Elevated insulin-like growth factor 1 receptor signaling induces antiestrogen resistance through the MAPK/ERK and PI3K/Akt signaling routes

Antibodies to rabbit anti-phospho-IGF-1Rβ (Tyr1131), anti-phospho-IGF-1Rβ (Tyr1135/Tyr1136), anti-ERK1/2, anti-phospho-ERK1/2 (Thr202/Tyr204), anti-Akt and anti-phospho-Akt (Ser473) (Cell Signaling Technology Danvers, MA, USA), mouse anti-IGF-1Rβ and rabbit anti-ERα (Santa Cruz Biotechnology, Santa Cruz, CA, USA), and mouse antitubulin (Sigma-Aldrich, St. Louis, MO, USA) were commercially purchased. Conjugated secondary antibodies included Alexa Fluor 488 antimouse (Jackson ImmunoResearch, West Grove, PA, USA), antimouse horseradish peroxidase (HRP) and antirabbit HRP (Jackson ImmunoResearch, West Grove, PA, USA), and antirabbit alkaline phosphatase (AP) (Tropix Western-SuperStar™ Immunodetection System; Applied Biosystems, Foster, CA, USA). Human IGF-1 (Sigma-Aldrich) was prepared in sterile H₂O (100 μg/mL). The estrogen compound 17β-estradiol (E2) and the antiestrogens 4-hydroxytamoxifen (4-OH-TAM) and FUL (Sigma-Aldrich) were dissolved in dimethyl sulfoxide (DMSO) to 1 mM stock concentration. The IGF-1R inhibitor BMS-536924 and dual PI3K/mammalian target of rapamycin (mTOR) inhibitor BEZ235 (Selleck Chemicals LLC Houston, TX, USA) and the mitogen-activated protein kinase kinase (MEK) inhibitor U0126 (Promega, Madison, WI, USA), were prepared in DMSO at 10 mM stock concentration.

MCF7 cells (American Type Culture Collection, Manassas, VA, USA) were cultured in RPMI 1640 medium (Gibco, Invitrogen, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS) and 100 U/mL penicillin-streptomycin (Invitrogen). The retroviral vector pMSCV-neo-IGF-1R containing neomycin resistance gene and expressing human wild-type IGF-1R cDNA was provided by Dr. R. Baserga [24]. IGF-1R-encoding retroviruses were produced by transfection of pMSCV-neo-IGF-1R into Phoenix Amphotropic packaging cells as previously described [25]. MCF7 cells were infected with freshly harvested retroviral supernatant. Two days later infected cells were selected by using 400 μg/mL neomycin. To establish individual positive clones from single cells, the neomycin selected cells were further subjected to limiting dilution in 96-well plates for two-week colony selection. To rule out clonal artefacts, we picked up the wells containing three to five individual positive clones and collected and expanded the multiple clones to generate stable MCF7/IGF-1R cells.

Cells were placed in six-or twelve-well plates at 60% to 70% confluence in regular growth medium. The next day cells were starved overnight with 1% FBS medium. Following an additional two-hour serum deprivation with serum-free medium (SFM), cells were exposed to IGF-1 at a given dose and for a given time course, which was directly followed by cell lysis. To inhibit IGF-1-stimulated signaling, cells were pretreated with a kinase inhibitor in SFM for two hours prior to IGF-1 exposure.

Cells were seeded with 10,000 cells/well in 96-well plates and incubated overnight. Prior to drug treatments, cells were starved for two days in phenol red-free RPMI 1640 medium (Gibco) supplemented with 5% charcoal/dextran-stripped FBS (CDFBS) (HyClone Laboratories, Thermo Scientific, Logan, UT, USA) devoid of steroid hormones. Starved cells were then subjected to individual or combined drug treatments in triplicate and allowed to proliferate for four days.

A sulforhodamine B (SRB) colorimetric assay [26, 27] was used to determine cell proliferation that occurred in given drug treatments. In short, drug-treated cells in 96-well plates were fixed with 30 μL of 50% trichloroacetic acid directly added to 100 μL of assay medium per well for 1 hour at 4°C on a shaker, gently washed five times with tap water and air-dried. Fixed cells were stained with 60 μL of 0.4% SRB (dissolved in 1% acetic acid) at room temperature on a shaker for 30 minutes to allow the SRB to bind to protein, rinsed five times with 1% acetic acid to remove unbound dye and then air-dried. Subsequently, the protein-bound SRB in each well was solubilized with 200 μL of 10 mM Tris base solution (pH > 10) for 10 minutes on a plate shaker and measured for its absorbance at 510 nm with a FLUOstar OPTIMA plate reader (BMG LABTECH, Offenburg, Germany). The SRB absorbance values at 510 nm are thus indicative of cell proliferation in response to drug treatments.

