1. IntroductionChronic myeloid leukemia (CML), the first human malignancy associated with a single acquired genetic defect, is well-known leukemia identified by the Philadelphia (Ph+) chromosome containing the Bcr-Abl1 fused oncogene [1]. This oncogene encodes the BCR-ABL1 oncoprotein, a constitutively active tyrosine kinase that activates multiple signaling pathways, ultimately leading to aberrant proliferation in CML [2,3]. Many tyrosine kinase inhibitors (TKIs) were developed to restrict this kinase, with imatinib being the most effective [4]. Even though imatinib was a highly effective drug for CML treatment, some patients acquired resistance after the initial response [5]. Several initiatives have been launched to resolve this issue, one of them being translational inhibition by microRNAs (miRNAs). Past in vitro and in vivo studies based on CML treatments have also shown that resistance to imatinib therapy can be reversed with miRNA transfection [6,7,8]. To thoroughly investigate how miRNAs regulate tumorigenesis, a metabolomic-based methodology is being incorporated into cancer therapeutics-based studies. Metabolomics is the scientific study of chemical processes involving metabolites, small molecule substrates, and intermediate products of cell metabolism. In the case of cancer, detailed analysis of altered metabolites may yield biomarkers for diagnostic and therapeutic purposes [9,10]. Something similar was observed in the case of CML, where metabolomics is currently the focus of intensive research, a panel of specific metabolites proved effective in monitoring the efficacy of TKI therapy [11,12].The most abundant antioxidant metabolism found in living organisms is glutathione (GSH) metabolism, which maintains cellular redox homeostasis. The key player in this metabolism is Glutathione (GSH) which maintains adequate cysteine levels, detoxifies xenobiotics, and regulates transcription factors related to redox signaling [13]. This antioxidant metabolite also confers therapeutic resistance to cancer cells, and GSH overexpression eventually promotes tumor progression [14]. This statement is a little perplexing because glutathione normally maintains cell health by regulating cellular homeostasis. However, tumor cells are different because they have highly reduced intracellular environments that are accompanied by significant levels of reduced glutathione (GSH) (intracellular level of GSH is 90–95% of its total concentration), so investigation into GSH metabolism would aid in understanding the role of GSH and its associated metabolites in the development of chemotherapeutic resistance in different tumors [14,15,16]. Not only GSH is responsible for chemoresistance in the GSH metabolism, but amino acids and inorganic compounds involved in this cellular metabolism also play a significant role [17]. These metabolites support the acquisition of anticancer resistance by providing essential building blocks for biosynthetic pathways and maintaining a balanced redox state, as well as by supporting the reduced intrinsic vulnerability of cancer stem cells to antineoplastic therapy [18]. Additionally, chemoresistance-based studies have reported that alterations in the abundance of amino acids and inorganic compounds related to GSH metabolism are frequently associated with resistance to drugs such as bortezomib, sorafenib, and cisplatin [19,20,21,22].Remarkably, microRNAs (miRNAs) have been discovered to be potent regulators, either negatively or positively, of various cellular metabolisms in a variety of cancers that ultimately affect cell proliferation [23]. In the case of GSH metabolism too, it gets regulated by several miRNAs such as miR-17-92, miR-125b, miR-126, miR-128a, miR-141, miR-200, miR-21, miR-34a, miR-210, miR-122, miR-335, and miR-320 [24,25,26,27,28,29,30,31]. Hence, in the context of the aforementioned investigations, the current study examined whether the expression of hsa-miR-203a-5p, that have previously been shown to overcome imatinib resistance in leukemic cells [32], could reconfigure the altered GSH metabolism of imatinib-resistant K562 cells (R). 2. Materials and Methods 2.1. Chemicals

RPMI-1640 cell culture media (HiMedia, Thane, India), Fetal bovine serum (FBS) (Gibco, Brough, UK), Antibiotic-Antimycotic (Gibco, Waltham, MA, USA), Imatinib (Sigma-Aldrich, Beijing, China), Phosphate buffer solution (PBS) (Gibco, Waltham, MA, USA), Trypsin (Hi-media, Thane, India), miRNA mimic: hsa-miR-203a-5p (Sigma-Aldrich, St. Louis, MO, USA), LipofectamineTM Transfection Reagent (Invitrogen, Waltham, MA, USA), Methanol (Fisher Scientific, Mumbai, India), Norvaline (Sigma-Aldrich, St. Louis, MO, USA), Dichloromethane (ThermoFisher Scientific, Scoresby, VIC, Australia), N-trimethylsilyl-N-methyl trifluoroacetamide (MSTFA) (Sigma-Aldrich, St. Louis, MO, USA). All compounds were of analytical grade and thus diluted in ultrapure water unless otherwise specified.

