Hypoxia negates hyperglycaemia-induced chemo-resistance in breast cancer cells: the role of insulin-like growth factor binding protein 2

Abstract

Background: Women who suffer from breast cancer and type II diabetes with associated hyperglycaemia respond less well to chemotherapy. We have shown that hyperglycaemia induces resistance to chemotherapy through upregulation of fatty acid synthase (FASN) in breast cancer cells and increased insulin-like binding protein 2 (IGFBP-2) in prostate cancer cells. As a tumour develops the tumour mass can outgrow the blood supply resulting in the cancer cells being exposed to hypoxia that stimulates many tumorigenic signalling pathways.

Methods: We used MCF-7 and T47D breast cancer cell lines. Trypan blue dye exclusion assay was employed to assess cell death and Western immunoblotting was used to determine changes in protein abundance. Hypoxia was induced both chemically by the addition of cobalt chloride (CoCl₂) and using a hypoxia chamber.

Results: IGFBP-2 abundance increased with increasing concentrations of glucose (0-25 mM) that contributed to hyperglycaemia-induced chemo-resistance as it was abrogated by downregulating IGFBP-2 using siRNA. Production of IGFBP-2 is ER dependent: pre-treatment of MCF-7 cells with β-estradiol increased IGFBP-2 and induced chemo-resistance to doxorubicin. The hyperglycaemia-induced chemo-resistance and increases in FASN and IGFBP-2 were negated in a hypoxic environment, with levels of cell death unaffected by glucose concentrations.

Conclusions: The sensitivity of breast cancer cells to chemotherapy is reduced in hyperglycaemic conditions but this effect is negated by hypoxia. These effects appear to be mediated via regulation of IGFBP-2 and FASN. Understanding the role of FASN and IGFBP-2 in chemo-resistance could provide a novel target for improving the effectiveness of breast cancer treatment.

Keywords: chemo-resistance; estrogen receptor; hyperglycaemia; hypoxia; insulin-like growth factor binding protein 2.

Figure 1. IGFBP-2 increases in response to different glucose levels

MCF-7 cells were treated with 5, 9, and 25mM glucose for 48h and the abundance of secreted (A) and endogenous (B) IGFBP-2 was examined by Western immunoblotting. The optical densitometry measurements of IGFBP-2 secreted (C) and in lysates (D) in MCF-7 cells. T47D cells were treated with 5, 9, and 25mM glucose for 48h and the abundance of secreted (E) and in lysates (F) IGFBP-2 was examined by Western immunoblotting. The optical densitometry measurements of IGFBP-2 secreted (G) and in lysates (H) in T47D cells. (I) MCF-7 cells were seeded at 0.4x10⁶ cells in 5mM glucose growth medium for 24h prior to switching to SFM of different glucose concentrations 5, 9, and 25mM glucose for 24h. The cells were lysed with Trizol and the level of IGFBP-2 mRNA expression was determined by real-time PCR. (J) T47D cells were treated the same way as MCF-7 cells in 5, 9, and 25mM glucose for 24h, and the level of IGFBP-2 expression was determined by real-time PCR. 18S was used a internal housekeeping gene. The graphs represent the mean±SEM of three independent repeats each conducted in triplicate.

Figure 2. The role of IGFBP-2 in hyperglycaemia-induced chemo-resistance

MCF-7 cells (A) and T47D cells (B) were incubated in different glucose conditions (5mM, 9mM and 25mM) and treated in the presence and absence of IGFBP-2 siRNA. The cells were treated with 1μM doxorubicin. Cell death was measured by a trypan blue dye exclusion assay. Inserts: IGFBP-2 abundance in 5, 9, 25mM glucose in the presence and absence of IGFBP-2 siRNA where the total cell extract was analysed by Western immunoblotting. The graph represents the mean±SEM of three independent repeats each conducted in triplicate; NS= non-silencing siRNA, BP2= IGFBP-2 siRNA.

Figure 3. Human recombinant IGFBP-2 promotes survival against chemotherapy in breast cancer cells

(A) MCF-7 cells were seed at 0.1x10⁶ cells then were treated with 100ng/ml human recombinant IGFBP-2 in 5mM, 9mM and 25mM glucose. The cells were exposed to 1μM doxorubicin for 24h and cell death was measured by a cell count and viability assay. (B) T47D cells were seed at 0.1x10⁶ cells then were treated with 200ng/ml human recombinant IGFBP-2 in 5mM, 9mM and 25mM glucose. The cells were exposed to 5μM doxorubicin for 24h and cell death was measured by a cell count and viability assay. The graphs represent the mean±SEM of three independent repeats each conducted in triplicate.

Figure 4. The effect sensitizing ERs on IGFBP-2 secretion

MCF-7 cells were seeded at 0.0125x10⁶ cells then were treated with various doses of 17β-estradiol and conjugated 17β-estradiol ranging from 0nM to 50nM for 24h. IGFBP-2 concentrations were determined by ELISA after treatment with 17β-estradiol (A) and conjugated 17β-estradiol (B) The graphs represents the mean±SEM of three independent repeats each conducted in triplicate (* P<0.05, *** P<0.001).

Figure 5. Nuclear ERs increase IGFBP-2 levels that is protective against chemotherapy

(A) MCF-7 cells were seeded at 0.1x10⁶ cells in 5mM growth medium then were serum starved for 24h in 5mM and 25mM glucose SFM. The cells were treated with 17β-estradiol for 48h and doxorubicin was introduced in the last 24h of incubation. Trypan blue exclusion assay was used to determine the level of cell death. (B) Western immunoblotting showing changes in IGFBP-2 and FASN protein abundance in 5mM and 25mM glucose in the presence and absence of 17β-estradiol. Optical densitometry measurements of IGFBP-2 (C) and FASN (D) in response to 17β-estradiol in 5mM and 25mM glucose. The graphs represent the mean±SEM of three independent experiments each performed in triplicate.

Figure 6. The effect of hypoxia on hyperglycaemia-induced chemo-resistance

(A) MCF-7 cells were treated with 500μM CoCl₂. Changes in the abundance of IGFBP-2 was analysed using Western immunoblotting. (B) Densitometry measurements of IGFBP-2 abundance in normoxic and hypoxic conditions in MCF-7 cells treated in 5mM and 25mM glucose. Hypoxia was chemically-induced in MCF-7 cells by 500μM CoCl₂ (C) and b low levels of oxygen (2% O₂) (D) in 5mM and 25mM glucose and the cells were treated with 1μM doxorubicin for 24h. The rate of cell death was analysed using a trypan blue dye exclusion assay. HIF-Iα was used as an indicator of successful induction of hypoxia by 500μM CoCl₂ (E) and 2% O₂ (F) α-tubulin was used as a loading control. (G) Changes in IGFBP-2 concentrations in MCF-7 cells treated in 5mM and 25mM glucose in the presence or absence of 500μM CoCl₂ and 1μM doxorubicin determined using ELISA. The graph represents the mean±SEM of three independent repeats each conducted in triplicate.

Figure 7. The involvement of the FASN/ERα/Akt pathway in hypoxia regulation of IGFBP-2

(A) MCF-7 cells were seeded at 0.1x10⁶ cells in 5mM growth medium. The cells were then treated in 5mM and 25mM glucose and were exposed to hypoxia. Changes in the abundance of HIF-Iα, FASN, ERα, Akt and p-Akt were analysed using Western immunoblotting. Densitometry measurements of changes in the abundance of FASN (B) ER (C) p-Akt/Akt (D) under normal and hypoxic conditions in 5mM and 25mM glucose. The graphs represent the mean±SEM of three independent repeats each conducted in triplicate.