Metabolomic Profiling of the Synergistic Effects of Melittin in Combination with Cisplatin on Ovarian Cancer Cells

Abstract

Melittin, the main peptide present in bee venom, has been proposed as having potential for anticancer therapy; the addition of melittin to cisplatin, a first line treatment for ovarian cancer, may increase the therapeutic response in cancer treatment via synergy, resulting in improved tolerability, reduced relapse, and decreased drug resistance. Thus, this study was designed to compare the metabolomic effects of melittin in combination with cisplatin in cisplatin-sensitive (A2780) and resistant (A2780CR) ovarian cancer cells. Liquid chromatography (LC) coupled with mass spectrometry (MS) was applied to identify metabolic changes in A2780 (combination treatment 5 μg/mL melittin + 2 μg/mL cisplatin) and A2780CR (combination treatment 2 μg/mL melittin + 10 μg/mL cisplatin) cells. Principal components analysis (PCA) and orthogonal partial least squares discriminant analysis (OPLS-DA) multivariate data analysis models were produced using SIMCA-P software. All models displayed good separation between experimental groups and high-quality goodness of fit (R2) and goodness of prediction (Q2), respectively. The combination treatment induced significant changes in both cell lines involving reduction in the levels of metabolites in the tricarboxylic acid (TCA) cycle, oxidative phosphorylation, purine and pyrimidine metabolism, and the arginine/proline pathway. The combination of melittin with cisplatin that targets these pathways had a synergistic effect. The melittin-cisplatin combination had a stronger effect on the A2780 cell line in comparison with the A2780CR cell line. The metabolic effects of melittin and cisplatin in combination were very different from those of each agent alone.

Metabolic Reprogramming of Chemoresistant Cancer Cells and the Potential Significance of Metabolic Regulation in the Reversal of Cancer Chemoresistance

“…Several natural compounds have been shown to reverse chemoresistance [ 75 , 76 , 77 ]. Ursolic acid (UA) is a natural compound with antitumor activity.…”

“…β-elemene, extracted from Curcuma zedoaria Roscoe , is potential to reverse the resistance to erlotinib [ 78 ]. Melittin, a main peptide with a potential for anticancer activity in bee venom, has a synergistic effect with cisplatin on ovarian cancer cells in several key metabolic pathways, such as TCA cycle, OXPHOS, purine and pyrimidine metabolism, and the arginine/proline pathway [ 77 ].…”

Monitoring for Response to Antineoplastic Drugs: The Potential of a Metabolomic Approach

“…Functional endpoints relevant to a response biomarker would include cell death or a reduction in cell proliferation. Unfortunately, most in-vitro studies on drug-induced changes in the metabolome primarily involve a description of changes in the metabolome that result from the drug itself, or it is unclear whether the changes are pharmacologic or due to impairment of tumor cell expansion [ 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 ].…”

“…There are data on the metabolomic consequences of effective chemotherapy from a variety of cell lines treated with diverse drugs and analyzed using different analytical platforms. Using LC–MS, a fall in intracellular ATP levels and a depletion of NAD+ was observed in association with cell death from melittin and cisplatin in ovarian cancer cells [ 51 ]. Lodi et al, using 1 H-NMR spectroscopy, characterized the metabolic fingerprints of two prostate cancer cell lines and a breast cancer cell line secondary to the pharmacological changes from a PI3K inhibitor and an HSP90 inhibitor, then identified changes that occurred due to loss of cell viability [ 53 ].…”

microbialPhenotypes: An R package that analyzes high-throughput microbial phenotype data

“…If there are data from other research domains that has a similar structure to the example described here, the utility of this package can be much more extended. For example, data from animal cells (Alonezi et al, 2016(Alonezi et al, , 2017 . In addition, knowing that there are resources for complicated machine learning algorithms to do classification of functions, our work here does not aim to improve those methods.…”