RA and ω-3 PUFA co-treatment activates autophagy in cancer cells

Abstract

Retinoic acid (RA), is a promising therapeutic agent for the treatment of breast cancer. However, metabolic disorders and drug resistance reduce the efficacy of RA. In this study, we found that RA and ω-3 polyunsaturated fatty acids (ω-3 PUFAs) synergistically induced cell death in vitro and in vivo and autophagy activation. Moreover, RA-induced hypercholesterolemia was completely corrected by ω-3 PUFA supplementation. In addition, we demonstrated that the effects of this combination on the autophagic flux were independent of the two major canonic regulatory complexes controlling autophagic vesicle formation. The treatment activated Gαq-p38 MAPK signaling pathways, which resulted in autophagy of breast cancer cells. Knockdown of Gαq or P38 expression prevented RA and ω-3 PUFAs from inducing autophagy. Data indicated that Gαq-p38activation was mediated by the co-activation of GPR40 and RARα in lipid rafts, rather than by the activation of GPR120, RARβ, or RARγ. The results of this study suggest that hyperlipidemic drug side effects may be ameliorated by the administration of ω-3 PUFAs. Thus, the therapeutic indexes of the corresponding drugs may be increased.

Keywords: autophagy; breast cancer; retinoic acid; ω-3 PUFAs.

Figure 1

Growth inhibition in three breast cancer cell lines treated with RA and ω-3 PUFAs (A-D) cells treated with RA and/or ω-3 PUFAs for 72 h. (A): Cell counts after treatment with ω-3 PUFAs. (B): Cell counts after treatment with RA. (C): Cell counts after combination treatment. (D): Cell viability after combination treatment. Data are given as mean ± SEM. ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001.

Figure 2. Autophagy induced by treatment with RA and ω-3PUFAs

(A): Cells were treated with RA or ω-3 PUFAs at indicated concentrations for 24h. Cell extracts were prepared and subjected to western blotting analysis. (B): Cells were treated with RA(20μM) or ω-3 PUFAs(80μM) alone or in combination for 24h.Cell extracts were prepared and subjected to western blotting analysis. (C): Cells were treated with RA(20μM) plus ω-3 PUFAs(80μM) with or without CQ(5μM) for 24h.Cell extracts were prepared and subjected to western blotting analysis. (D): Cells were treated with RA(20μM) + EPA(80μM) for 24h prior to examination by TEM. Electron micrographs of MCF-7, SKBR-3, MDA-MB-231, control (C), RA-treated (R), EPA-treated (E), and RA-plus-EPA-treated (E+R) cells. Magnified images of the boxed regions of i showing autolysosomes orautophagosomes(magnified) (red arrows).

Figure 3. Autophagy induction by RA and ω-3 PUFA treatment is independent of two classical pathways

(A): Cells were treated with RA(20μM) or ω-3 PUFAs(80μM) alone or in combination for 24h. Cell extracts were prepared and subjected to western blotting analysis. (B): Cells were treated with RA(20μM) +ω-3 PUFAs(80μM) with or without insulin (2μM) for 24h.Cell extracts were prepared and subjected to western blotting analysis. (C): Cell morphology of MCF-7 cells treated with RA(20μM) + ω-3 PUFAs(80μM) with or without insulin (2μM) for 24h. (D): Cell counts of MCF-7 cells treated with RA(20μM) plus ω-3 PUFAs(80μM) with or without insulin (2μM) for 24h. (E): Cells were treated with RA(20μM) or ω-3 PUFAs(80μM) alone or in combination for 24h.Cell extracts were prepared and subjected to western blotting analysis. (F): Cells were treated with RA(20μM) +ω-3 PUFAs (80μM) with or without 3-MA (5mM) for 24h.Cell extracts were prepared and subjected to western blotting analysis. (G): Cell morphology of MCF-7 treated with RA(20μM) +ω-3 PUFAs(80μM) with or without 3-MA(5mM) for 24h. (H): Cell counts of MCF-7 treated with RA(20μM) plus ω-3 PUFAs(80μM) with or without 3-MA(5mM) for 24h. Data are given as mean ± SEM. ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001.

