signaling Technology, 9102  Polyclonal  1:1000  AB_330744  IRS1    IRS1  Santa Cruz Biotechnology, sc7200  Polyclonal  1:1000  AB_649615      Insulin Receptor β  Cell Signaling Technology, 3025  Monoclonal  1:1000  AB_2280448  β-actin    Monoclonal Anti-Actin antibody produced in mouse  Thermo Fisher Scientific, A3853  Monoclonal  1:10,000  AB_262137  Abbreviations: mAb, monoclonal antibody; MAPK, mitogen-activated protein kinase; RRID, Research Resource Identifier. View Large Abbreviations: bFGF fibroblast growth factor-basic CI combination index CSS charcoal stripped serum E2 17β-estradiol EC50 half maximal effective concentration EGF epidermal growth factor ER estrogen receptor IGF insulinlike growth factor IGF1 insulinlike growth factor-1 IGF1R insulinlike growth factor-1 receptor IRS1 insulin receptor substrate-1 LTED long-term estrogen deprivation MEK mitogen-activated protein kinase kinase mTOR mammalian target of rapamycin p- phosphorylated PDGF platelet-derived growth factor PI3K phosphoinositide 3-kinase qRT-PCR quantitative reverse transcription polymerase chain reaction RNA-seq RNA-sequencing siRNA small interfering RNA TamR tamoxifen-resistant VEGF vascular endothelial growth factor WT wild-type. Acknowledgments The authors thank Dr. Rachel Schiff for the gift of MCF-7 and ZR75-1 TamR and LTED cell models, the assistants from all Lee-Oesterreich laboratories, and all of the annotated grant sources. Financial Support: This study was in part supported by funds from the Nicole Meloche Memorial Breast Cancer Fund, the Pittsburgh Penguins Foundation, the Penguins Alumni Association, the Mario Lemieux Foundation, Fashion Footwear of New York, Breast Cancer Research Foundation (to A.V.L., S.O., and B.H.P.), and Avon Foundation for Women (to B.H.P.) through a Pilot Award from the Institute for Precision Medicine at the University of Pittsburgh (to S.O.) and through Grant P30CA047904 from the National Institutes of Health/National Cancer Institute. S.O. and A.V.L. are recipients of Scientific Leadership awards from Susan G. Komen. P.W. was supported by a China Scholarship Council award through Tsinghua Medical School (Beijing, China). Z.L. was supported by the John S. Lazo Cancer Pharmacology Fellowship. K.M.L. was supported by National Institutes of Health F30 Grant CA203154. Disclosure Summary: The authors have nothing to disclose. References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin . 2016; 66( 1): 7– 30. Google Scholar CrossRef Search ADS PubMed  2. Hayes EL, Lewis-Wambi JS. Mechanisms of endocrine resistance in breast cancer: an overview of the proposed roles of noncoding RNA. Breast Cancer Res . 2015; 17( 1): 40. Google Scholar CrossRef Search ADS PubMed  3. Clarke R, Tyson JJ, Dixon JM. Endocrine resistance in breast cancer--An overview and update. Mol Cell Endocrinol . 2015; 418( Pt 3): 220– 234. Google Scholar CrossRef Search ADS PubMed  4. Toy W, Shen Y, Won H, Green B, Sakr RA, Will M, Li Z, Gala K, Fanning S, King TA, Hudis C, Chen D, Taran T, Hortobagyi G, Greene G, Berger M, Baselga J, Chandarlapaty S. ESR1 ligand-binding domain mutations in hormone-resistant breast cancer. Nat Genet . 2013; 45( 12): 1439– 1445. Google Scholar CrossRef Search ADS PubMed  5. Fanning SW, Mayne CG, Dharmarajan V, Carlson KE, Martin TA, Novick SJ, Toy W, Green B, Panchamukhi S, Katzenellenbogen BS, Tajkhorshid E, Griffin PR, Shen Y, Chandarlapaty S, Katzenellenbogen JA, Greene GL. Estrogen receptor alpha somatic mutations Y537S and D538G confer breast cancer endocrine resistance by stabilizing the activating function-2 binding conformation. eLife . 2016; 5: e12792. Google Scholar CrossRef Search ADS PubMed  6. Harrod A, Fulton J, Nguyen VT, et al.  . Genomic modelling of the ESR1 Y537S mutation for evaluating function and new therapeutic approaches for metastatic breast cancer. Oncogene . 2017; 36: 2286– 2296. Google Scholar CrossRef Search ADS PubMed  7. Mao C, Livezey M, Kim JE, Shapiro DJ. Antiestrogen Resistant Cell Lines Expressing Estrogen Receptor α Mutations Upregulate the Unfolded Protein Response and are Killed by BHPI. Sci Rep . 2016; 6: 34753. Google Scholar CrossRef Search ADS PubMed  8. Gelsomino L, Gu G, Rechoum Y, Beyer AR, Pejerrey SM, Tsimelzon A, Wang T, Huffman K, Ludlow A, Andò S, Fuqua SAW. ESR1 mutations affect anti-proliferative responses to tamoxifen through enhanced cross-talk with IGF signaling [published correction appears in Breast Cancer Res Treat. 2017;163:639–640]. Breast Cancer Res Treat . 2016; 157( 2): 253– 265. Google Scholar CrossRef Search ADS PubMed  9. Bahreini A, Li Z, Wang P, Levine KM, Tasdemir N, Cao L, Weir HM, Puhalla SL, Davidson NE, Stern AM, Chu D, Park BH, Lee AV, Oesterreich S. Mutation site and context dependent effects of ESR1 mutation in genome-edited breast cancer cell models. Breast Cancer Res . 2017; 19( 1): 60. Google Scholar CrossRef Search ADS PubMed  10. Creighton CJ, Casa A, Lazard Z, Huang S, Tsimelzon A, Hilsenbeck SG, Osborne CK, Lee AV. Insulin-like growth factor-I activates gene transcription programs strongly associated with poor breast cancer prognosis. J Clin Oncol . 2008; 26( 25): 4078– 4085. Google Scholar CrossRef Search ADS PubMed  11. Mauro L, Salerno M, Panno ML, Bellizzi D, Sisci D, Miglietta A, Surmacz E, Andò S. Estradiol increases IRS-1 gene expression and insulin signaling in breast cancer cells. Biochem Biophys Res Commun . 2001; 288( 3): 685– 689. Google Scholar CrossRef Search ADS PubMed  12. Molloy CA, May FE, Westley BR. Insulin receptor substrate-1 expression is regulated by estrogen in the MCF-7 human breast cancer cell line. J Biol Chem . 2000; 275( 17): 12565– 12571. Google Scholar CrossRef Search ADS PubMed  13. Cicenas J. The potential role of Akt phosphorylation in human cancers. Int J Biol Markers . 2008; 23( 1): 1– 9. Google Scholar CrossRef Search ADS PubMed  14. Robey RB, Hay N. Is Akt the “Warburg kinase”?—Akt-energy metabolism interactions and oncogenesis. Semin Cancer Biol . 2009; 19: 25– 31. 15. Parisot JP, Hu XF, DeLuise M, Zalcberg JR. Altered expression of the IGF-1 receptor in a tamoxifen-resistant human breast cancer cell line. Br J Cancer . 1999; 79( 5-6): 693– 700. Google Scholar CrossRef Search ADS PubMed  16. Fagan DH, Uselman RR, Sachdev D, Yee D. Acquired resistance to tamoxifen is associated with loss of the type I insulin-like growth factor receptor: implications for breast cancer treatment. Cancer Res . 2012; 72( 13): 3372– 3380. Google Scholar CrossRef Search ADS PubMed  17. Jeselsohn R, Yelensky R, Buchwalter G, Frampton G, Meric-Bernstam F, Gonzalez-Angulo AM, Ferrer-Lozano J, Perez-Fidalgo JA, Cristofanilli M, Gómez H, Arteaga CL, Giltnane J, Balko JM, Cronin MT, Jarosz M, Sun J, Hawryluk M, Lipson D, Otto G, Ross JS, Dvir A, Soussan-Gutman L, Wolf I, Rubinek T, Gilmore L, Schnitt S, Come SE, Pusztai L, Stephens P, Brown M, Miller VA. Emergence of constitutively active estrogen receptor-α mutations in pretreated advanced estrogen receptor-positive breast cancer. Clin Cancer Res . 2014; 20( 7): 1757– 1767. Google Scholar CrossRef Search ADS PubMed  18. Jeselsohn R, Buchwalter G, De Angelis C, Brown M, Schiff R. ESR1 mutations—a mechanism for acquired endocrine resistance in breast cancer. Nat Rev Clin Oncol . 2015; 12( 10): 573– 583. Google Scholar CrossRef Search ADS PubMed  19. Kotin RM, Siniscalco M, Samulski RJ, Zhu XD, Hunter L, Laughlin CA, McLaughlin S, Muzyczka N, Rocchi M, Berns KI. Site-specific integration by adeno-associated virus. Proc Natl Acad Sci USA . 1990; 87( 6): 2211– 2215. Google Scholar CrossRef Search ADS PubMed  20. Sander JD, Joung JK. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat Biotechnol . 2014; 32( 4): 347– 355. Google Scholar CrossRef Search ADS PubMed  21. Ross RL, Askham JM, Knowles MA. PIK3CA mutation spectrum in urothelial carcinoma reflects cell context-dependent signaling and phenotypic outputs. Oncogene . 