2.2. Cell CultureThe human CML blast crisis cell line K562 (acquired from NCCS Pune, India) was grown in RPMI-1640 culture medium supplemented with 1% penicillin-streptomycin and 10% FBS and was kept in an incubator (at 37 °C under 5% CO2). When sufficient numbers of cells were developed, some of the cells subsequently developed into imatinib-resistant K562 (IR-K562) cells, as described in a previously published study from our laboratory [33]. After successfully establishing IR-K562 cells, hsa-miR-203a-5p was transfected into imatinib-resistant K562 (IR-K562/R) cells using Lipofectamine transfection reagent according to the protocol outlined by the manufacturer. The cells were harvested 24 h post-transfection. The following three sample sets were utilized for the metabolic profiling: (1) K562 cells (C), (2) IR-K562 cells (R), and (3) IR-K562 cells transfected with hsa-miR-203a-5p (R + miR-203). 2.3. Metabolite Extraction

All K562 cell lines were centrifuged, and the culture media was disposed of. After that, the cells were gently washed twice with PBS before being resuspended in ice-cold methanol. The supernatants of different cell sets were separated in a fresh falcon as it contained the metabolite extract after centrifugation at 3000× g at 4 °C for 10 min. The falcons were stored at −80 °C until further analysis.

2.4. Preparation of Samples for GC-MS AnalysisThe sample preparation for GC-MS analysis was carried out as explained in previous studies [34,35]. Stored −80 °C falcons were taken out and thawed at room temperature, after which norvaline (1 mg/mL) was added as an internal standard to methanol containing cellular extracts of C, R, and R + miR-203; and the entire solution was transferred to glass vials. The glass vial solutions were then gently dried in an Eppendorf Vacufuge® plus Vacuum Concentrator with a gentle stream of nitrogen gas.Subsequently, the derivatization process was carried out to volatilize and polarize metabolites such as carbohydrates, steroids, amino acids, and fatty acids [36]. Fifty microliters of dichloromethane (DCM) plus 50 μL of MSTFA (derivatization reagent) was added to the residue left after evaporation in the vacuum concentrator during the derivatization process. The vials were then vortexed and heated at 80 °C for 30 min. Then, 50 μL of the cooled derivatized solutions containing metabolite extracts from different sets of K562 cells was transferred into screw-top autosampler glass vials for GC-MS analysis. 2.5. Instrumentation and Chromatographic Parameters

The metabolic profile of all samples (C, R, and R + miR-203) was obtained using GCMS-QP2010 Ultra (Shimadzu) at the Central Instrumentation Laboratory (CIL), Central University of Punjab, Bathinda. The column temperature was set to 70 °C for 2 min, then ramped at 15 °C/min to 250 °C, held for 2 min, and at last, raised by 10 °C/min to 300 °C, then withstood for 8 min, for a complete run time of 29 min. The samples (1 μL) were introduced in the instrument in the split mode of a 10:1 ratio with the injector temperature set at 250 °C and the flux rate of the carrier gas, helium, as 1 mL/min. The temperature of the transfer line was 250 °C, that of the manifold was 40 °C and the ion source had a temperature of 200 °C.

The electron multiplier was set to relative mode with auto-tune, and the ionization was accomplished with a 1.02 kV electron beam. Electron Impact (EI) mode was active on the MS detector. The mass spectra were recorded at a rate of six scans per second over a range of 50–1000 m/z. The intensity and resolution of chromatographic peaks in all samples were investigated using norvaline as an internal standard.

2.6. Data Acquisition for GC-MSData files from GC-TQ-MS were extracted as Computable Document Format (CDF) files. To make the instrumental data set more comprehensible for data analysis, any biases such as background noise and retention time (RT) fluctuations were eliminated to make the instrumental data set more understandable for data analysis. For data pre-processing, MZmine 2.21 software was used [37]. The parameters were chosen for this procedure as: RT range 3.0–29.0 min; m/z range 50–600; MS data noise level 2.5 × 104; m/z tolerance 0.5; chromatogram baseline level 2.0 × 104; peak duration range 0.05–0.50 min. 2.7. Identification of Metabolites

The National Institute of Standards and Technology Mass Spectral Library (NIST17M1.Lib) database was used for the identification of specific compounds. In addition, the name, molecular weight, and CAS number of each compound were ascertained by NIST WebBook and PubChem.

When metabolites were identified, their area under curve (AUC) values were calculated and the data were normalized by dividing the area of each metabolite by the area of norvaline (internal standard). Eventually, the normalized AUC values were considered as the concentration of each metabolite; and metabolic profiling and statistical analysis of each sample (C, R, and R + miR-203) were performed afterward. SMPDB (Small Molecule Pathway Database), Human Metabolic Database (HMDB), and Kyoto Encyclopedia of Genes and Genomes were used to obtain data for pathways and metabolites (KEGG).

2.8. Analysis of Metabolic PathwaysMetabolite Set Enrichment Analysis (MSEA), notably pathway over-representation analysis (ORA), was executed on the GC-MS metabolite data using the free online software MetaboAnalyst 5.0 [38]. This was used to identify biologically significant patterns in resistant cells and to verify which metabolic pathways were altered and contributed the most to the distinction between different K562 cell sets (C, R, and R + miR-203). 2.9. Statistical Analysis

Following enrichment analysis, the retrieved data was statistically analyzed using GraphPad Prism version 5 (GraphPad Software, San Diego, CA, USA).