Figure 4. Autophagy induction by RA and ω-3 PUFA treatment is dependent on Gαq-P38 activation

(A): Cells were treated with RA(20μM) or ω-3 PUFAs(80μM) alone or in combination for 24h. Cell extracts were prepared and subjected to western blotting analysis. (B): Cell counts of breast cancer cells treated with RA(20μM) plus ω-3 PUFAs(80μM) with or without P38-knockdown for 24h. (C): Cells were treated with RA(20μM) +ω-3 PUFAs(80μM) with or withoutP38-knockdown for 24h.Cell extracts were prepared and subjected to western blotting analysis. (D): Cells were treated with RA(20μM) + EPA(80μM) with or without Gαq-knockdown for 24h.Cell extracts were prepared and subjected to western blotting analysis. (E): MCF-7 cells were pretreated with the indicated chemical inhibitors for 30min, followed by 15 min treatment with RA(20μM) + EPA(80μM).Cell extracts were prepared and subjected to western blotting analysis. Data are given as mean ± SEM. ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001.

Figure 5. The combination of RA and ω-3 PUFAs induces Gαq-P38 activation through RARα and GPR40

(A): Cells were treated with RA(20μM) + EPA(80μM) with or without GPR120-knockdown for 15 min. Cell extracts were prepared and subjected to western blotting analysis. (B): Cells were treated with RA(20μM) + EPA(80μM) with or without GPR40-knockdown for 15 min. Cell extracts were prepared and subjected to western blotting analysis. (C): Cells were treated with RA(20μM) + EPA(80μM) with or without RARα-knockdown for 15 min. Cell extracts were prepared and subjected to western blotting analysis. (D): Cells were treated with RA(20μM) + EPA(80μM) with or without RARβ-knockdown for 15 min. Cell extracts were prepared and subjected to western blotting analysis. (E): Cells were treated with RA(20μM) + EPA(80μM) with or without RARγ-knockdown for 15 min. Cell extracts were prepared and subjected to western blotting analysis. (F): MCF-7 cells were treated with RA(20μM) + EPA(80μM) and their extracts fractionated using an iodixanol density gradient, as described in the Materials and Methods section. Each fraction was subjected to SDS-PAGE and immunoblot analysis using antibodies against the indicated proteins. (G): MCF-7 cells were pretreated with the indicated concentrations of methyl-β-cyclodextrin (MβCD) for 1 h, followed by 15 min treatment with RA(20μM) + EPA(80μM). Cell lysates were prepared and subjected to SDS-PAGE and immunoblot analysis. (H): MCF-7 cells were pretreated with the indicated concentrations of methyl-β-cyclodextrin (MβCD) for 1 h followed by 24h treatment with RA(20μM) + EPA(80μM), and then subjected to cell counts.

Figure 6. The effects of RA and EPA alone or in combination on tumor growth in vivo

(A): Effects of ω-3 PUFA and RA treatments on tumor growth in a xenograft model. Tumors were measured at the indicated time points. N=10 per group. (B): Effects of ω-3 PUFA and RA treatments on body weight. (C): Typical size and macroscopic appearance of tumors. (D): Tumor weight in each group. (E): Representative images of IHC staining of Ki67 and Ki67-positivecell numbers. The results represent the mean percentage of Ki67-positive cells relative to that of the control. (F-G): Western blot analysis of xenograft models treated with RA and EPA alone or in combination. (H): Tumor tissues were fixed in 2% glutaraldehyde solution and were then examined by TEM. Electron micrographs of control (C), RA treated (R), EPA treated (E), and RA + EPA treated (E+R) tissues. Magnified images of the boxed regions of d showing autolysosomes or autophagosomes (magnified) (red arrows). (I): TG, TC, HDL-c, and LDL-c levels were determined by a programmable automatic biochemical analyzer (Mindray, BS480, China) according to the manufacturer’s instructions. Data are given as mean ± SEM. ^*p < 0.05, ^**p < 0.01, and ^***p < 0.001.

Figure 7. Signal transduction pathway