Adiponectin triggers breast cancer cell death via fatty acid metabolic reprogramming

Abstract

Background: Adiponectin, the most abundant adipokine derived from adipose tissue, exhibits a potent suppressive effect on the growth of breast cancer cells; however, the underlying molecular mechanisms for this effect are not completely understood. Fatty acid metabolic reprogramming has recently been recognized as a crucial driver of cancer progression. Adiponectin demonstrates a wide range of metabolic activities for the modulation of lipid metabolism under physiological conditions. However, the biological actions of adiponectin in cancer-specific lipid metabolism and its role in the regulation of cancer cell growth remain elusive.

Methods: The effects of adiponectin on fatty acid metabolism were evaluated by measuring the cellular neutral lipid pool, free fatty acid level, and fatty acid oxidation (FAO). Colocalization between fluorescent-labeled lipid droplets and LC3/lysosomes was employed to detect lipophagy activation. Cell viability and apoptosis were examined by MTS assay, caspase-3/7 activity measurement, TUNEL assay, and Annexin V binding assay. Gene expression was determined by real time-quantitative polymerase chain reaction (RT-qPCR) and western blot analysis. The transcriptional activity of SREBP-1 was examined by a specific dsDNA binding assay. The modulatory roles of SIRT-1 and adiponectin-activated mediators were confirmed by gene silencing and/or using their pharmacological inhibitors. Observations from in vitro assays were further validated in an MDA-MB-231 orthotopic breast tumor model.

Results: Globular adiponectin (gAcrp) prominently decreased the cellular lipid pool in different breast cancer cells. The cellular lipid deficiency promoted apoptosis by causing disruption of lipid rafts and blocking raft-associated signal transduction. Mechanistically, dysregulated cellular lipid homeostasis by adiponectin was induced by two concerted actions: 1) suppression of fatty acid synthesis (FAS) through downregulation of SREBP-1 and FAS-related enzymes, and 2) stimulation of lipophagy-mediated lipolysis and FAO. Notably, SIRT-1 induction critically contributed to the adiponectin-induced metabolic alterations. Finally, fatty acid metabolic remodeling by adiponectin and the key role of SIRT-1 were confirmed in nude mice bearing breast tumor xenografts.

Conclusion: This study elucidates the multifaceted role of adiponectin in tumor fatty acid metabolic reprogramming and provides evidence for the connection between its metabolic actions and suppression of breast cancer.

Keywords: Adiponectin; Breast cancer; Cancer metabolism; Lipophagy; SIRT-1; SREBP-1.

Fig. 1

Effect of globular adiponectin on cellular lipid reservation in breast cancer cells. A-C MCF-7 (A), T47D (B), and MDA-MB-231 (C) cells were treated with gAcrp (1 μg/mL) for indicated time periods. Cells were then incubated with Bodipy 493/503 for 30 min. Neutral lipid accumulation was determined as mean fluorescence intensity (MFI) using flow cytometry analysis. D-E MCF-7 (D) and MDA-MB-231 (E) cells were treated with gAcrp for 24 h, followed by labeling of lipid droplets with Nile red. Representative images for gAcrp-treated and control cells were shown along with quantification of lipid droplet area in respect to DAPI, in which the lipid droplet areas were determined by total area of red fluorescence using Image J software (scale bar: 20 μm). * denotes p < 0.05 compared to control cells (n=3 except where specifically indicated in Figures)

Fig. 2

Critical roles of lipid metabolic remodeling in breast cancer cell death induction by globular adiponectin. A-B MCF-7 (A) and MDA-MB-231 cells (B) were treated with gAcrp (1 μg/mL) or TVB-3166 (200 nM), a pharmacological inhibitor of FASN, for 48 h. Lipid raft microdomains were labeled with Alexa fluor 488-conjugated Cholera toxin B (CT-B) as described in Methods (green color). Representative images from three independent experiments were presented (upper panel) along with quantification of the corrected total cell fluorescence (CTCF) for CT-B (lower panel), in which CTCF was determined by integrated density - (area of selected cell × mean fluorescence of background readings) as previously described [38]. C-D MDA-MB-231 cells were treated with gAcrp for 48 h, followed by further stimulated with Wnt3a (100 ng/mL) for 6 h. C Lipid raft (green) was stained with Alexa fluor 488-conjugated CT-B and membrane p-LRP6 was labeled with a specific primary antibody and an Alexa fluor 594-conjugated secondary antibody (red). Scale bar: 20 μm. D p-LRP6, LRP6, and non-phospho- (active) β-catenin levels were examined by western blot analysis. Bar diagram shows quantification of blots. E-I MCF-7 (E and G) and MDA-MB-231 (F, H, and I) cells were treated with gAcrp (1 μg/mL) or TVB-3166 (200 nM) for 48 h in serum-free culture media in the absence of presence of BSA-conjugated palmitic acid (50 mM). Cell viability (E-F) and capase-3/7 activity (G-H) were determined at the end of treatment period as described in Methods. I Cells were double-stained with FITC-annexin V and 7-AAD, and subjected to flow cytometry analysis. Bar diagram shows the percentage of annexin V-positive cells that represents apoptotic cell population. * indicates p < 0.05 compared to control cells; # indicates p < 0.05 compared to cells treated with gAcrp alone; $ indicates p < 0.05 compared to cells treated with TVB-3166 alone; n=3 except where specifically indicated in Figures