4. DiscussionDespite significant therapeutic advances, drug resistance remains the primary obstacle to effective cancer treatment, and the same holds for imatinib in the case of CML patients. Many factors contribute to drug resistance, one of which is cellular metabolism, which is generally beneficial to normal cells but causes tumorigenesis in cancer cells after being altered by overstimulated growth factors. This altered metabolism along with mitochondrial dysfunction, and persistent genetic alterations in tumor cells causes a high yield of reactive oxygen species (ROS), resulting in the accumulation of oxidized proteins, DNA, and lipids [46]. As a result, cancer cells have elevated levels of ROS-scavenging molecules as an adaptive response, with glutathione (GSH) being the most effective free radical scavenger. This scavenging molecule maintains a highly reducing milieu in the cellular compartments (cytosol, nucleus, mitochondrial matrix, and peroxisome) to facilitate proper folding and activity; furthermore, GSH prevents apoptosis by reacting with ROS in mitochondria [14]. Other than this, high GSH levels in cancer cells also incentivize metastasis and contribute to the emergence of chemotherapy drug resistance [47,48]. Since miRNAs play a role in the regulation of cellular metabolism by directly targeting enzymes or indirectly by modulating the expression of transcription factors [49,50], a miRNA-based approach to restrain metabolisms may be expected to overcome imatinib resistance.Taking into account all of these facts, the current study was conducted in the anticipation that hsa-miR-203a-5p would reduce GSH metabolism to overcome imatinib resistance, as the GC-MS results demonstrated that GSH metabolism is prominent in imatinib-resistant leukemic (IR-K562/R) cells. L-alanine, 5-oxoproline/pyroglutamic acid, L-glutamic acid, glycine, and phosphoric acid (Pi), which were identified through MetaboAnalyst 5.0 to be related to GSH metabolism, were found to be differentially expressed among different K562 cell sets (Figure 4). L-alanine, the metabolite that enters the cell and initiates GSH metabolism, has been recognized as a new cancer-associated biomarker because of its elevated levels in malignant cells [51,52]. Since L-alanine is a glucogenic amino acid, it contributes to gluconeogenesis and meets the immediate energy requirements of cancer cells whenever its levels increase in cancer patients [53,54]. Higher concentrations of pyroglutamic acid, also known as 5-oxoproline, signal upregulated GSH metabolism and oxidative stress [55]. Oxidative stress, which means ROS generation, has long been involved in cancer as tumor cells use ROS to activate protumorigenic signaling processes that support cancer cell growth, survival, angiogenesis, and metastasis [56,57]. Adenosine triphosphate (ATP), the driving force behind all metabolisms, leads to the generation of phosphoric acid (Pi) while regulating the autophosphorylation of signaling pathway components [58]. Phosphoric acid (Pi) maintains an acidic environment in cancer cells, which promotes immunosuppression and unregulated cancer cell proliferation [59]. The increased production of phosphoric acid (Pi) in resistant cells indicated that ATP catabolism occurred more rapidly in resistant cells than in control cells, suggesting that glutathione metabolism was speeding up to meet the reduced milieu demand of resistant leukemic cells.Since GSH (L-glutamyl-L-cysteinyl-glycine) is a tripeptide that incorporates L-glutamic acid and glycine, the rate of GSH synthesis is influenced by these amino acids and both are well-studied GSH metabolites in the context of cancer [60]. The determined values of L-glutamic acid and glycine in the current study assisted in differentiating the metabolic profiles of imatinib-sensitive, imatinib-resistant, and miRNA-transfected leukemic K562 cells and better understanding the cause of drug resistance. In tumor cells, in addition to participating in GSH metabolism, L-glutamic acid also serves as a nitrogen source in cancer cells alongside L-glutamine [56]. An intriguing fact about L-glutamic acid is that all of its derivatives have the potential to be used therapeutically, and they all behave in a way that is completely different from how L-glutamic acid causes tumorigenesis [55,61]. As a result, L-glutamic acid, its conjugated derivatives, and glutathione metabolism have suddenly become the target of cancer-related research. Glycine, an essential component of GSH, has been identified as a new target for cancer therapeutic interventions as this amino acid contributes to the biosynthesis of purines, and also its catabolism encourages tumorigenesis and malignancy [62]. It has been postulated that glycine availability is the limiting factor in glutathione synthesis [63]. Since cancer cells rely more on L-glutamic acid and glycine, these two glutathione-related metabolites have been identified as possible targets.Overall, this study uncovered a strong correlation between glutathione metabolism and the emergence of resistance. We can say this with assurance because results are consistent with a previous study by Kalinina et al., which proved that the emergence of cellular drug resistance against doxorubicin (DOX) in erythroleukemic K562 cells was caused by elevated levels of antioxidant enzymes and GSH [64]. Besides that, it was exposed that imatinib resistance could be overcome by lowering the levels of the five aforementioned metabolites through miRNA transfection, as their concentrations in hsa-miR-203a-5p transfected resistant K562 cells (R+203) were equivalent to those in imatinib-sensitive K562 cells (C).