In addition to the SRB assay, a modified 3D culture was established to further examine cellular responses to growth factor IGF-1 stimulation and drug treatments. Briefly, a 96-well plate was coated with 30 μL/well of 100% Matrigel (354230, BD Matrigel™ Basement Membrane Matrix Growth Factor Reduced; BD Biosciences, Franklin Lakes, NJ, USA) at 37°C for 45 minutes. Single cells (10,000 cells/well) were suspended in 120 μL of starving medium (phenol red-free RPMI 1640) containing 2% soluble Matrigel and 5% CDFBS (instead of complete medium) in which E2 was depleted, and the cells were then seeded on top of coating Matrigel. For given treatments, the concentrations of the drugs were adjusted for the total volume of 150 μL. Fresh medium was added every three or four days. After two weeks of culturing on Matrigel, the cells were fixed and stained with 3.7% formaldehyde in phosphate-buffered saline (PBS) containing 0.2% Triton X-100, 0.25 μM rhodamine phalloidin (Sigma) and 1 μg/mL Hoechst blue dye. Confocal imaging of 3D cell culture was performed with a Nikon ECLIPSE TE2000-E (Nikon Instruments, Tokyo, Japan).

To prepare cell lysates for Western blot analysis, cells were washed three times with ice-cold PBS and lysed on ice for 30 minutes with lysis buffer (50 mM Tris, pH 7.5, 150 mM NaCl, 2 mM ethylenediaminetetraacetic acid, 0.1% sodium dodecyl sulfate (SDS), 1% Nonidet P-40 and 1% deoxycholic acid) freshly supplemented with 100-fold diluted protease inhibitor cocktail (P8340-ML; Sigma-Aldrich). Harvested lysate supernatant was measured for cellular protein concentration using the BCA™ Protein Assay Kit (Thermo Scientific, Rockford, IL, USA). A quantity of 30 μg/lane total protein were separated by SDS-polyacrylamide gel electrophoresis on 7.5% acrylamide gel and electrophoretically transferred to polyvinylidene fluoride membrane (Millipore, Billerica, MA, USA). Prior to primary antibody probe, membrane was blocked for 1 hour at room temperature with 5% bovine serum albumin (BSA) in Tris-buffered saline Tween 20 (TBST) buffer (100 mM Tris, pH 7.4, 500 mM NaCl, 0.05% Tween 20) or with I-Block buffer (Tropix, Applied Biosystems). Phospho-IGF-1Rβ (Tyr1131) and phospho-ERK1/2 (Thr202/Tyr204) were probed in 5% BSA-TBST buffer, whereas phospho-IGF-1Rβ (Tyr1135/Tyr1136) and phospho-Akt (Ser473) were probed in I-Block buffer. HRP-or AP-conjugated secondary antibody incubation was performed in 5% BSA-TBST or I-Block buffer, corresponding to the primary antibodies used. Protein bands were visualized by using the ECL Plus method (GE Healthcare, Little Chalfont, Buckinghamshire, UK), after which the membrane was scanned by using a Typhoon 9400 imager (GE Healthcare) or a Tropix Western-SuperStar™ procedure (Applied Biosystems) by placing the membrane in contact with standard X-ray film (GE Healthcare).

Cells seeded onto coverslips were washed once with PBS, fixed with 80% acetone in H₂O and then blocked with 5% normal goat serum in PBS containing 0.05% Tween 20. Expression of IGF-1R was detected by mouse monoclonal antibody against IGF-1Rβ and labeled with Alexa Fluor 488 antimouse antibody. Nuclear DNA was stained with 4',6-diamidino-2-phenylindole. The stained cells were visualized under a fluorescence microscope (Nikon Eclipse E600; Nikon Instruments) at a × 60 lens objective.

To silence target genes in cells, 50 nM SMARTpool siRNA mix (Dharmacon Technologies, Thermo Scientific, Lafayette, CO, USA) was delivered into 15,000 cells/well in 96-well plates by using a standard transfection method with DharmaFECT 4 transfection reagent (Dharmacon Technologies) according to the manufacturer's instructions. After 24 hours, the small interfering RNA (siRNA) transfection mixture was replaced with complete medium or with 5% CDFBS starving medium if drug treatment and SRB proliferation assay were included. Cells were kept in culture for one more day before analysis of knockdown or further treatment.