2013; 32( 6): 768– 776. Google Scholar CrossRef Search ADS PubMed  22. Spoerke JM, Gendreau S, Walter K, Qiu J, Wilson TR, Savage H, Aimi J, Derynck MK, Chen M, Chan IT, Amler LC, Hampton GM, Johnston S, Krop I, Schmid P, Lackner MR. Heterogeneity and clinical significance of ESR1 mutations in ER-positive metastatic breast cancer patients receiving fulvestrant. Nat Commun . 2016; 7: 11579. Google Scholar CrossRef Search ADS PubMed  23. Manning BD, Cantley LC. AKT/PKB signaling: navigating downstream. Cell . 2007; 129( 7): 1261– 1274. Google Scholar CrossRef Search ADS PubMed  24. Peterson TR, Laplante M, Thoreen CC, Sancak Y, Kang SA, Kuehl WM, Gray NS, Sabatini DM. DEPTOR is an mTOR inhibitor frequently overexpressed in multiple myeloma cells and required for their survival. Cell . 2009; 137( 5): 873– 886. Google Scholar CrossRef Search ADS PubMed  25. Hunzicker-Dunn ME, Lopez-Biladeau B, Law NC, Fiedler SE, Carr DW, Maizels ET. PKA and GAB2 play central roles in the FSH signaling pathway to PI3K and AKT in ovarian granulosa cells. Proc Natl Acad Sci USA . 2012; 109( 44): E2979– E2988. Google Scholar CrossRef Search ADS PubMed  26. Baserga R, Peruzzi F, Reiss K. The IGF-1 receptor in cancer biology. Int J Cancer . 2003; 107( 6): 873– 877. Google Scholar CrossRef Search ADS PubMed  27. Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P, Hemmings BA. Mechanism of activation of protein kinase B by insulin and IGF-1. EMBO J . 1996; 15( 23): 6541– 6551. Google Scholar PubMed  28. Mulvihill MJ, Cooke A, Rosenfeld-Franklin M, Buck E, Foreman K, Landfair D, O’Connor M, Pirritt C, Sun Y, Yao Y, Arnold LD, Gibson NW, Ji QS. Discovery of OSI-906: a selective and orally efficacious dual inhibitor of the IGF-1 receptor and insulin receptor. Future Med Chem . 2009; 1( 6): 1153– 1171. Google Scholar CrossRef Search ADS PubMed  29. Ross-Innes CS, Stark R, Teschendorff AE, Holmes KA, Ali HR, Dunning MJ, Brown GD, Gojis O, Ellis IO, Green AR, Ali S, Chin SF, Palmieri C, Caldas C, Carroll JS. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature . 2012; 481( 7381): 389– 393. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Endocrinology Oxford University Press

Upregulation of IRS1 Enhances IGF1 Response in Y537S and D538G ESR1 Mutant Breast Cancer Cells

Loading next page...
 
/lp/ou_press/upregulation-of-irs1-enhances-igf1-response-in-y537s-and-d538g-esr1-JzXm8n8DzZ
Publisher
Endocrine Society
Copyright
Copyright © 2018 Endocrine Society
ISSN
0013-7227
eISSN
1945-7170
D.O.I.
10.1210/en.2017-00693
Publisher site
See Article on Publisher Site

Abstract

Abstract Increased evidence suggests that somatic mutations in the ligand-binding domain of estrogen receptor [ER (ERα/ESR1)] are critical mediators of endocrine-resistant breast cancer progression. Insulinlike growth factor-1 (IGF1) is an essential regulator of breast development and tumorigenesis and also has a role in endocrine resistance. A recent study showed enhanced crosstalk between IGF1 and ERα in ESR1 mutant cells, but detailed mechanisms are incompletely understood. Using genome-edited MCF-7 and T47D cell lines harboring Y537S and D538G ESR1 mutations, we characterized altered IGF1 signaling. RNA sequencing revealed upregulation of multiple genes in the IGF1 pathway, including insulin receptor substrate-1 (IRS1), consistent in both Y537S and D538G ESR1 mutant cell line models. Higher IRS1 expression was confirmed by quantitative reverse transcription polymerase chain reaction and immunoblotting. ESR1 mutant cells also showed increased levels of IGF-regulated genes, reflected by activation of an IGF signature. IGF1 showed increased sensitivity and potency in growth stimulation of ESR1 mutant cells. Analysis of downstream signaling revealed the phosphoinositide 3-kinase (PI3K)–Akt axis as a major pathway mediating the enhanced IGF1 response in ESR1 mutant cells. Decreasing IRS1 expression by small interfering RNA diminished the increased sensitivity to IGF1. Combination treatment with inhibitors against IGF1 receptor (IGF1R; OSI-906) and ER (fulvestrant) showed synergistic growth inhibition in ESR1 mutant cells, particularly at lower effective concentrations. Our study supports a critical role of enhanced IGF1 signaling in ESR1 mutant cell lines, pointing toward a potential for cotargeting IGF1R and ERα in endocrine-resistant breast tumors with mutant ESR1. Estrogen deprivation has become a prevalent therapeutic strategy for >70% of breast cancers that harbor estrogen receptor (ER)α expression (1). Despite improved patient outcomes, acquired endocrine resistance develops in a subset of ERα-positive breast cancers via various molecular mechanisms (2, 3). Recent studies have documented a group of recurrent hotspot mutations in ESR1, Y537S and D538G being the most frequent, that cluster in the ligand-binding domain and overlap with activation function 2 (4). Molecular modeling indicates that the hotspot mutations at positions Y537 and D538 alter the position of helix 12 to favor an agonist conformation of the receptor (5). Recent mechanistic studies performed by us and others, employing either transfected or genome-edited ESR1 mutant cells, show that mutant ESR1 has ligand-independent activity and diminished sensitivity toward antiestrogen drugs (4). Clinical studies have documented rare ESR1 mutations in primary breast cancers, but increased frequency in metastatic lesions and circulation-free DNA, suggesting a potential role of acquired ESR1 mutations in facilitating metastasis (4). Given studies suggesting poor outcomes in patients harboring breast cancers with ESR1 mutations (4), identification of new targets and design of novel therapeutic strategies have gained urgency. Multiple recent preclinical studies have uncovered promising therapeutic targets in breast cancer cells harboring ESR1 mutations. Harrod et al. (6) highlighted the potential utility of a CDK7 inhibitor to block growth in MCF-7 cells with CRISPR-edited Y537S. Mao et al. (7) found increased unfolded protein response in CRIPSR-edited Y537S and D538G ESR1 mutant cells. Recent findings from Gelsomino et al. (8) identified enhanced crosstalk between mutant ERα and insulinlike growth factor-1 receptor (IGF1R), proposing a role in tamoxifen resistance, indicating a potential for combination therapy by cotargeting ERα and mammalian target of rapamycin (mTOR) in ESR1 mutant tumors. Our recent transcriptomic analysis of genome-edited MCF-7 and T47D Y537S and D538G ESR1 cell lines revealed mutation site- and context-dependent gene expression changes compared with wild-type (WT) ESR1 (9). Similar to the findings by Gelsomino et al. (8), RNA-sequencing (RNA-seq) analysis revealed altered expression of insulinlike growth factor-1 (IGF1) pathway members in both ESR1 mutant cell lines. In this study, we performed a preclinical study to characterize the mechanisms underlying the augmented IGF1 response in ESR1 mutant cells and evaluated the strategy of cotargeting ER and IGF1R for future therapeutic development. Materials and Methods Cell culture T47D and MCF-7 cells were obtained from American Type Culture Collection. Both cell lines were authenticated at the University of Arizona Genetics Core. T47D and MCF-7 cells were maintained in RPMI 1640 plus 10% fetal bovine serum and Dulbecco’s modified Eagle medium plus 10% fetal bovine serum, respectively. Tamoxifen-resistant (TamR) and long-term estrogen deprivation (LTED) MCF-7 and ZR75-1 cell lines were gifts from Dr. Rachel Schiff (Baylor College of Medicine, Houston, TX). For hormone treatment experiments, cells were deprived of steroid hormones by placement in phenol red–free Iscove’s modified Eagle medium with 10% and 5% charcoal stripped serum (CSS) for T47D and MCF-7, respectively. CSS was purchased from Gibco (catalog no. 12676; Waltham, MA). 