Fig. 3

Modulation of fatty-acid-synthesis related genes by globular adiponectin in breast cancer cells. MCF-7 (A, C, E, and H) and MDA-MB-231 (B, D, F, G, and I) cells were treated with gAcrp (1 μg/mL) for indicated time periods. A-B The expression levels of precursor and nuclear SREBP-1 were determined by western blot analysis. C-D SREBP-1 mRNA levels were measured by RT-qPCR. E-F The SREBP-1 specific dsDNA binding activity was examined as described in Methods. G The expression of the genes modulating fatty acid synthesis, including FASN, ACC-1, ACLY, FADS1, FADS2, and SCD-1, were analyzed by RT-qPCR. H-I The protein levels of FASN were examined by western blot analysis. * indicates p < 0.05 compared to control; n=3 except where specifically indicated in Figures

Fig. 4

Activation of lipophagy by globular adiponectin in breast cancer cells. A-B MCF-7 (A) and MDA-MB-231 (B) cells were treated with 1 μg/mL of gAcrp for different time periods. The expression levels of autophagy-related genes, including p62, Beclin-1, Atg5, and LC3I/II were examined by western blot analysis. C-F Cells were cultured in the media containing 10 μM of oleic acid/BSA to induce lipid droplet formation and treated with 1 μg/mL of gAcrp for 6 h. C-D Lipid droplets and LC3 were detected by labelling with Nile red (red), and incubation with an anti-LC3 primary antibody followed by Alexa fluor 488-conjugated antibody (green), respectively. E-F Lipid droplets and lysosomes were labeled with Bodipy 493/503 (green) and Lysotracker (red). Co-localization rate was determined by Mander’s overlap coefficient using Image J software. Scale bar: 20 μm. G-H Cells were treated with gAcrp for the indicated time duration. Free fatty acid levels were measured at the end of treatment periods as indicated in the methods. I MDA-MB-231 cells were treated with gAcrp (1 μg/mL) for 8 h. Fatty acid oxidation (FAO) was determined as described in the methods. * denotes p < 0.05 compared to control cells; n=3 except where specifically indicated in Figures

Fig. 5

Roles of SIRT-1 in fatty acid metabolic reprogramming by globular adiponectin in breast cancer cells. A-B MCF-7 (A) and MDA-MB-231 (B) cells were treated with gAcrp as indicated. SIRT-1 expression was determined by western blot analysis. C MCF-7 cells were pretreated with EX527 (5 μM) for 2 h, followed by incubation with gAcrp (1 μg/mL) for further 8 h. The protein levels of SREBP-1 and FASN were examined by western blot analysis. D MCF-7 cells were transfected with a siRNA targeting SIRT-1 or a scramble control siRNA for 36 h, followed by treatment with gAcrp (1 μg/mL) for 8 h. (Upper panel) The gene silencing efficiency was monitored by western blot analysis. (Lower panel) The expression levels of FASN and SREBP-1 in SIRT-1 knockdown cells were examined after gAcrp treatment. E MCF-7 cells were pretreated with EX527 for 2 h, followed by further incubation with gAcrp (1 μg/mL) for 12 h. mRNA levels of FASN, ACC-1, ACLY, and FADS2 were measured by RT-qPCR. F-H MCF-7 cells were transfected with SIRT-1 siRNA (25 nM) for 36 h, followed by treatment with 1 μg/mL of gAcrp for 24 h (F), 48 h (G), or 6 h (H). F Cellular neutral lipid content was determined by Bodipy 493/503 uptake assay. G Cells were labeled with Alexa fluor 488-conjugated CT-B and lipid raft microdomains were observed under a confocal microscope. H Lipid droplets were stained with Nile red and autophagosomes were labeled with an Alexa fluor 488-conjugated anti-LC3 antibody. The overlapping between lipid droplets (red) and autophagosomes (green) were observed under a confocal microscope. The Mander’s overlap coefficient was used to test the colocalization of lipid droplets and autophagosomes. Scale bar: 20 μm. I-M MCF-7 cells were transfected with SIRT-1 siRNA (25 nM) (I) or pretreated with EX527 for 2 h (J-M), followed by further treatment with 1 μg/mL of gAcrp for 48 h. I and J Cell viability was measured by MTS assay. K-L The apoptosis level was determined using caspase-3/7 activity (K) and TUNEL assay (L) as indicated in the methods. M Expression levels of Bax and Bcl2 were examined by western blot analysis. * denotes p < 0.05 compared to control; # denotes p < 0.05 compared to cells treated with gAcrp alone; n=3 except where specifically indicated in Figures