Cells were resuspended in antibiotic-free culture medium, and 40,000 cells/well were seeded into 48-well plates. By use of Lipofectamine PLUS reagent (Invitrogen), cells were transiently transfected with 0.16 μg of the estrogen response element (ERE)-thymidine kinase-luciferase plasmid kindly provided by Dr. R. Michalides [28]. After three hours, cells were starved with 5% CDFBS medium for two days. Following 12-hour treatments as indicated, cells were washed once with PBS and lysed with 1 × passive lysis buffer (Dual-Luciferase Assay Kit; Promega). ERE-luciferase activity was measured using a luminometer (CentroXS³ LB 960; Berthold Technologies, Bad Wildbad, Germany).

Each average SRB absorbance value was derived from triplicate samples. Statistical analyses of all experimental data were performed using a two-sided Student's t-test. Significance was set at P < 0.05.

To establish a breast cancer cell line stably overexpressing IGF-1R, human MCF7 breast cancer cells were retrovirally transduced with a pMSCV-neo-IGF-1R vector and subjected to single-cloning selection following limiting dilution. The established MCF7/IGF-1R cell line stably expressed ectopic IGF-1R (Figure 1a), with expression approximately 10-fold that of parental MCF7 cells (Figure 1b). The proliferative response of MCF7/IGF-1R cells to IGF-1 (half-maximal effective concentration (EC₅₀) = 2.4 ng/mL) was increased compared to MCF7 cells (EC₅₀ = 16.0 ng/mL) (Figure 1c).

IGF-1R contains a triple tyrosine cluster in the kinase domain, Tyr1131, Tyr1135 and Tyr1136, which is required for full kinase activation of IGF-1R [29]. To demonstrate IGF-1R autoactivation by ligand binding, time course (Figure 1b) and dose range exposures to IGF-1 (Figure 1d) were performed. Triple tyrosine IGF-1R phosphorylation was initiated rapidly and sustained for long time periods, reaching maximal levels at 100 ng/mL IGF-1. Overall, MCF7/IGF-1R cells displayed stronger IGF-1R autophosphorylation than parental MCF7 cells, indicating that MCF7/IGF-1R cells gained elevated intrinsic IGF-1R tyrosine kinase activity, which is necessary for the activation of the IGF-1-stimulated downstream signaling cascades.

IGF-1R signal transduction involves several major phosphorylation cascades, including the MAPK and PI3K pathways [30, 31]. To confirm canonical IGF-1R signal transduction in the cell lines used, the activity of the downstream kinases ERK and Akt was determined. Concurrently with IGF-1R autophosphorylation, both ERK and Akt kinases became phosphorylated in parallel (Figure 1b). While maximal Akt phosphorylation was induced to a similar degree, maximal ERK phosphorylation was clearly higher in MCF7/IGF-1R cells than in MCF7 cells (Figures 1b and 1d), appearing to be consistent with IGF-1R phosphorylation levels. These data indicate that the MAPK/ERK and PI3K/Akt signal transduction cascades are induced via ligand-activated IGF-1R kinase activity, with increased MAPK/ERK signaling in the MCF7/IGF-1R cells.

Of note, equal levels of ERα were detected in both MCF7 and MCF7/IGF-1R cells (Figure 1b), indicating that overexpression and activation of IGF-1R does not influence ERα expression in the MCF7/IGF-1R cell model.

Next, we evaluated the sensitivity of MCF7/IGF-1R cells to E2 and the antiestrogens 4-OH-TAM and FUL. MCF7/IGF-1R cells were responsive to proliferative effects of E2 (EC₅₀ = 5 × 10^-11 M) (Additional file 1), similar to parental MCF7 cells. The proliferation induced by E2 (1 nM) in both cell types was antagonized by 4-OH-TAM and FUL (IC₅₀ = 4 × 10^-8 M) (Figure 2), indicating that ectopic IGF-1R expression in MCF7/IGF-1R cells does not affect ERα responses.