17β-Estradiol (E2) and fulvestrant were obtained from Sigma-Aldrich (catalog no. E8875; St. Louis, MO) and Tocris Bioscience (catalog no. 1047; Bristol, United Kingdom), respectively. IGF1 was purchased from GroPep Bioreagents (catalog no. AM001; Tbarton, South Australia, Australia). IGF-II (catalog no. 110-12), vascular endothelial growth factor (VEGF; catalog no. 100-20), platelet-derived growth factor (PDGF)–BB (catalog no. 100-14B), and fibroblast growth factor-basic (bFGF; catalog no. 100-18B) were from PeproTech (Rocky Hill, NJ). OSI-906 (catalog no. S1091), wortmannin (catalog no. S2758), BX-795 (catalog no. S1274), and BMS-754807 (catalog no. S1124) were obtained from Selleck Chemicals (Houston, TX). U0126 (catalog no. 1144) and rapamycin (catalog no. 1292) were purchased from Tocris Bioscience. Immunoblotting After 3 days in CSS, 120,000 (MCF-7) and 90,000 cells (T47D) were plated in six-well plates. For IGF1 stimulation, cells were starved in serum-free Iscove’s modified Eagle medium for 24 hours and then treated with IGF1 for 15 minutes. For small interfering RNA (siRNA) knockdown experiments, cells were reverse transfected with siRNA-targeting insulin receptor substrate-1 (IRS1; catalog no. L-003015; Dharmacon, Lafayette, CO) and/or IGF1R (catalog no. L-003012; Dharmacon) for 24 hours and subjected to IGF1 stimulation. The cells were lysed with RIPA buffer, sonicated, 60 μg of protein separated using a sodium dodecyl sulfate polyacrylamide gel electrophoresis gel, and then transferred onto polyvinylidene difluoride membrane. Membranes were immunostained using the following antibodies: phosphorylated (p-)IGF1R-Y1135 (catalog no. 3918), IGF1R (catalog no.9750), p-IRS1–S320 (catalog no. 2384), p-AKT–T308 (catalog no. 13038), p-AKT–S473 (catalog no. 4060), AKT (catalog no. 2920), p-ERK1/2–T202/T204 (catalog no. 4377), ERK (catalog no. 9102), and IR (catalog no. 3025), all purchased from Cell Signaling Technology (Danvers, MA). IRS1 (catalog no. sc-7200) and β-actin (catalog no. A3853) antibodies were obtained from Santa Cruz Biotechnology (Dallas, TX) and Sigma-Aldrich, respectively. Quantifications were performed using ImageJ software (National Institutes of Health). Immunoblots were repeated twice unless otherwise stated. RNA analysis using quantitative reverse transcription polymerase chain reaction Posthormone-deprived cells were seeded into six-well plates and treated with 1 nM E2, 1 μM ICI 182,780, or their combination. RNA samples were extracted after 24 hours and complementary DNA synthesized using an iScript kit (catalog no. 1708890; Bio-Rad, Hercules, CA). Quantitative reverse transcription polymerase chain reaction (qRT-PCR) reactions were performed with SYBR Green Supermix (catalog no. 1726275; Bio-Rad), the ΔΔ threshold cycle method was used to analyze relative messenger RNA fold changes, and RPLP0 levels were measured as the internal control. Primers sequences are as follows: RPLP0 (forward), 5′-TAAACCCTGCGTGGCAATC-3′; RPLP0 (reverse), 5′-TTGTCTGCTCCCACAATGAAA-3′; ESR1 (forward), 5′-GAGTATGATCCTACCAGACCCTTC-3′; ESR1 (reverse), 5′-CCTGATCATGGAGGGTCAAATC-3′; IRS1 (forward), 5′-TCTGCTCAGCGTTGGTG-3′; and IRS1 (reverse), 5′-GTGCATGCTCTTGGGTTTG-3′. Growth assays As previously described (8), individual MCF-7 or T47D CRISPR-edited clones were evenly pooled after 3 days of hormone deprivation in CSS and plated into 96-well plates using 2500 cells/well (MCF-7) or 4000 cells/well (T47D). After 24 hours, the cells were treated with various concentrations of growth factors or inhibitors, as well as vehicle control. For time-course growth assay, the cells were harvested after 0, 2, 4, 6, and 9 days and quantified with the FluoReporter kit (catalog no. F2962; Life Technologies, Carlsbad, CA) following the manufacturer’s protocol, and half maximal inhibitory concentrations or half maximal effective concentrations (EC50s) were calculated using the Prism statistical package (GraphPad Software, La Jolla, CA). For analysis of drug synergy, the combination index values were calculated using the CalcuSync package (Biosoft, Great Shelford, United Kingdom). All experiments were performed with six biological replicates. Calculation of IGF activation score The IGF activation score was calculated for each cell line as described previously (10). Briefly, the activation score was calculated as the Pearson correlation between the IGF gene signature pattern (by using 1 and −1 for up and down, respectively) and the gene expression values of each cell line. The gene expression values were the log2 (transcripts per million + 1) of Salmon (v.0.6.0) mapped reads to Ensembl gene-level identification numbers from Bahreini et al. (9). The microarray probes from the IGF upregulated and downregulated genes were mapped to Ensembl identification numbers using the hgu133a2.db annotation package [hgu133a2.db: Affymetrix Human Genome U133A 2.0 Array annotation data (chip hgu133a2); Affymetrix, Santa Clara, CA]. This was performed using R version 3.2.2, and the R code is provided as a supplemental text file. Statistical analysis Each experiment was performed with corresponding biological and technical replicates and repeated, as indicated. Multiple statistical tests were used to assess the statistical significance depending on the design of the experiments, and the P value was calculated accordingly (*P < 0.05; **P < 0.01). Results Y537S and D538G mutant cells exhibit upregulation of IRS1 We have previously shown a notable similarity between genes for which expression is altered by E2 or IGF1 (10). In this study, we investigated levels of these same IGF1-stimulated genes in ESR1 mutant cells by performing a correlation analysis between the RNA-seq data sets of ESR1 mutant cells and the previously reported IGF1 signature panel of 976 transcripts (814 genes) (10). Replicates from each group of RNA-seq results were compared with the IGF1 signature individually, and Pearson correlation coefficient values were calculated to indicate the similarities (Fig. 1A). In the absence of E2, gene expression in both MCF-7 and T47D ESR1 mutant cells showed a significantly increased similarity to the IGF gene signature pattern compared with WT groups. Consistent with previous results (10), E2 stimulation led to a higher similarity to the IGF gene signature in all cell line models. We then directly examined genes in the IGF1 signaling pathway (n = 39; Supplemental Table 1) that showed a fold change >1.5 (P < 0.005) in ESR1 mutant cells (Fig. 1B). We found that IGF2, IGFBP6, IRS1, IGFBP4, and INSR were consistently altered in both ESR1 mutant cell line models compared with WT cells. IRS1, which encodes an essential adaptor in the IGF1 pathway, was the most consistently upregulated gene in both mutations and cell lines in the entire panel. IGF1R levels were also induced in MCF-7 ESR1 mutant cells. Interestingly, the INSR gene exhibited decreased levels in both cell lines (Supplemental Fig. 1A and 1B), perhaps reflecting a secondary negative-feedback regulation following increased IGF1R/IRS1 signaling. Moreover, the transcript reads from RNA-seq suggest no difference in IR isoform ratios (IR-A/IR-B) between different mutants of each cell line (Supplemental Fig. 1C). Upregulation of IRS1 was validated using immunoblotting (Fig. 1C) and qRT-PCR (Fig. 1D). Increased protein levels of IRS1 were observed in both ESR1 mutant cell lines under hormone-deprived conditions, whereas higher IGF1R expression was only seen in MCF-7 mutant cells, a result that correlated well with the RNA-seq data. The qRT-PCR results also demonstrated higher basal level of IRS1 in ESR1 mutant cell lines, with the exception of T47D-D538G clones in which the increase was not noticeable. Similar data were observed when measuring IRS1 messenger RNA in individual clones that constitute the pool (Supplemental Fig. 2A). qRT-PCR results indicated that IRS1 levels were induced by E2 and inhibited by fulvestrant treatment, which is consistent with previous studies characterizing IRS1 as an ER-induced gene (11, 12). Decreasing ESR1 expression with siRNA attenuated the higher IRS1 levels in ESR1 mutant cells, confirming the dependency of the IRS1 upregulation on the genome-edited mutant ESR1 (Supplemental Fig. 2B). Figure 1. View largeDownload slide MCF-7 and T47D cells with CRISPR-edited ESR1 Y537S and D538G mutations exhibit upregulation of IGF signaling components and downstream activity. (A) RNA-seq results from all four replicates of each group were aligned and compared with the IGF gene signature panel. Pearson values from vehicle groups were compared between WT and two mutants with one-way analysis of variance (ANOVA) . (B) The heat maps from our previous RNA-seq shows normalized log2 fold change of IGF1 pathway related genes differentially regulated in mutants vs WT (fold change >1.5; P < 0.005). (C) Posthormone-deprived cells were pooled, and protein samples were extracted and subjected for immunoblotting with β-actin as the internal control. (D) Hormone-deprived MCF-7 and T47D cells were treated with vehicle, 1 nM of E2, or combination of 1 nM E2 and 1 μM fulvestrant for 24 hours. RNA was isolated, and qRT-PCR was performed (one-way ANOVA for comparison of basal level). Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Figure 1. View largeDownload slide MCF-7 and T47D cells with CRISPR-edited ESR1 Y537S and D538G mutations exhibit upregulation of IGF signaling components and downstream activity. (A) RNA-seq results from all four replicates of each group were aligned and compared with the IGF gene signature panel. Pearson values from vehicle groups were compared between WT and two mutants with one-way analysis of variance (ANOVA) . (B) The heat maps from our previous RNA-seq shows normalized log2 fold change of IGF1 pathway related genes differentially regulated in mutants vs WT (fold change >1.5; P < 0.005). (C) Posthormone-deprived cells were pooled, and protein samples were extracted and subjected for immunoblotting with β-actin as the internal control. (D) Hormone-deprived MCF-7 and T47D cells were treated with vehicle, 1 nM of E2, or combination of 1 nM E2 and 1 μM fulvestrant for 24 hours. RNA was isolated, and qRT-PCR was performed (one-way ANOVA for comparison of basal level). Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Y537S and D538G ESR1 mutant cells display enhanced proliferation in response to IGF1 Given the alteration in IGF signaling components, including the constitutive expression of IRS1, we hypothesized that ESR1 mutant cells would show an elevated response to IGF1. We therefore performed time-course proliferation assays with or without IGF1 after hormone deprivation and serum-free starvation (Fig. 2A). A hallmark of ESR1 mutant cells is the greater estrogen-independent growth in CSS (9). In the complete absence of serum, however, the ligand-independent growth of T47D mutant cells was not discerned (Fig. 2A), whereas MCF-7 mutant cells still maintained a stronger proliferation rate compared with WT controls, suggesting that the ligand-independent growth of T47D mutant cells depends on additional factors in CSS. When exposed to IGF1, T47D-Y537S cells exhibited a significantly (P < 0.0001) and approximately twofold stronger IGF1 response compared with WT control cells, whereas D538G mutant cells showed only a weak IGF1-induced proliferation that did not reach significance (Fig. 2A, bottom panels). In MCF-7, both mutations resulted in marked IGF1-mediated induction of proliferation, with D538G mutant cells showing a stronger response (Fig. 2A, top panels). Figure 2. View largeDownload slide ESR1 mutant cells display enhanced IGF1-stimulated growth response. (A) MCF-7 and T47D cells were hormone deprived for 3 days, followed by incubation in serum-free medium for 24 hours. Cells were then seeded into 96-well plates and treated with or without 100 ng/mL of IGF1 for up to 8 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Growth was normalized to day 0 and presented as fold change (FC). Two-way analysis of variance (ANOVA) was performed to compare the time-course growth rates between WT and mutant cells. Representative data shown from two independent experiments. (B) EC50 values were determined by IGF1 dose-response growth curves following the identical procedure as in (A) with IGF1 concentrations as indicated and 20 pM of E2. Mean EC50s are shown in the table with three (T47D) or two (MCF-7) independent experiments. One-way ANOVA was performed to compare the EC50s of WT and mutant cells. (C) Hormone-deprived and serum-free starved MCF-7 and T47D cells were treated with 100 ng/mL IGF1, 100 ng/mL IGF-II, 10 ng/mL VEGF, 10 ng/mL fibroblast growth factor-basic (bFGF), 10 ng/mL PDGF-BB, or 50 ng/mL EGF for 6 (MCF-7) or 7 (T47D) days. FCs induced by growth factors were obtained by normalizing to the average of vehicle controls. One-way ANOVA was applied to compare the growth factor effects between WT and mutant cells. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Figure 2. View largeDownload slide ESR1 mutant cells display enhanced IGF1-stimulated growth response. (A) MCF-7 and T47D cells were hormone deprived for 3 days, followed by incubation in serum-free medium for 24 hours. Cells were then seeded into 96-well plates and treated with or without 100 ng/mL of IGF1 for up to 8 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Growth was normalized to day 0 and presented as fold change (FC). Two-way analysis of variance (ANOVA) was performed to compare the time-course growth rates between WT and mutant cells. Representative data shown from two independent experiments. (B) EC50 values were determined by IGF1 dose-response growth curves following the identical procedure as in (A) with IGF1 concentrations as indicated and 20 pM of E2. Mean EC50s are shown in the table with three (T47D) or two (MCF-7) independent experiments. One-way ANOVA was performed to compare the EC50s of WT and mutant cells. (C) Hormone-deprived and serum-free starved MCF-7 and T47D cells were treated with 100 ng/mL IGF1, 100 ng/mL IGF-II, 10 ng/mL VEGF, 10 ng/mL fibroblast growth factor-basic (bFGF), 10 ng/mL PDGF-BB, or 50 ng/mL EGF for 6 (MCF-7) or 7 (T47D) days. FCs induced by growth factors were obtained by normalizing to the average of vehicle controls. One-way ANOVA was applied to compare the growth factor effects between WT and mutant cells. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. Dose-response studies were performed to evaluate the sensitivity and potency of IGF1 (Fig. 2B). Both mutations resulted in increased potency of IGF1-stimulated proliferation, particularly at concentrations >1 ng/mL of IGF1 (Fig. 2B, top panels), and this was observed in both cell line models. Examining sensitivity, EC50 calculated from the dose response curves suggested no difference between T47D WT and mutant cells, whereas MCF-7 Y537S and D538G mutants both showed a threefold to fourfold decrease in EC50 (Fig. 2B, bottom panel). We next determined whether the enhanced IGF1 response was a unique phenotype in ESR1 mutant cells or whether it is present in other endocrine-resistant models. We therefore tested IGF1 response in TamR breast cancer cells and in LTED models, mimicking resistance to aromatase inhibitors (Supplemental Fig. 3). These analyses showed that there was no noteworthy difference in IGF1 response between parental and the endocrine-resistant TamR and LTED models (Supplemental Fig. 3A and 3B). Furthermore, the dose-response curves in MCF-7 cells indicated that both TamR and LTED cells exhibited reduced sensitivity toward IGF1 exposure: a 4-fold and 18-fold increase in EC50 values was detected in the MCF-7–LTED and TamR models, respectively (Supplemental Fig. 3C). Together, these results suggest that the enhanced IGF1 response might be a unique property in endocrine-resistant cells with ESR1 mutations. We next determined the effect of other growth factors known to play a role in breast cancer progression to determine whether the IGF1 effects are unique or whether the ESR1 mutations sensitize the cells to activation of other growth factor receptors. Therefore, we measured growth response of ESR1 WT and mutant cells to IGF-II, VEGF, PDGF, fibroblast