The paracrine effects of adipocytes on lipid metabolism in doxorubicin-treated triple negative breast cancer cells

Abstract

Adipocytes in the breast tumour microenvironment promotes acquired treatment resistance. We used an in vitro adipocyte-conditioned media approach to investigate the direct paracrine effects of adipocyte secretory factors on MDA-MB-231 breast cancer cells treated with doxorubicin to clarify the underlying treatment resistance mechanisms. Cell-viability assays, and Western blots were performed to determine alterations in apoptotic, proliferation and lipid metabolism protein markers. Free fatty acids (FFA) and inflammatory markers in the collected treatment-conditioned media were also quantified. Adipocyte secretory factors increased the cell-viability of doxorubicin-treated cells (p < 0.0001), which did not correspond to apoptosis or proliferation pathways. Adipocyte secretory factors increased the protein expression of hormone-sensitive lipase (p < 0.05) in doxorubicin-treated cells. Adipocyte secretory factors increased the utilization of leptin (p < 0.05) and MCP-1 (p < 0.01) proteins and possibly inhibited release of linoleic acid by doxorubicin-treated cells (treatment-conditioned media FFA profiles). Adipocyte secretory factors induced doxorubicin treatment resistance, by increasing the utilization of inflammatory mediators and inhibiting the release of FFA by doxorubicin-treated cells. This further promotes inflammation and lipid metabolic reprogramming (lipid storage) in the tumour microenvironment, which breast cancer cells use to evade the toxic effects induced by doxorubicin and confers to acquired treatment resistance.

Keywords: Adipocytes; breast cancer; fatty acids; inflammation; lipolysis; treatment resistance.

Figure 1.

The effect of adipocyte secretory factors in conditioned-medium on cell-viability in MDA-MB-231 triple negative breast cancer cells treated with/without doxorubicin for after 24 hours and 48 hours. Results are presented as mean ± SEM (n = 3). Two-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. **** = p < 0.0001. Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio)

Figure 2.

The effect of adipocyte secretory factors in conditioned-medium on (a) PARP and (b) caspase-3 protein cleavage in MDA-MB-231 triple negative breast cancer cells treated with/without doxorubicin for 48-hours. Results are presented as mean ± SEM, and as a fold change (n = 3 PARP and n = 4 Caspase-3). One-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. * = p < 0.05. Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio)

Figure 3.

The effect of adipocyte secretory factors in conditioned-medium on proliferation pathways. (a) PI3K phosphorylation (Tyr 467, Tyr 199), (b) Akt phosphorylation (Ser 473), (c) ERK1/2 phosphorylation (pERK1, Thr 202/Tyr 204; pERK2, Thr185/Tyr187) and (d) NFĸB-p65 protein expression in MDA-MB-231 triple negative breast cancer cells treated with/without doxorubicin for 48-hours (n = 3). Results are presented as mean ± SEM and as a fold change (except for NFĸB-p65). One-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. * = p < 0.05, ** = p < 0.01, and **** = p < 0.0001. Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio)

Figure 4.

The effect of adipocyte secretory factors in conditioned-medium on EMT markers in MDA-MB-231 triple negative breast cancer cells treated with/without doxorubicin for 48-hours. (a) Vimentin, and (b) E-cadherin. Results are presented as mean ± SEM (n = 3). One-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. * = p < 0.05 and ** = p < 0.01. Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio)

Figure 5.

The effect of adipocyte secretory factors in conditioned-medium on markers of lipid metabolism in MDA-MB-231 triple negative breast cancer cells treated with/without doxorubicin for 48-hours (a) FAS, (b) SCD-1, (c) ATGL, and (d) HSL. Results are presented as mean ± SEM (n = 3, except HSL n = 4). One-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. * = p < 0.05, ** = p < 0.01 and *** = p < 0.001. ATGL, Adipose triglyceride lipase; FAS, Fatty acid synthase; Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio)

Figure 6.

Individual FFA (% of total µg FFA) in treatment-conditioned media of MDA-MB-231 triple negative breast cancer cells treated with/without adipocyte-conditioned media and/or doxorubicin for 48-hours. (a) Palmitic Acid (PA), (b) Stearic Acid (SA), (c) cis-Vaccenic Acid (cVA), (d) Oleic Acid (OA), and (e) Linoleic Acid (LA). Results are presented as mean ± SEM (n = 3). One-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. * = p < 0.05, ** = p < 0.01 and *** = p < 0.001. Adipocyte CM; 100% Adipocyte-conditioned media; Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio). Value above bars represents the mean value

Figure 7.

Inflammatory marker concentrations in treatment-conditioned media of MDA-MB-231 triple negative breast cancer cells treated with/without adipocyte-conditioned media and/or Dox for 48-hours. (a) Adiponectin, (b) Leptin, (c) MCP-1, (d) IL-1β, and (e) Resistin. Results are presented as mean ± SEM (n = 4). One-way ANOVA with Fishers LSD post hoc correction was employed. p < 0.05 was considered as statistically significant. * = p < 0.05, ** = p < 0.01 and **** = p < 0.0001. Adipocyte CM, 100% Adipocyte-conditioned media; Control, normal growth media; CM, 30% adipocyte-conditioned media and 70% growth media ratio; Dox, 2.5 µM Doxorubicin; Dox+CM, 2.5 µM Doxorubicin + CM (30% adipocyte-conditioned media and 70% growth media ratio)

Figure 8.

Summary of the in vitro model findings. HSL, Hormone sensitive lipase; MCP-1, macrophage chemoattractant protein-1; MET, mesenchymal-to-epithelial transition