NSCLC Cells Resistance to PI3K/mTOR Inhibitors Is Mediated by Delta-6 Fatty Acid Desaturase (FADS2)

Abstract

Hyperactivation of the phosphatidylinositol-3-kinase (PI3K) pathway is one of the most common events in human cancers. Several efforts have been made toward the identification of selective PI3K pathway inhibitors. However, the success of these molecules has been partially limited due to unexpected toxicities, the selection of potentially responsive patients, and intrinsic resistance to treatments. Metabolic alterations are intimately linked to drug resistance; altered metabolic pathways can help cancer cells adapt to continuous drug exposure and develop resistant phenotypes. Here we report the metabolic alterations underlying the non-small cell lung cancer (NSCLC) cell lines resistant to the usual PI3K-mTOR inhibitor BEZ235. In this study, we identified that an increased unsaturation degree of lipid species is associated with increased plasma membrane fluidity in cells with the resistant phenotype and that fatty acid desaturase FADS2 mediates the acquisition of chemoresistance. Therefore, new studies focused on reversing drug resistance based on membrane lipid modifications should consider the contribution of desaturase activity.

Keywords: BEZ235; PI3K/Akt/mTOR pathway; drug resistance; lipid metabolism; non-small-cell lung cancer.

Figure 1

(A) Dose-response curve of the clones treated with BEZ235 at different concentrations. The means and standard deviation (SD) of at least three independent experiments are plotted. Statistical analysis was carried out through the two-way ANOVA test and the Bonferroni post hoc test for multiple comparisons and is reported in Supplemental Table S3. (B) Histograms representing the relative expression of MDR-1 mRNA levels. Actin was used as a housekeeping gene for normalization. The means of three replicates and SD are reported. (C) Histograms representing the relative expression of ABCG2 mRNA levels. Actin was used as a housekeeping gene for normalization. The means of three replicates and SD are reported. (D) Western blot analysis of MAPK and PI3K/mTOR pathway proteins in wt, wt_bez, G12C and G12C_bez clones at basal conditions. Ran was used as a loading control. (E–L) Dose-response curves of the clones treated with PI3K/Akt/mTOR inhibitors at increasing concentrations. On each graph, the specific drug with the corresponding target is reported. The means and standard deviation (SD) of at least three independent experiments are plotted.

Figure 2

(A) Pie chart summarizing the metabolic class distribution of the statistically significant metabolites between the wt and wt_bez clones. (B) Pie chart summarizing the metabolic class distribution of the statistically significant metabolites between G12C and G12C_bez clones. (C) Sum of intracellular concentrations of lysophosphatidylcholines (lysoPC), phosphatidylcholines (PC), and sphingomyelins (SM) in wt, wt_bez, G12C, and G12C_bez clones. ** indicates statistically significant differences with p < 0.01 (two-way ANOVA and Tukey Kramer post-hoc test). (D) Sum and distribution of intracellular concentrations of saturated (SFA), monounsaturated (MUFA), and polyunsaturated (PUFA) lipid species. ** and *** indicate statistically significant differences with p < 0.01 and p < 0.001, respectively (two-way ANOVA and Tukey Kramer post-hoc test). (E) Representative images of lipid droplets detected by Nile Red fluorescent dye (green) in all the clones. DAPI (blue) was used to counterstain nuclei. The scale bar is 20.0 µM. (F) Intracellular concentration of free cholesterol. (G) Membrane fluidity expressed as relative fluorescence units (RFUs) of the ratio between PDA dimer and monomers, detected by a membrane fluidity kit (Abcam, see Section 2). * and ** indicate statistically significant differences with p < 0.05 and p < 0.01, respectively (one-way ANOVA and Bonferroni post-hoc test). LysoPC, Lysophosphatidylcholine, PC, Phosphatidylcholine, PE, lysophosphatidylethanolamine, CE, Ceramide, SM, Sphingomyelin, MG, monoacylglyceride, and TG, Triglyceride.

Figure 3

(A) Western blot analysis of SREBP1 protein expression in wt, wt_bez, G12C, and G12C_bez clones at basal conditions. Ran was used as a loading control. (B) Western blot analysis of SREBP1 protein expression in nuclear (left) and cytosolic (right) extracts of wt, wt_bez, G12C, and G12C_bez clones at basal conditions. Actin was used as a loading control. Lamin B was used as a nuclear extract marker. (C) Western blot analysis of ACC, FASN, and SCD1 protein expression in wt, wt_bez, G12C, and G12C_bez clones at basal conditions. Ran was used as a loading control. (D) Intracellular levels of Palmitic and Oleic acid in wt, wt_bez, G12C, and G12C_bez clones. (E) Histograms representing relative mRNA expression of the ELOVL2 and ELOVL5 genes. Actin was used as a housekeeping gene for normalization. The means of three replicates and SD are reported. (F) Histograms representing relative mRNA expression of the FADS1 and FADS2 genes. Actin was used as a housekeeping gene for normalization. The means of three replicates and SD are reported. (G) Estimated desaturase activity in the clones based on MUFA/SFA and PUFA/SFA ratio. **, ***, and **** highlight statistically significant differences with a p < 0.01, p < 0.001, and p < 0.0001, respectively (one-way ANOVA and Tukey Kramer post-hoc test).