growth factor-basic (bFGF), and epidermal growth factor (EGF) (Fig. 2C). Although we did observe increased response of the Y537S mutant cells to IGF-II, bFGF, PDGF, and EGF, consistent increase in both cell lines for both mutations was only observed for IGF1. Phosphoinositide 3-kinase–Akt axis mediates the enhanced IGF1 response in ESR1 mutant cells We next investigated which downstream signaling pathways mediate the enhanced IGF1 response in ESR1 mutant cells (Fig. 3). Cells were stimulated with increasing concentrations of IGF1 for 15 minutes, and the phosphorylation status of IGF1R (Y1135) and canonical IGF1 downstream substrates Akt (T308 and S473) and ERK1/2 (T202/T204) were measured (Fig. 3). Consistent with the growth response data, greater IGF1R phosphorylation at Y1135 was observed in both MCF-7 (Fig. 3A and 3B) and T47D (Fig. 3C and 3D) and Y537S and D538G mutant cells, particularly at higher concentrations of IGF1 stimulation (80 or 100 ng/mL). The two phosphorylation sites (T308 and S473) of Akt are phosphorylated by different kinases and involved in distinct biological functions (13, 14). We noted that only pT308-Akt was increased in ESR1 mutant cells, whereas phosphorylation at pS473-Akt displayed higher basal levels but weaker or no IGF1-induced stimulation. In contrast, p-ERK1/2 (T202/T204) showed a cell line–specific induction, with greater induction in T47D but not in MCF-7 ESR1 mutants compared with WT. Figure 3. View largeDownload slide PI3K-Akt axis mediates the enhanced IGF1 response in ESR1 mutant cells. (A and C) MCF-7 and T47D cells were hormone deprived for 3 days and starved in serum-free medium for 24 hours. Cells were then stimulated with various concentrations of IGF1 for 15 minutes. Proteins were isolated, and immunoblots were performed with antibodies as indicated, with β-actin as loading control. (B and D) The intensities of bands were quantified by ImageJ (National Institutes of Health), and relative phosphorylation levels were calculated by correcting phosphorylation levels to total protein and then normalizing to vehicle controls. Representative data shown from two independent experiments. Figure 3. View largeDownload slide PI3K-Akt axis mediates the enhanced IGF1 response in ESR1 mutant cells. (A and C) MCF-7 and T47D cells were hormone deprived for 3 days and starved in serum-free medium for 24 hours. Cells were then stimulated with various concentrations of IGF1 for 15 minutes. Proteins were isolated, and immunoblots were performed with antibodies as indicated, with β-actin as loading control. (B and D) The intensities of bands were quantified by ImageJ (National Institutes of Health), and relative phosphorylation levels were calculated by correcting phosphorylation levels to total protein and then normalizing to vehicle controls. Representative data shown from two independent experiments. A parallel set of time-dependent IGF1 stimulation experiments were performed to compare the rates of IGF1 downstream substrate phosphorylation response in the various cell lines (Supplemental Fig. 4). In T47D WT cells, p-IGF1R (Y1135) was maximal after 45 minutes and declined at 60 minutes, whereas in both Y537S and D538G mutant cells p-IGF1R increased rapidly with a maximal response within only 10 minutes (Supplemental Fig. 4B). In MCF-7 cells, similar time-dependent effects were seen, but differences in p-IGF1R were found in ESR1 mutant cells compared with WT cells (Supplemental Fig. 4A). Consistently, pT308-Akt not pS473 site was differentially affected in cells with both types of ESR1 mutations. pT308-Akt reached the peak intensity within 5 and 10 minutes of IGF1 stimulation in T47D and MCF-7 mutant cells, whereas 10 and 30 minutes were required for the corresponding WT cells. Consistently, phosphorylation of S6 kinase also exhibited a stronger response in both ESR1 mutant cells, with a higher basal level in T47D cell line. We also discerned a more rapid ERK response in MCF-7-Y537S mutant cell line but not in other mutant cells. Given the observed effects on IGF sensitivity and potency, we next used a number of inhibitors to test the dependency of the observed increased IGF1 signaling on the phosphoinositide 3-kinase (PI3K)–Akt axis (Supplemental Fig. 4C). Specifically, we used wortmannin, BX-579, UO126, and rapamycin to inhibit PI3K, PDK1, mitogen-activated protein kinase kinase (MEK), and mTOR, respectively. In addition, as a control, we inhibited IGF1R with OSI-906 that, as expected, completely abrogated IGF1-induced growth. Effects of inhibition of PI3K, mTOR, and MEK were cell line dependent, with mTOR inhibitors blocking IGF1-induced growth in ESR1 mutant T47D cells, but not MCF-7 cells. In contrast, inhibition of PI3K and MEK was effective in ESR1 mutant MCF-7, but not in T47D cells. The only inhibitor that consistently resulted in inhibition of IGF1-induced growth in both cell line models was BX-579, an inhibitor of PDK1, a kinase that specifically catalyzes T308-Akt phosphorylation. In addition, to further confirm this signaling enhancement is a unique mechanism for ESR1 mutation, we examined the status of IGF1 downstream regulators in TamR and LTED models. TamR and LTED cell models showed decreased IGF1-induced phosphorylation of IGF1R (Y1135), IRS1 (S320), and Akt (T308 and S473) compared with parental controls (Supplemental Fig. 5A), which is consistent with previous studies using these cells and TamR breast cancers (15, 16). In contrast, an identical IGF1 stimulation triggered consistent and robust enhanced IGF1 downstream substrate phosphorylation in ESR1-mutant cells (Supplemental Fig. 5B). Knockdown of IRS1 attenuates the enhanced IGF1 response in ESR1 mutant cells Given the consistent and high upregulation of IRS1 in the ESR1 mutant cells, we next performed knockdown studies to test the functional significance of IRS1 in the increased IGF1 response. Growth curves demonstrated that loss of IRS1 reduced IGF1-induced proliferation in both WT and mutant cell lines, with stronger effects in the mutant cells (Fig. 4A). Figure 4. View largeDownload slide Knockdown of IRS1 attenuates the IGF1 response in ESR1 mutant cells. (A) IRS1 or scramble siRNAs were reverse transfected into hormone-deprived MCF-7 for 24 hours. Cells were then starved in serum-free medium and seeded into 96-well plates with 2500 (MCF-7) cells per well and treated with or without 100 ng/mL IGF1 for up to 6 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Cell numbers in scramble and IRS1 knockdown groups with IGF1 simulation were compared by Student t test. (B) MCF-7 and T47D cells were hormone deprived and transfected with IRS1, IGF1R, or scramble siRNA (siScramble). Cells were then starved in serum-free medium and treated with 100 ng/mL IGF1 for 15 minutes. Immunoblots were performed with antibodies as indicated. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. veh, vehicle. Figure 4. View largeDownload slide Knockdown of IRS1 attenuates the IGF1 response in ESR1 mutant cells. (A) IRS1 or scramble siRNAs were reverse transfected into hormone-deprived MCF-7 for 24 hours. Cells were then starved in serum-free medium and seeded into 96-well plates with 2500 (MCF-7) cells per well and treated with or without 100 ng/mL IGF1 for up to 6 days. Cell numbers were quantified by FluoReporter kit (Life Technologies). Cell numbers in scramble and IRS1 knockdown groups with IGF1 simulation were compared by Student t test. (B) MCF-7 and T47D cells were hormone deprived and transfected with IRS1, IGF1R, or scramble siRNA (siScramble). Cells were then starved in serum-free medium and treated with 100 ng/mL IGF1 for 15 minutes. Immunoblots were performed with antibodies as indicated. Representative data shown from two independent experiments. *P < 0.05; **P < 0.01. veh, vehicle. To investigate the corresponding signaling changes, we studied IGF1-induced phosphorylation of downstream targets in the absence and presence of IRS1 (Fig. 4B). The immunoblotting of scramble controls correlated strongly with the concentration and time-gradient stimulation results in Fig. 3. As a positive control, knockdown of IGF1R markedly reduced IGF1-induced activation of IRS1, Akt, and ERK1/2. Intriguingly, knockdown of IRS1 resulted in diminished IGF1R phosphorylation in all groups, indicating a yet-to-be-identified feedback mechanism. Consequently, the downstream activation of Akt (both S473 and T308) and ERK1/2 (T202/T204) were blocked, albeit the effect on p-ERK1/2 was weaker in MCF-7 normalized to scramble control groups. The enhanced phosphorylation of AKT (T308) was also diminished in MCF-7 cell line after IRS1 knockdown. Combination of anti-IGF1R (OSI-906) and anti-ER (fulvestrant) exhibits stronger synergism in ESR1 mutant cells Finally, we asked whether activation of the IGF1 pathway altered ESR1 mutant cells response to small-molecule inhibitors targeting IGF1R. First, we examined the effect of BMS-754807 and OSI-906 on two-dimensional growth of ESR1 mutant cells (Supplemental Fig. 6). MCF-7 cells were more sensitive to IGF1R inhibition compared with T47D. Both MCF-7 Y537S and D538G mutant cells exhibited greater half maximal inhibitory concentration toward both IGF1R inhibitors, with Y537S showing a stronger resistance. The similar resistance trend was also detected in response of T47D cells to BMS-754807; however, no obvious differences were detected with OSI-906. Combination therapies cotargeting IGF1R and ERα in breast cancer have been developed and evaluated (4). With respect to this therapeutic strategy, we examined the combined effect of OSI-906 and fulvestrant (Fig. 5A). Therefore, we treated WT and ESR1 mutant T47D and MCF-7 cells with increasing concentrations of fulvestrant (1 nM to 100 nM) and OSI-906 (0.1 μM to 10 μM). Effects on growth rates are shown in Fig. 5A, and combination index values are shown in Fig. 5B. These data show that under EC50s, OSI-906 and fulvestrant display synergistic effects [combination index (CI) <1] in both mutant cell line models in both MCF-7 and T47D, with the exception of T47D-Y537S cells that showed no effect (CI = 1). In contrast, antagonistic effects (CI > 1) were observed in both WT cell line models. At 75% and 90% effective dose, the drugs displayed synergistic activities in all mutant cell line models, with consistently stronger effects compared with WT cells. Figure 5. View largeDownload slide Combination treatment with IGF1R inhibitors and fulvestrant exhibits synergistic effects in ESR1 mutant cells. (A) Hormone-deprived MCF-7 and T47D clones were pooled and treated with the corresponding combined concentrations of OSI-906 and fulvestrant in the presence of 20 pM of E2. Cell numbers were quantified with FluoReporter kit (Life Technologies) after 6 (MCF-7) and 7 (T47D) days. Percentages of growth were calculated by normalizing to the averages of the vehicle control groups. The data were calculated by combining two independent experiments. (B) Combination index mean values at 50% effective dose (ED50), 75% effective dose (ED75), and 90% effective dose (ED90) were calculated based on two independent experiments. Figure 5. View largeDownload slide Combination treatment with IGF1R inhibitors and fulvestrant exhibits synergistic effects in ESR1 mutant cells. (A) Hormone-deprived MCF-7 and T47D clones were pooled and treated with the corresponding combined concentrations of OSI-906 and fulvestrant in the presence of 20 pM of E2. Cell numbers were quantified with FluoReporter kit (Life Technologies) after 6 (MCF-7) and 7 (T47D) days. Percentages of growth were calculated by normalizing to the averages of the vehicle control groups. The data were calculated by combining two independent experiments. (B) Combination index mean values at 50% effective dose (ED50), 75% effective dose (ED75), and 90% effective dose (ED90) were calculated based on two independent experiments. Discussion Recent data indicate that mutations in the ligand-binding domain of ESR1 confer constitutive ligand-independent gene expression and that this may lead to endocrine resistance and poor survival (17, 18). Using two breast cancer cell lines that are genome-edited with two of the most common ESR1 mutations, we have characterized augmented IGF1 response in these cell lines, which is partially attributed to the upregulation of IRS1. Elevated levels of IRS1 in ESR1 mutant cells increased IGF1 potency through the PI3K-Akt axis and triggered enhanced IGF1-induced growth (Fig. 6). Of note, employment of an IGF1R inhibitor (OSI-906) in combination with a selective ER downregulator (fulvestrant) showed synergistic growth inhibitory effects in ESR1 mutant cells, yet showed antagonism in WT MCF-7 cells. This study sheds light on a putative therapeutic strategy for ESR1 mutant breast cancers. Figure 6. View largeDownload slide Schematic view of enhanced IGF1 response in ESR1 mutant cells. In ESR1 WT cells, IGF1 activates IGF1R, which recruits either IRS1 or Shc to increase PI3K-Akt and/or mitogen-activated protein kinase (MAPK) pathways. With the acquisition of Y537S or D538G mutations, mutant ER binds to IRS1 promoter and increases IRS1 expression levels. Upregulation of IRS1 further amplifies the IGF1 response via the PI3K-Akt axis and thus enhances cell proliferation. Figure 6. View largeDownload slide Schematic view of enhanced IGF1 response in ESR1 mutant cells. In ESR1 WT cells, IGF1 activates IGF1R, which recruits either IRS1 or Shc to increase PI3K-Akt and/or mitogen-activated protein kinase (MAPK) pathways. With the acquisition of Y537S or D538G mutations, mutant ER binds to IRS1 promoter and increases IRS1 expression levels. Upregulation of IRS1 further amplifies the IGF1 response via the PI3K-Akt axis and thus enhances cell proliferation. Initial reports from Gelsomino et al. (8) identified enhanced IGF1R-ERα crosstalk in breast cancer cells transfected with mutant ESR1, with a focus on the higher basal IGF1R activation levels in ESR1 mutant cells. We find a similar increase in IGF1R-ERα crosstalk in genome-edited cells, corroborating the study of Gelsomino et al. (8); however, there are several important differences. Gelsomino et al. (8) found that IGF1R and IRS1 levels were unaltered in ESR1-transfected cells, despite these being estrogen-regulated genes. In contrast, we found elevated IGF1R and IRS1. One potential reason for these discrepancies is the difference in cell model construction, perhaps mainly the CRISPR vs transfection, with CRISPR heterozygous mutant clones more closely mimicking the genetic alteration in human breast cancers (19, 20). In addition, our findings were shown in two cell lines with two distinct high-frequent mutations allowing us to identify different context-dependent effects (e.g., greater IGF sensitization MCF-7 mutant ESR1 clones) (21, 22). This pattern matched the observation of IGF1R upregulation in MCF-7 mutant cells according to RNA-seq, which in part increases ligand activation and downstream signaling. According to our previous RNA-seq analysis, although estrogen-independent gene expression patterns were detected in all cell models, genes overlapping between Y537S and D538G mutant cells were only 20.1% and 43.5% of all of the constitutively expressed genes in T47D and MCF-7, respectively, suggesting mutation-specific ligand-independent genes and their consequential phenotypes. For instance, it is interesting to observe the enhanced bFGF response in T47D-Y537S and EGF response in Y537S mutant exclusively of both cell lines (Fig. 2C). It could be presumably explained by either the upregulation of signaling downstream kinases (PRKCA) or transcriptional factors (STAT5A and MYC), particularly under these signaling receptors. We also discerned higher basal phosphorylation levels in D538G but limited levels in Y537S mutants of MCF-7 cell line. This may also be explained by either the exclusively downregulation of TIMP3 (a matrix metalloproteinase inhibitor) or upregulation of GAB2 (a