To address whether IGF-1 stimulation affects MCF7/IGF-1R cellular sensitivity to antiestrogens, cells were treated with a concentration range of 4-OH-TAM (Figure 2a) or FUL (Figure 2b) in combination with E2 (1 nM), IGF-1 (100 ng/mL) or a combination of E2 and IGF-1 as indicated. While 4-OH-TAM and FUL at 10^-6 M completely blocked E2-induced proliferation, IGF-1 stimulation desensitized MCF7/IGF-1R cells to the antiproliferative effects of 4-OH-TAM and FUL, sustaining proliferation level similar to that induced by IGF-1 alone and close to that induced by the IGF-1 and E2 co-treatment. IGF-1 stimulation also led to partial resistance of MCF7 cells to the antiestrogens at 10^-6 M, but it was much less significant than that of MCF7/IGF-1R cells. Together, these results indicate that enhanced IGF-1R signaling upon IGF-1 stimulation directly renders MCF7/IGF-1R cells insensitive to 4-OH-TAM and FUL, leading to strong antiestrogen resistance.

4-OH-TAM and FUL regulate ER function differently. 4-OH-TAM is a competitive E2 antagonist and blocks ER transcriptional activity [32], whereas FUL antagonizes E2 and degrades ER protein [33]. This differential ER regulation by 4-OH-TAM and FUL was equal in both MCF7 and MCF7/IGF-1R cells. 4-OH-TAM did not affect ERα expression, while FUL decreased ERα protein levels (Figure 3a).

To investigate whether tamoxifen resistance of MCF7/IGF-1R cells is related to ER function, we silenced ERα by siRNA transfection, reaching > 80% knockdown of ERα (Figure 3b). Compared to siRNA control (siCtrl), ERα silencing (siERα) significantly inhibited proliferation by E2 (65% to 70%; P < 0.01) (Figure 3c), but not by IGF-1 either alone or in cotreatment with E2 plus 4-OH-TAM (1 μM) (Figure 3d). In addition, the ERE-luciferase assay showed that IGF-1 did not affect ERE-mediated transcription (Figure 3e). While E2 led to a high level of ERE-mediated transcription, 4-OH-TAM sustained its antagonistic action on E2 and suppressed ER transcriptional ability in the presence of IGF-1 (Figure 3e). Furthermore, mRNA expression analysis demonstrated that IGF-1 stimulation did not significantly interfere with the expression of ER target genes, for instance, the known E2-responsive genes CXCL12 (chemokine (C-X-C motif) ligand 12; also named stromal cell-derived factor 1) and FOXC1 (forkhead box C1) [34, 35]. The induction of both CXCL12 (Additional file 2a) and FOXC1 (Additional file 2b) gene expression in response to E2 was significantly inhibited by 4-OH-TAM antagonistic activity in either the absence or the presence of IGF-1.

These results suggest that IGF-1R signal-mediated antiestrogen resistance in MCF7/IGF-1R cells does not involve ERα-dependent processes. IGF-1-stimulated proliferation may run parallel to E2-induced proliferation and therefore is not inhibited by the antiestrogens 4-OH-TAM and FUL.

Activation of IGF-1R signaling upon IGF-1 exposure led to downstream ERK and Akt phosphorylation (Figures 1b and 1c). To evaluate the role of these kinase pathways in IGF-1R signal-induced antiestrogen resistance in MCF7/IGF-1R cells, we first used several specific kinase inhibitors, including the IGF-1R inhibitor BMS-536924 [36], the MEK inhibitor U0126 [37] and the PI3K inhibitor BEZ235 [38]. BMS-536924 efficiently blocked IGF-1R autophosphorylation by IGF-1 stimulation and downstream phosphorylation of both ERK and Akt by IGF-1R signal transduction (Figure 4a). U0126 (10 μM) largely inhibited ERK phosphorylation without interfering with activation of either IGF-1R or Akt. BEZ235 (1 μM) completely blocked Akt phosphorylation, leaving IGF-1R and ERK signaling intact (Figure 4a). These results suggest that, indeed, ERK and Akt are downstream from IGF-1R signaling in a linear fashion.

Next, the engagement of the ERK and Akt pathways in IGF-1R signal-mediated antiestrogen resistance in MCF7/IGF-1R cells was examined by use of the respective kinase inhibitors (Figures 4b and 4c). IGF-1 stimulation increased MCF7/IGF-1R proliferation by 60% to 64% (P < 0.001) over the cotreatment of 4-OH-TAM or FUL (1 μM) with E2, whereas this IGF-1-promoted antiestrogen resistance was significantly abrogated by BMS-536924, U0126 and BEZ235 equivalently to their inhibitory effects on relative kinase phosphorylation.