Fig. 6

Involvement of SIRT-1 induction and mTOR signaling in the modulation of SREBP-1 by globular adiponectin. A-D MCF-7 cells were treated with gAcrp (1 μg/mL) for 8 h (A-B) or transfected with 25 nM SIRT-1 siRNA for 36 h before treatment with gAcrp for 8 h. (A and C) Acetylated SREBP-1 level was determined by immunoprecipitation with an anti-acetyl lysine antibody and western blot analysis using an anti-SREBP-1 primary antibody. B and D Ubiquitinated SREBP-1 level was measured by immunoprecipitation with an anti-ubiquitin antibody, followed by western blot analysis using an anti-SREBP-1 primary antibody. E-F Cells were treated with gAcrp (1 μg/mL) for different time periods. The total and phospho-mTOR levels were determined by immunoblotting. G-H MCF-7 cells were transfected with SIRT-1 siRNA for 36 h (G) or pretreated with EX527 for 2 h (H), followed by treatment with gAcrp for additional 3 h. p-mTOR/mTOR levels were examined by western blot analysis. I-J MCF-7 cells were treated with different concentrations of rapamycin (I) and MHY1485 (J) for 8 h. SREBP-1 and FASN expression levels were measured by western blot analysis. K-L MCF-7 cells were treated with rapamycin and MHY1485 for 24 h. Intracellular lipid accumulation (K) and free fatty acid (L) levels were determined as described above. * indicates p < 0.05 compared to control; # indicates p < 0.05 compared to gAcrp-treated cells; n=3

Fig. 7

Role of SIRT-1 in adiponectin modulation of in vivo breast tumor lipid metabolism and growth. MDA-MB-231-luc orthotopic breast tumors were generated in BALB/c nude mice, followed by treatment with gAcrp alone or gAcrp in combination with EX527 for 28 days. A and B Luminescent images of tumors (A) and tumor growth rate were monitored by luminescent in vivo imaging during treatment (B). C and D Tumor tissues were harvested after 4 weeks of treatment. Isolated tumors were captured at the end of experiment (C) and tumor weight was recorded (D). E Tissue section was prepared, and cleavage of caspase-3 was examined by immunohistochemistry (IHC). The percentage of cleaved caspase-3 positive tumor cells was determined by Image J software. F The expression levels of Bax and Bcl2 were measured by western blot analysis. The representative images from 3 mice each group were shown along with blot quantification for all collected tumor tissues. G The expression levels of FASN and SREBP-1 were analyzed by immunoblotting analysis. H SREBP-1 was detected in tumor tissues by IHC. The proportion of nuclear SREBP-1 positive cells were presented in bar diagram. Scale bar: 100 μm. I The mRNA levels of SREBP-1, FASN, ACC-1, FADS2, and ACLY in tumor tissues were measured by RT-qPCR. J The protein levels of p-mTOR, mTOR, β-catenin, and LC3I/II were determined by western blot analysis. K-L Single cells were isolated from tumor tissues by incubating with collagenase solution. K Tumor cells were incubated with Bodipy 493/503 for 15 min at 37^oC, followed by flow cytometry analysis. L The free fatty acid level was measured in tumor cells and normalized to tumor cell number

Fig. 8

Proposed model for the modulation of tumor fatty acid metabolism by adiponectin. Adiponectin has been reported to potently suppress breast cancer growth. This study focuses on effects of adiponectin on tumor fatty acid metabolism and provides a novel mechanism for its breast cancer suppressing activity. SIRT-1 plays a central role in metabolic actions of adiponectin in breast cancer cells. On the one hand, SIRT-1 induction leads to downregulation of SREBP-1 by direct deacetylation and destabilization of nuclear SREBP-1 or by suppressing SREBP-1 expression through inhibition of mTOR signaling. SREBP-1 suppression leads to decreased expression of key enzymes in fatty acid synthesis (FAS) pathway and resultant blockage of FAS. On the other, SIRT-1 induction by adiponectin stimulates lipophagy to degrade lipid droplets and promote utilization of fatty acids for energy production via fatty acid oxidation (FAO). Inhibition of FAS accompanied by elevated FAO result in impairment in cellular fatty acid pool, which in turn causes disruption of lipid rafts and raft-dependent signal transduction, and cell apoptosis as a final consequence