signaling adaptor) in the MCF7-D538G line, which potentially enhances the AKT basal levels through integrin or ER crosstalk. Intriguingly, we found constitutive phosphorylation of S473 on AKT in ESR1 mutant clones. Akt sites T308 and S473 sites are phosphorylated by PDK1 and mTORC2, respectively (23). Previous studies have identified site-specific regulatory mechanisms of Akt phosphorylation and downstream signaling (23). The Ser473 site phosphorylation is reported to be exclusively regulated by integrin-linked kinase, protein kinase Cα, and other receptor tyrosine kinase activation (4). It is possible that ESR1 mutant cells have these pathways activated, which increase basal pS473-Akt. Moreover, the mutant or cell line–specific upregulations of kinase components genes (DEPTOR) (24) or signaling adaptor genes (GAB2) (25) will potentially enhances the basal pS473-Akt. Additionally, our previous study (9) reveals that D538G mutation of T47D cell line exhibits relatively lower ligand-independent growth rate but stronger E2 response after hormone deprivation, which indicates a weaker constitutive ligand-independent activation gene pattern in this particular line, including the group IGF1 pathway genes shown in Fig. 1B. Consequently, the IGF1-induced proliferation was consistently weaker in T47D-D538G cell line, compared with other ESR1 mutation cell lines (Fig. 2A). Furthermore, as a ligand of insulin receptor (IR), IGF1 has the capacity to activate IR (26, 27). Interestingly, IR consistently decreased in all the mutant lines at both RNA (Supplemental Fig. 1A) and protein levels (Supplemental Fig. 1B), implicating that stronger growth response dominantly depends on IGF1R activation in ESR1 mutant cells. Similarly, the inhibitory effects of OSI-906 also mainly rely on the block of IGF1 activity rather than IR, even IR serves as one of the targets of OSI-906 (28). The equivalent IR isoform ratios (IR-A/IR-B) between different mutants of each cell line (Supplemental Fig. 1C) rules out the possibility that the altered IR isoform ratios leads to differentiated IGF-II response (Fig. 2C). We also examined the ER binding patterns on the IRS1 promoter in our cell models via chromatin immunoprecipitation quantitative polymerase chain reaction, with the primers designed based on the two reported putative binding sites (29) (Supplemental Fig. 7A). However, only MCF7-Y537S exhibits ligand-independent ER recruitment (Supplemental Fig. 7B and 7C). It is possible that the ER binding is present as a cell line–specific property or other potential ER-related indirect binding exists in T47D cell line. Finally, our data indicate that the enhanced IGF1 response is exclusive to ESR1 mutant cells and not found in TamR and LTED cells, indicating that it is not a general endocrine resistance mechanism. Consistent with this, Fagan et al. (16) reported the loss of IGF1R expression and IGF1 response in TamR cell models. These phenotypic distinctions further stress the need to dissect out the specific mechanisms of resistance and suggest that precise treatment of endocrine resistance may require the monitoring of ESR1 mutations, in which IGF1R inhibitors may have effect. To this end, further analysis examining gene profiles and identifying novel pathways, together with clinical archived materials, is warranted. Appendix. Antibody Table Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Monoclonal or Polyclonal  Dilution Used  RRID  p-IGF1R-Y1135    Phospho-IGF-I Receptor β (Tyr1135) (DA7A8) Rabbit mAb  Cell Signaling Technology, 3918  Monoclonal  1:200  AB_10548764  IGF1R    IGF-I Receptor β (D23H3) XP® Rabbit mAb  Cell Signaling Technology, 9750  Monoclonal  1:1000  AB_10950969  p-IRS1-S320    Phospho-IRS-1 (Ser302) Antibody  Cell Signaling Technology, 2384  Polyclonal  1:500  AB_330360  p-AKT-T308    Phospho-Akt (Thr308) (D25E6) XP® Rabbit mAb  Cell Signaling Technology, 13038  Monoclonal  1:1000  AB_2629447  p-AKT-S473    Phospho-Akt (Ser473) (D9E) XP® Rabbit mAb  Cell Signaling Technology, 4060  Monoclonal  1:1000  AB_2315049  AKT    Akt (pan) (40D4) Mouse mAb  Cell Signaling Technology, 2920  Monoclonal  1:1000  AB_1147620  p-ERK1/2-T202/T204    Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (197G2) Rabbit mAb  Cell Signaling Technology, 4377  Monoclonal  1:1000  AB_331775  ERK    p44/42 MAPK (Erk1/2) Antibody  Cell Signaling Technology, 9102  Polyclonal  1:1000  AB_330744  IRS1    IRS1  Santa Cruz Biotechnology, sc7200  Polyclonal  1:1000  AB_649615      Insulin Receptor β  Cell Signaling Technology, 3025  Monoclonal  1:1000  AB_2280448  β-actin    Monoclonal Anti-Actin antibody produced in mouse  Thermo Fisher Scientific, A3853  Monoclonal  1:10,000  AB_262137  Peptide/Protein Target  Antigen Sequence (if Known)  Name of Antibody  Manufacturer, Catalog No.  Monoclonal or Polyclonal  Dilution Used  RRID  p-IGF1R-Y1135    Phospho-IGF-I Receptor β (Tyr1135) (DA7A8) Rabbit mAb  Cell Signaling Technology, 3918  Monoclonal  1:200  AB_10548764  IGF1R    IGF-I Receptor β (D23H3) XP® Rabbit mAb  Cell Signaling Technology, 9750  Monoclonal  1:1000  AB_10950969  p-IRS1-S320    Phospho-IRS-1 (Ser302) Antibody  Cell Signaling Technology, 2384  Polyclonal  1:500  AB_330360  p-AKT-T308    Phospho-Akt (Thr308) (D25E6) XP® Rabbit mAb  Cell Signaling Technology, 13038  Monoclonal  1:1000  AB_2629447  p-AKT-S473    Phospho-Akt (Ser473) (D9E) XP® Rabbit mAb  Cell Signaling Technology, 4060  Monoclonal  1:1000  AB_2315049  AKT    Akt (pan) (40D4) Mouse mAb  Cell Signaling Technology, 2920  Monoclonal  1:1000  AB_1147620  p-ERK1/2-T202/T204    Phospho-p44/42 MAPK (Erk1/2) (Thr202/Tyr204) (197G2) Rabbit mAb  Cell Signaling Technology, 4377  Monoclonal  1:1000  AB_331775  ERK    p44/42 MAPK (Erk1/2) Antibody  Cell Signaling Technology, 9102  Polyclonal  1:1000  AB_330744  IRS1    IRS1  Santa Cruz Biotechnology, sc7200  Polyclonal  1:1000  AB_649615      Insulin Receptor β  Cell Signaling Technology, 3025  Monoclonal  1:1000  AB_2280448  β-actin    Monoclonal Anti-Actin antibody produced in mouse  Thermo Fisher Scientific, A3853  Monoclonal  1:10,000  AB_262137  Abbreviations: mAb, monoclonal antibody; MAPK, mitogen-activated protein kinase; RRID, Research Resource Identifier. View Large Abbreviations: bFGF fibroblast growth factor-basic CI combination index CSS charcoal stripped serum E2 17β-estradiol EC50 half maximal effective concentration EGF epidermal growth factor ER estrogen receptor IGF insulinlike growth factor IGF1 insulinlike growth factor-1 IGF1R insulinlike growth factor-1 receptor IRS1 insulin receptor substrate-1 LTED long-term estrogen deprivation MEK mitogen-activated protein kinase kinase mTOR mammalian target of rapamycin p- phosphorylated PDGF platelet-derived growth factor PI3K phosphoinositide 3-kinase qRT-PCR quantitative reverse transcription polymerase chain reaction RNA-seq RNA-sequencing siRNA small interfering RNA TamR tamoxifen-resistant VEGF vascular endothelial growth factor WT wild-type. Acknowledgments The authors thank Dr. Rachel Schiff for the gift of MCF-7 and ZR75-1 TamR and LTED cell models, the assistants from all Lee-Oesterreich laboratories, and all of the annotated grant sources. Financial Support: This study was in part supported by funds from the Nicole Meloche Memorial Breast Cancer Fund, the Pittsburgh Penguins Foundation, the Penguins Alumni Association, the Mario Lemieux Foundation, Fashion Footwear of New York, Breast Cancer Research Foundation (to A.V.L., S.O., and B.H.P.), and Avon Foundation for Women (to B.H.P.) through a Pilot Award from the Institute for Precision Medicine at the University of Pittsburgh (to S.O.) and through Grant P30CA047904 from the National Institutes of Health/National Cancer Institute. S.O. and A.V.L. are recipients of Scientific Leadership awards from Susan G. Komen. P.W. was supported by a China Scholarship Council award through Tsinghua Medical School (Beijing, China). Z.L. was supported by the John S. Lazo Cancer Pharmacology Fellowship. K.M.L. was supported by National Institutes of Health F30 Grant CA203154. Disclosure Summary: The authors have nothing to disclose. References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2016. CA Cancer J Clin . 