The key components involved in the PI3K/Akt and MAPK/ERK pathways have been well defined [39]. Using several other specific kinase inhibitors (Figure 5a) and targeting siRNA (Figure 5b and Additional file 3), we further delineated the functional activities of the known PI3K/Akt and MAPK/ERK components in the 4-OH-TAM resistance of MCF7/IGF-1R cells and positioned them in the IGF-1R signaling map (Figure 5c, dashed circles) by using MetaCore Pathway Analysis software (GeneGo, St. Joseph, MI, USA) according to fold changes in proliferation (Additional file 3 Table S1). As expected, inhibition of IGF-1R by either kinase inhibitor or siRNA silencing blocked IGF-1-stimulated proliferation (Figure 5c, exp 1 and 2, blue bars). Insulin receptor substrate 1 (IRS-1), which is known to be phosphorylated by autoactivated IGF-1R upon ligand binding, was shown to be involved in IGF-1R-mediated 4-OH-TAM resistance (Figure 5c, exp 2, blue bar). Kinase inhibitors blocking the signaling of the MAPK/ERK and PI3K/Akt pathways led to overall inhibition of proliferation (Figure 5c, blue bars in exp 1). Silencing of the major MAPK/ERK and PI3K/Akt kinases c-Raf-1 (Raf1), ERK2 (MAPK1), PI3K cat class A (PIK3CA), Akt (AKT1), mTOR (FRAP1) and p70 S6 kinase 1 (RPS6KB1) largely abrogated IGF-1-induced 4-OH-TAM resistance (Figure 5c, exp 2, blue bars). The PI3K negative regulators PI3K reg class IA (PI3KR1) and phosphatase and tensin homolog (PTEN) greatly increased proliferation when knocked down (Figure 5c, exp 2, red bars), further indicating the role of the PI3K pathway in the 4-OH-TAM resistance of MCF7/IGF-1R cells. In a novel way, our results reveal that, when targeted by siRNA, phosphoinositide-dependent kinase (PDK) (PDPK1) and p90Rsk (RPS6KA2), which are known positive regulators of transcription factor CREB1 (cyclic adenosine monophosphate-responsive element binding protein 1), decreased cell proliferation (Figure 5c, exp 2, blue bars), whereas inhibitor of nuclear factor κB kinase subunit (IKK) (CHUK) and glycogen synthase kinase 3 (GSK3) (GSK3B), which are known inhibitors of transcription factors NF-κB and c-Myc, significantly increased cell proliferation (Figure 5c, exp 2, red bars). In addition, silencing of apoptosis signal-regulating kinase 1 (ASK1) (MAP3K5), which is negatively regulated by IGF-1R phosphorylation to prevent apoptosis, enhanced cell proliferation (Figure 5c, exp 2, red bars).

Together, our results confirm that IGF-1R signaling incites multiple downstream cascades in which the key MAPK/ERK and PI3K/Akt components constitute the central signaling nodes that regulate IGF-1/IGF-1R signal-mediated proliferation and antiestrogen resistance in MCF7/IGF-1R cells.

Next, we further investigated the effect of IGF-1 stimulation on MCF7/IGF-1R 4-OH-TAM resistance in a more structurally and physiologically relevant context by use of a modified 3D culture. Both parental MCF7 (Figure 6a) and MCF7/IGF-1R cells (Figure 6b) were responsive to E2 or IGF-1 by forming acini on Matrigel, but the response was significantly larger in MCF7/IGF-1R cells with altered acinar morphogenesis (zoom images at bottom of Figures 6a and 6b). 4-OH-TAM effectively blocked E2-induced acini in both cell types (4-OH-TAM + E2). However, addition of IGF-1 (4-OH-TAM + E2 + IGF-1) overcame the antiproliferative effects of 4-OH-TAM and allowed acini growth of MCF7/IGF-1R cells but not MCF7 cells, demonstrating IGF-1-stimulated 4-OH-TAM resistance of MCF7/IGF-1R cells in 3D culture. Similarly to the SRB 2D assay, inhibition of IGF-1R/ERK/Akt signaling by respective kinase inhibitors restored 4-OH-TAM sensitivity of MCF7/IGF-1R cells in 3D culture (Figure 6c).

