Identification of the metabolic alterations associated with the multidrug resistant phenotype in cancer and their intercellular transfer mediated by extracellular vesicles

Abstract

Multidrug resistance (MDR) is a serious obstacle to efficient cancer treatment. Overexpression of P-glycoprotein (P-gp) plays a significant role in MDR. Recent studies proved that targeting cellular metabolism could sensitize MDR cells. In addition, metabolic alterations could affect the extracellular vesicles (EVs) cargo and release. This study aimed to: i) identify metabolic alterations in P-gp overexpressing cells that could be involved in the development of MDR and, ii) identify a potential role for the EVs in the acquisition of the MDR. Two different pairs of MDR and their drug-sensitive counterpart cancer cell lines were used. Our results showed that MDR (P-gp overexpressing) cells have a different metabolic profile from their drug-sensitive counterparts, demonstrating decreases in the pentose phosphate pathway and oxidative phosphorylation rate; increases in glutathione metabolism and glycolysis; and alterations in the methionine/S-adenosylmethionine pathway. Remarkably, EVs from MDR cells were capable of stimulating a metabolic switch in the drug-sensitive cancer cells, towards a MDR phenotype. In conclusion, obtained results contribute to the growing knowledge about metabolic alterations in MDR cells and the role of EVs in the intercellular transfer of MDR. The specific metabolic alterations identified in this study may be further developed as targets for overcoming MDR.

Figure 1. GO analysis of all the differentially expressed proteins (DEPs) identified with the Progenesis QI software in both pairs of counterpart drug-sensitive and MDR cancer cell lines: K562 versus K562Dox and NCI-H460 versus NCI-H460/R.

(A) GO - Cellular component analysis of the identified proteins; (B) GO - Molecular functional analysis of the identified proteins; and (C) GO - Biological process analysis of the identified proteins.

Figure 2. Validation by Western blot analysis of some DEPs initially identified by quantitative proteomics.

Representative blots were chosen from three independent experiments. Actin was used as a loading control.

Figure 3. Comparison of GSH and ROS levels between drug-sensitive and MDR cancer cell lines.

(A) GSH levels in K562 and K562Dox cells. (B) GSH levels in NCI-H460 and NCI-H460/R cells. GSH levels are represented as Relative Luminescence Units (RLU). (C) ROS levels in K562 and K562Dox cells. (D) ROS levels in H460 and NCI-H460/R cells. The ROS levels are represented as mean fluorescence. Staurosporine was used as a positive control. Results are the mean ± SEM of 3 independent experiments. *p ≤ 0.05 and **p ≤ 0.01.

Figure 4. Global metabolic differences between the NCI-H460 and NCI-H460/R counterpart cell lines.

Cells were metabolically profiled using Seahorse XF-24 Analyser. (A) Representative results of a glycolysis stress test, which measures extracellular acidification rate (ECAR) following addition of glucose-free media (blue line A), glucose (blue line B), oligomycin (blue line C) and 2-deoxyglucose (blue line D). (B) Representative results of a mitochondrial stress test, measuring the oxygen consumption rate (OCR) in glucose-containing media, following sequential addition of media (blue line A), oligomycin (blue line B), FCCP (blue line C) and rotenone (blue line D). Results are the mean ± SEM from three independent experiments, with four to eight replicates per experiment. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001 of NCI- H460 vs. NCI-H460/R cells.

Figure 5. Alterations in NCI-H460 cell metabolism after treatment with a PPP inhibitor.

NCI-H460 cells were treated with 5 mM DCA for 24 h and subsequently metabolically profiled, using the Seahorse XF-24 Analyser. Data represent mean ± SEM from three independent experiments, with four replicates per experiment. *p ≤ 0.05; **p ≤ 0.01; NCI-H460 vs. NCI-H460+DCA. ECAR - extracellular acidification rate; OCR - oxygen consumption rate.

Figure 6. Alterations in the NCI-H460/R cellular metabolism after treatment with a P-gp inhibitor.

NCI-H460/R cells were treated with 2 μM verapamil for 15 h and subsequently metabolically profiled using the Seahorse XF-24 Analyser for measuring alterations in glycolysis, glycolytic capacity, glycolytic reserve, non-glycolytic acidification, basal respiration and non-mitochondrial respiration. Data represent three independent experiments, with four replicates per experiment. *p ≤ 0.05; **p ≤ 0.01; NCI-H460/R vs. NCI-H460/R+verapamil. ECAR - extracellular acidification rate; OCR - oxygen consumption rate.

Figure 7. Alterations in the resistance level of NCI-H460 and NCI-H460/R cells after treatment with PPP and P-gp inhibitors, respectively.

Cells with or without pre-treatment with DCA or verapamil were treated with 37.5 nM of doxorubicin and the ratio between cells with and without pre-treatment was calculated. (A) NCI-H460 cells were pre-treated with 5 mM DCA for 24 h. (B)- NCI-H460/R cells were pre-treated with 2 μM verapamil for 15 h. Data represent mean ± SEM from three independent experiments. *p ≤ 0.05.

Figure 8. Alterations observed in NCI-H460 cellular metabolism after treatment with EVs shed by MDR cells.

NCI-H460 cells were treated with 18 × 10⁸ EVs for 15 h and subsequently metabolically profiled using Seahorse XF-24 Analyser. The results of glycolysis stress test are shown as ECAR measurements after addition of glucose-free media (blue line A), glucose (blue line B), oligomycin (blue line C) and 2-deoxyglucose (blue line D). (A) NCI-H460 cells were treated with their own EVs and with EVs isolated from NCI-H460/R cells. (B) NCI-H460 cells were treated with EVs isolated from the pair of chronic myeloid leukaemia counterpart cells, K562 and K562Dox. Data represents the results from two independent experiments, with four replicates each. A two-tailed *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001 NCI-H460 vs. NCI-H460+EVs NCI-H460; NCI-H460 vs. NCI-H460+EVs NCI-H460/R; NCI.H460 vs. NCI-H460+EVs K562; NCI-H460 vs. NCI-H460+EVs K562Dox.

Figure 9. Schematic representation of the observed metabolic differences between drug-sensitive cells and their counterpart MDR cells.

P-gp overexpressing MDR cancer cells have various protective metabolic strategies. These include: increasing rates of glycolysis (1) and methylation capacity (2), alterations in the GSH metabolism (3), decreasing rates of PPP (4) and OXPHOS (5) and finally changing the phenotype of the surrounding drug-sensitive cells by EVs-mediated transfer of new features (6). Filled lines and arrows: increased pathways; Dashed lines and arrows: decreased pathways; Bold and bigger fonts: increased metabolic processes; Smaller fonts: decreased metabolic processes.