Identification of a Resistance Mechanism to IGF-IR Targeting in Human Triple Negative MDA-MB-231 Breast Cancer Cells

Abstract

Triple negative breast cancer (TNBC) is associated with unfavorable prognosis and high relapse rates following chemotherapy. There is an urgent need to develop effective targeted therapy for this BC subtype. The type I insulin-like growth factor receptor (IGF-IR) was identified as a potential target for BC management. We previously reported on the production of the IGF-Trap, a soluble IGF-1R fusion protein that reduces the bioavailability of circulating IGF-1 and IGF-2 to the cognate receptor, impeding signaling. In nude mice xenotransplanted with the human TNBC MDA-MB-231 cells, we found variable responses to this inhibitor. We used this model to investigate potential resistance mechanisms to IGF-targeted therapy. We show here that prolonged exposure of MDA-MB-231 cells to the IGF-Trap in vitro selected a resistant subpopulation that proliferated unhindered in the presence of the IGF-Trap. We identified in these cells increased fibroblast growth factor receptor 1 (FGFR1) activation levels that sensitized them to the FGFR1-specific tyrosine kinase inhibitor PD166866. Treatment with this inhibitor caused cell cycle arrest in both the parental and resistant cells, markedly increasing cell death in the latter. When combined with the IGF-Trap, an increase in cell cycle arrest was observed in the resistant cells. Moreover, FGFR1 silencing increased the sensitivity of these cells to IGF-Trap treatment in vivo. Our data identify increased FGFR1 signaling as a resistance mechanism to targeted inhibition of the IGF-IR and suggest that dual IGF-1R/FGFR1 blockade may be required to overcome TNBC cell resistance to IGF-axis inhibitors.

Keywords: FGFR1; IGF signaling; drug resistance; the IGF-Trap; triple negative breast cancer.

Figure 1

Ligand induced IGF-1R signaling in MDA-MB-231 cells. Cells were serum starved for 24 h and then stimulated with 100 ng/mL IGF-1 for the indicated time intervals. Shown in (A) is a representative Western blot of the IGF-IR and downstream signaling mediators following IGF-1 stimulation. Shown in the bar graphs (B–D) are the mean (and SE (standard error)) of 3 experiments expressed as pIGF-1R/IGF-1R, pERK/ERK, and pAkt/Akt ratios, respectively, normalized to the levels of unstimulated cells that were assigned a value of 1. * p < 0.05.

Figure 2

The IGF-Trap blocks IGF-1R signaling and inhibits the proliferation of MDA-MB-231 cells. Cells were serum starved for 24 h and then stimulated with 100 ng/mL IGF-1 in the presence of IGF-Trap added at a molar ratio of 2:1 (IGF-Trap:IGF-1) for the indicated time intervals. Shown in (A) is a representative Western blot of the IGF-1R and downstream signaling mediators following IGF-1 stimulation. Shown in the bar graphs (B–D) are the means (±SE) of 3 experiments expressed as pIGF-1R/IGF-1R, pERK/ERK, and pAkt/Akt ratios, respectively, normalized to the levels of unstimulated cells that were assigned a value of 1. The anti-proliferative effect of the IGF-Trap was measured by the MTT assay. Cells were treated with the indicated concentrations of the IGF-Trap in the presence of serum for 3 days. Shown in (E,F) are mean % (±SE) (top) and fold change (bottom) relative to control, vehicle-treated cells that were assigned a value of 100% and 1, respectively (n = 3). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 3

Acquired resistance to the growth inhibitory effect of IGF-Trap and augmented ERK signaling following sustained exposure of MDA-MB-231 to IGF-Trap in vitro. MDA-MB-231 exposed to gradually increasing concentrations of IGF-Trap for 3 months (MDA-MB-231-R) and the parental cells were treated with the IGF-Trap (315 μg/mL) for 72 h. Cell proliferation was measured using the MTT assay. Shown in (A) are means ± SE of the results (n = 5) expressed as fold change relative to the respective vehicle-treated controls that were assigned a value of 1. ERK activation (B) was analyzed by Western blotting. Shown in (B) is a representative immunoblot and in (C) the means ± SE (n = 3) expressed as fold change in the pERK/ERK ratio relative to vehicle-treated cells that were assigned a value of 1. Shown in (D) is a representative Western blot of additional signal transduction mediators analyzed in these cells, and in the bar graph (E) the means (±SE) of 3 experiments expressed as phosphorylated to total protein ratios normalized to the levels of parental MDA-MB-231 cells that were assigned a value of 1. * p < 0.05, ** p < 0.01, ns—not significant.