We have demonstrated that IGF-1 treatment of MCF7/IGF-1R cells overruled the antiproliferative effect of tamoxifen (at 1 μM). However, we also note that at low concentrations of 4-OH-TAM, an increase in proliferation was triggered on top of IGF-1-driven proliferation (Figure 2a), suggesting a potential role for enhanced intrinsic IGF-1R signaling in tamoxifen resistance. To investigate to what extent IGF-1R signaling causes this 4-OH-TAM agonistic effect, cells were coexposed to 4-OH-TAM in doses of 10, 33 or 100 ng/mL IGF-1, which were shown to cause different levels of sustained receptor and downstream signaling activation (Figure 1d). Consistently, IGF-1 promoted a 4-OH-TAM promitogenic effect in MCF7/IGF-1R cells but not in MCF7 cells, with the highest potential occurring at 4-OH-TAM 10^-9 to 10^-8 M (Figure 7a). This 4-OH-TAM promitogenic effect was most strongly induced by IGF-1 between 33 and 100 ng/mL, concentrations which were shown to induce maximal IGF-1R signaling (Figure 1d).

As expected, at 10 nM 4-OH-TAM, a marginal antagonistic effect was observed to block E2 (1 nM)-mediated proliferation (Figure 7b) and ERE-luciferase activity (Figure 7c) in both MCF7 and MCF7/IGF-1R cells. Interestingly, 10 nM 4-OH-TAM provided a mild promitogenic effect, increasing IGF-1-dependent proliferation (17% to 27%; P < 0.05) (Figure 7d) and ERE-mediated ER transcriptional activity in MCF7/IGF-1R cells (P < 0.05) (Figure 7e), whereas it remained inhibitory in parental MCF7 cells (15% to 20%; P < 0.05) (Figures 7d and 7e). These agonistic effects were not observed for FUL.

To further evaluate whether ERα is functionally required for this 4-OH-TAM agonistic switch in MCF7/IGF-1R cells, siRNA ERα knockdown experiments were performed. In siRNA control (siCrtrl) conditions, 10 nM 4-OH-TAM maintained a promitogenic effect on IGF-1-stimulated MCF7/IGF-1R cells (18% to 25%; P < 0.05), but ERα knockdown (siERα) eliminated the proliferation increase induced by 10 nM 4-OH-TAM (Figure 7f). These results imply that enhanced IGF-1R intrinsic kinase activity triggered this switch of 4-OH-TAM (10 nM) from an ERα antiproliferative modulator into a promitogenic modulator, indicating an agonistic effect on ER.

We also investigated whether the agonistic behavior of 4-OH-TAM (10 nM) in IGF-1-stimulated MCF7/IGF-1R cells is conferred via the ERK and Akt signaling pathways. Inhibition of IGF-1R signaling by BMS-536924 effected complete abrogation of IGF-1-stimulated proliferation and 4-OH-TAM (10 nM) agonistic action (Figure 8a). While 1 μM U0126 inhibited neither phosphorylation of ERK (Figure 4a) nor IGF-1-stimulated proliferation, it did clearly attenuate tamoxifen agonistic behavior (P < 0.01), with a subsequent competence in proliferative inhibition at 10 μM (Figure 8a). IGF-1-stimulated proliferation was drastically restrained by BEZ235 at either 1 or 10 μM (Figure 8a). Similar inhibitory patterns by the kinase inhibitors were observed in parental MCF7 cells, except for the 4-OH-TAM (10 nM) promitogenic effect (Additional file 4a).

We noted that although the BEZ235 inhibitory effect (even at 1 μM) was drastic, the 4-OH-TAM (10 nM) agonistic action seemed not to be completely abolished (Figure 8a). Therefore, further experiments with a lower dose range of BEZ235 were performed. While IGF-1R and ERK signaling remained unaffected in response to IGF-1, the phosphorylation level of Akt was diminished with an increase in the dose of BEZ235, largely at 0.1 μM and entirely at 0.5 μM (Figure 8b). Likewise, the cell proliferation rate decreased correspondingly to Akt phosphorylation levels, significantly dropping at 0.1 μM BEZ235 (Figure 8c), which occurred similarly in parental MCF7 cells (Additional file 4b). Intriguingly, the 4-OH-TAM (10 nM) agonistic effect on MCF7/IGF-1R cells appeared irreversible by BEZ235 (Figure 8c). These data suggest that 4-OH-TAM (10 nM) agonist conversion results from elevated intrinsic IGF-1R/MAPK/ERK signaling in MCF7/IGF-1R cells.