2016; 66( 1): 7– 30. Google Scholar CrossRef Search ADS PubMed  2. Hayes EL, Lewis-Wambi JS. Mechanisms of endocrine resistance in breast cancer: an overview of the proposed roles of noncoding RNA. Breast Cancer Res . 2015; 17( 1): 40. Google Scholar CrossRef Search ADS PubMed  3. Clarke R, Tyson JJ, Dixon JM. Endocrine resistance in breast cancer--An overview and update. Mol Cell Endocrinol . 2015; 418( Pt 3): 220– 234. Google Scholar CrossRef Search ADS PubMed  4. Toy W, Shen Y, Won H, Green B, Sakr RA, Will M, Li Z, Gala K, Fanning S, King TA, Hudis C, Chen D, Taran T, Hortobagyi G, Greene G, Berger M, Baselga J, Chandarlapaty S. ESR1 ligand-binding domain mutations in hormone-resistant breast cancer. Nat Genet . 2013; 45( 12): 1439– 1445. Google Scholar CrossRef Search ADS PubMed  5. Fanning SW, Mayne CG, Dharmarajan V, Carlson KE, Martin TA, Novick SJ, Toy W, Green B, Panchamukhi S, Katzenellenbogen BS, Tajkhorshid E, Griffin PR, Shen Y, Chandarlapaty S, Katzenellenbogen JA, Greene GL. Estrogen receptor alpha somatic mutations Y537S and D538G confer breast cancer endocrine resistance by stabilizing the activating function-2 binding conformation. eLife . 2016; 5: e12792. Google Scholar CrossRef Search ADS PubMed  6. Harrod A, Fulton J, Nguyen VT, et al.  . Genomic modelling of the ESR1 Y537S mutation for evaluating function and new therapeutic approaches for metastatic breast cancer. Oncogene . 2017; 36: 2286– 2296. Google Scholar CrossRef Search ADS PubMed  7. Mao C, Livezey M, Kim JE, Shapiro DJ. Antiestrogen Resistant Cell Lines Expressing Estrogen Receptor α Mutations Upregulate the Unfolded Protein Response and are Killed by BHPI. Sci Rep . 2016; 6: 34753. Google Scholar CrossRef Search ADS PubMed  8. Gelsomino L, Gu G, Rechoum Y, Beyer AR, Pejerrey SM, Tsimelzon A, Wang T, Huffman K, Ludlow A, Andò S, Fuqua SAW. ESR1 mutations affect anti-proliferative responses to tamoxifen through enhanced cross-talk with IGF signaling [published correction appears in Breast Cancer Res Treat. 2017;163:639–640]. Breast Cancer Res Treat . 2016; 157( 2): 253– 265. Google Scholar CrossRef Search ADS PubMed  9. Bahreini A, Li Z, Wang P, Levine KM, Tasdemir N, Cao L, Weir HM, Puhalla SL, Davidson NE, Stern AM, Chu D, Park BH, Lee AV, Oesterreich S. Mutation site and context dependent effects of ESR1 mutation in genome-edited breast cancer cell models. Breast Cancer Res . 2017; 19( 1): 60. Google Scholar CrossRef Search ADS PubMed  10. Creighton CJ, Casa A, Lazard Z, Huang S, Tsimelzon A, Hilsenbeck SG, Osborne CK, Lee AV. Insulin-like growth factor-I activates gene transcription programs strongly associated with poor breast cancer prognosis. J Clin Oncol . 2008; 26( 25): 4078– 4085. Google Scholar CrossRef Search ADS PubMed  11. Mauro L, Salerno M, Panno ML, Bellizzi D, Sisci D, Miglietta A, Surmacz E, Andò S. Estradiol increases IRS-1 gene expression and insulin signaling in breast cancer cells. Biochem Biophys Res Commun . 2001; 288( 3): 685– 689. Google Scholar CrossRef Search ADS PubMed  12. Molloy CA, May FE, Westley BR. Insulin receptor substrate-1 expression is regulated by estrogen in the MCF-7 human breast cancer cell line. J Biol Chem . 2000; 275( 17): 12565– 12571. Google Scholar CrossRef Search ADS PubMed  13. Cicenas J. The potential role of Akt phosphorylation in human cancers. Int J Biol Markers . 2008; 23( 1): 1– 9. Google Scholar CrossRef Search ADS PubMed  14. Robey RB, Hay N. Is Akt the “Warburg kinase”?—Akt-energy metabolism interactions and oncogenesis. Semin Cancer Biol . 2009; 19: 25– 31. 15. Parisot JP, Hu XF, DeLuise M, Zalcberg JR. Altered expression of the IGF-1 receptor in a tamoxifen-resistant human breast cancer cell line. Br J Cancer . 1999; 79( 5-6): 693– 700. Google Scholar CrossRef Search ADS PubMed  16. Fagan DH, Uselman RR, Sachdev D, Yee D. Acquired resistance to tamoxifen is associated with loss of the type I insulin-like growth factor receptor: implications for breast cancer treatment. Cancer Res . 2012; 72( 13): 3372– 3380. Google Scholar CrossRef Search ADS PubMed  17. Jeselsohn R, Yelensky R, Buchwalter G, Frampton G, Meric-Bernstam F, Gonzalez-Angulo AM, Ferrer-Lozano J, Perez-Fidalgo JA, Cristofanilli M, Gómez H, Arteaga CL, Giltnane J, Balko JM, Cronin MT, Jarosz M, Sun J, Hawryluk M, Lipson D, Otto G, Ross JS, Dvir A, Soussan-Gutman L, Wolf I, Rubinek T, Gilmore L, Schnitt S, Come SE, Pusztai L, Stephens P, Brown M, Miller VA. Emergence of constitutively active estrogen receptor-α mutations in pretreated advanced estrogen receptor-positive breast cancer. Clin Cancer Res . 2014; 20( 7): 1757– 1767. Google Scholar CrossRef Search ADS PubMed  18. Jeselsohn R, Buchwalter G, De Angelis C, Brown M, Schiff R. ESR1 mutations—a mechanism for acquired endocrine resistance in breast cancer. Nat Rev Clin Oncol . 2015; 12( 10): 573– 583. Google Scholar CrossRef Search ADS PubMed  19. Kotin RM, Siniscalco M, Samulski RJ, Zhu XD, Hunter L, Laughlin CA, McLaughlin S, Muzyczka N, Rocchi M, Berns KI. Site-specific integration by adeno-associated virus. Proc Natl Acad Sci USA . 1990; 87( 6): 2211– 2215. Google Scholar CrossRef Search ADS PubMed  20. Sander JD, Joung JK. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat Biotechnol . 2014; 32( 4): 347– 355. Google Scholar CrossRef Search ADS PubMed  21. Ross RL, Askham JM, Knowles MA. PIK3CA mutation spectrum in urothelial carcinoma reflects cell context-dependent signaling and phenotypic outputs. Oncogene . 2013; 32( 6): 768– 776. Google Scholar CrossRef Search ADS PubMed  22. Spoerke JM, Gendreau S, Walter K, Qiu J, Wilson TR, Savage H, Aimi J, Derynck MK, Chen M, Chan IT, Amler LC, Hampton GM, Johnston S, Krop I, Schmid P, Lackner MR. Heterogeneity and clinical significance of ESR1 mutations in ER-positive metastatic breast cancer patients receiving fulvestrant. Nat Commun . 2016; 7: 11579. Google Scholar CrossRef Search ADS PubMed  23. Manning BD, Cantley LC. AKT/PKB signaling: navigating downstream. Cell . 2007; 129( 7): 1261– 1274. Google Scholar CrossRef Search ADS PubMed  24. Peterson TR, Laplante M, Thoreen CC, Sancak Y, Kang SA, Kuehl WM, Gray NS, Sabatini DM. DEPTOR is an mTOR inhibitor frequently overexpressed in multiple myeloma cells and required for their survival. Cell . 2009; 137( 5): 873– 886. Google Scholar CrossRef Search ADS PubMed  25. Hunzicker-Dunn ME, Lopez-Biladeau B, Law NC, Fiedler SE, Carr DW, Maizels ET. PKA and GAB2 play central roles in the FSH signaling pathway to PI3K and AKT in ovarian granulosa cells. Proc Natl Acad Sci USA . 2012; 109( 44): E2979– E2988. Google Scholar CrossRef Search ADS PubMed  26. Baserga R, Peruzzi F, Reiss K. The IGF-1 receptor in cancer biology. Int J Cancer . 2003; 107( 6): 873– 877. Google Scholar CrossRef Search ADS PubMed  27. Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P, Hemmings BA. Mechanism of activation of protein kinase B by insulin and IGF-1. EMBO J . 1996; 15( 23): 6541– 6551. Google Scholar PubMed  28. Mulvihill MJ, Cooke A, Rosenfeld-Franklin M, Buck E, Foreman K, Landfair D, O’Connor M, Pirritt C, Sun Y, Yao Y, Arnold LD, Gibson NW, Ji QS. Discovery of OSI-906: a selective and orally efficacious dual inhibitor of the IGF-1 receptor and insulin receptor. Future Med Chem . 2009; 1( 6): 1153– 1171. Google Scholar CrossRef Search ADS PubMed  29. Ross-Innes CS, Stark R, Teschendorff AE, Holmes KA, Ali HR, Dunning MJ, Brown GD, Gojis O, Ellis IO, Green AR, Ali S, Chin SF, Palmieri C, Caldas C, Carroll JS. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer. Nature . 2012; 481( 7381): 389– 393. Google Scholar CrossRef Search ADS PubMed  Copyright © 2018 Endocrine Society

Journal

EndocrinologyOxford University Press

Published: Jan 1, 2018

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