Figure 4

Upregulated expression and increased activation of FGFR1 in IGF-Trap resistant MDA-MB-231-R cells. MDA-MB-231 and MDA-MB-231-R cells were treated with 315 μg/mL IGF-Trap in the presence of serum for 72 h. A phopsho-RTK (receptor tyrosine kinase) array was then used (A) to profile phosphorylated RTKs in total cell lysates derived from these cells. Western blotting (B) and qPCR (D–F) were used to analyze activated FGFR1 (B) and mRNA expression (D–F) levels of the indicated transcripts in cells treated with 160 μg/mL IGF-Trap for 72 hr prior to lysis. Shown in (B) is a representative immunoblot and in (C) the data expressed as pFGFR1/FGFR1 ratios normalized to the level in vehicle-treated MDA-MB-231 cells that were assigned a value of 1. Data in (D–F) were normalized to GAPDH and are expressed as means (±SE) fold change in transcript expression in MDA-MB-231-R cells relative to MDA-MB-231 cells that were assigned a value of 1 (n = 4). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 5

Increased sensitivity of MDA-MB-231-R cells to FGFR1 signaling blockade. Cells in complete medium were treated for 72 h with the FGFR1 inhibitor PD166866 and cell proliferation measured by the MTT assay. Shown in (A) is a dose response curve and in (B) a bar graph depicting the anti-proliferative effect of PD166866. Results are expressed as means ± SE (n = 3) relative to the respective, vehicle-treated controls that were assigned a value of 100% and 1, respectively. * p < 0.05, *** p < 0.001.

Figure 6

Cell cycle arrest in MDA-MB-231 and MDA-MB-231-R cells treated with PD166866. Cells were treated with 10 μM PD166866 or vehicle in complete medium for 72 h, fixed, stained with propidium iodide and cell cycle analysis performed by flow cytometry. Shown are representative flow cytometry histograms of MDA-MB-231 (A,B) and MDA-MB-231-R (D,E) cells. Shown in the bar graphs are the calculated proportions of cells at different cell cycle phases for MDA-MB-231 (C) and MDA-MB-231-R cells (F) and in (G) the proportions of dead cells for each cell type. Results are based on three independent experiments and are expressed as mean percentage (±SE) of total cells in (C,F) and percent of all events analyzed in (G). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 6

Cell cycle arrest in MDA-MB-231 and MDA-MB-231-R cells treated with PD166866. Cells were treated with 10 μM PD166866 or vehicle in complete medium for 72 h, fixed, stained with propidium iodide and cell cycle analysis performed by flow cytometry. Shown are representative flow cytometry histograms of MDA-MB-231 (A,B) and MDA-MB-231-R (D,E) cells. Shown in the bar graphs are the calculated proportions of cells at different cell cycle phases for MDA-MB-231 (C) and MDA-MB-231-R cells (F) and in (G) the proportions of dead cells for each cell type. Results are based on three independent experiments and are expressed as mean percentage (±SE) of total cells in (C,F) and percent of all events analyzed in (G). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 7

FGFR1 silencing enhances MDA-MB-231 cell sensitivity to the IGF-Trap. FGFR1 was silenced using lentiviral FGFR1 short hairpin RNA (shRNA). Shown in (A,C) are representative Western blots confirming FGFR1 silencing in MDA-MB-231 and MDA-MB-231-R cells, respectively. Cells were treated with the indicated concentrations of the IGF-Trap in the presence of serum for 48 hr and proliferation (B,D) measured using the colorimetric MTT assay. Data are expressed as means ± SE (n = 3), relative to controls that were assigned a value of 1. ** p < 0.01, *** p < 0.001.

Figure 8

Combination FGFR and IGF-IR targeting enhances cell cycle arrest in MDA-MB-231-R cells. Cells were starved overnight and then cultured in DMEM medium containing 10% fetal bovine serum (FBS) with (or without) IGF-Trap (160 μg/mL), PD166866 (20 μM) or both for 48 h prior to cell cycle analysis. A single representative cell cycle profile is shown for MDA-MB-231-R (A–D) and MDA-MB-231-R cells in which FGFR1 was silenced using shRNA (MDA-MB-231-R/FGFR1 KD cells (F–I) treated with vehicle (A,F), IGF-Trap (B,G), PD166866 (C,H) or combination of the two (D,I). The proportions of cells (%) in the different cell cycle phases are shown in the bar graphs (E,J). * p < 0.05, ** p < 0.01, *** p < 0.001.

Figure 9

FGFR1 silencing increases the sensitivity of MDA-MB-231-R cells to IGF-Trap treatment in vivo. FGFR1-silenced MDA-MB-231-R cells in Matrigel were injected into the mammary fatpads of female NSG mice and the mice treated, twice weekly, with 5mg/kg IGF-Trap, intravenously. Tumor volumes were recorded twice weekly. Shown are mean tumor volumes (±SE) in mice injected with wild type (A), control transfected (B) and FGFR1 silenced (C) MDA-MB-231-R cells (n = 5). Note the initial delay in tumor appearance in mice injected with FGFR1 silenced MDA-MB-231 cells. * p < 0.05 as analyzed by the Student’s t-test.

Figure 10

A postulated model for the mechanism underlying IGF-Trap resistance in MDA-MB-231. The IGF-Trap inhibits IGF-IR signaling, decreasing ERK phosphorylation and cell cycle progression and inducing apoptosis in MDA-MB-231 cells (Left). Our findings suggest that MDA-MB-231 cells that are continuously exposed to the IGF-Trap lose the sensitivity to the growth inhibitory effect of the Trap, due to increased autocrine FGFR1 signaling and ERK activation, mediated by increased FGFR1/FGF1 expression (Right). The increase in constitutive ERK phosphorylation may also contribute to increased mTORC1/S6K activation and protein synthesis in these cells, either directly or through crosstalk with the PI3K-AKT pathway [36]. Please note that postulated interactions are indicated by dotted arrows, while interactions elucidated by our data are indicated by solid arrows.