3.3 Assessing the intracellular metabolic changes of Hela cells exposed to DOX
The above results arrive at the conclusion that N2-A has been identified as the most optimal method to acquire intracellular metabolites with minimal loss during Hela carcinoma sample preparation. Metabolomics is emerging as an important tool for understanding the molecular mechanisms underlying the response of drug delivery systems. Further, in this study, a case study with the acquired intracellular metabolomics data from Hela carcinoma cell under both 2D monolayer culture and 3D MTS in response to anticancer drug DOX treatment was illustrated using this optimal sample preparation protocol. DOX is a kind of anti-tumor antibiotics widely used in clinics and prevents the resealing of DNA double helix and subsequently creates a double stranded DNA break which leads to cell death [62-64]. In our study, we focused on the intracellular metabolite changes of Hela cells exposed to DOX. We determined the cytotoxicity of the drug to 2D monolayer culture and 3D MTSs through cell counting kit-8 (CCK-8). As shown inFigure S7 , compared with the 2D control, 3D MTSs showed higher resistance to DOX treatment. After 48 hours of DOX treatment, the half maximal inhibitory concentration (IC50) values of 3D MTSs and 2D monolayer culture cells were 13.13 μM and 1.108 μM, respectively. As expected, the result indicated that 3D Hela MTSs were less sensitive to DOX than 2D monolayer cells.
As shown in Figure 10 , extracellular concentrations of glucose, lactic acid, glutamine and ammonia were measured. It has been acknowledged that cancer cells rely on aerobic glycolysis instead of mitochondrial respiration for the rapid provision of ATP and precursors, a phenomenon termed “the Warburg effect” [65]. Apparently, 3D cells consumed more glucose for aerobic glycolysis and produced a larger amount of lactate as compared to 2D cells (Figure 10 ). Therefore, 3D MTSs conserved a more tumor metabolic phenotype in terms of Warburg effect. After adding 1μM DOX in 2D cells, extracellular glucose was hardly consumed, and the level of lactate remained basically unchanged, which was around 0.2 g/L with the dosing time. By contrast, the ability of 3D MTSs to consume glucose and produce lactic acid was significantly stronger than 2D cells. This might indicate that the addition of DOX had a greater impact on the respiration capacity of 2D cells than 3D cells. In addition, it was found that when glucose and glutamine were present at the same time, cancer cells preferably used glutamine first. The addition of DOX did not change the utilization of glutamine in both 2D and 3D cells. Regarding ammonia metabolism, the ammonia produced by 2D cells was gradually accumulated, while 3D cells accumulated ammonia in the early stage of culture, and re-consumed ammonia in the later stage of culture. This seemed reasonable because metabolic recycle of ammonia has been reported to favor tumor cell growth [6].
After the DOX treatment, the hierarchical cluster analysis of the acquired intracellular metabolites showed that the treatment on 3D MTSs elicited more pronounced metabolic changes than 2D monolayer cultures (Figure 11 ). Specifically, 2D monolayer cells before and after dosing showed significant regulation on 13 metabolites among which G3P, proline (Pro) and pyruvate (PYR) were upregulated while glycine (Gly), ribose-5-phosphate (R5P), serine (Ser), nicotinamide adenine dinucleotide phosphate (NADP), glutamine (Gln), NAD, cysteine (Cys), aspartate (Asp), threonine (Thr), and malate (MAL) were down-regulated. In contrast, 24 metabolites were significantly changed in 3D MTS after dosing with DOX, where citrate (CIT), G3P, Ser, Cys, Ala, Thr, glutamine, valine, and lysine were up-regulated while PYR, Glu, 3PG, succinate (SUC), NAD, 6PG, NADP, leucine (Leu), AMP, ATP, AKG, Pro, Asn, E4P, and PEP were down-regulated. Armiñán et al. found that 2D MCF7 cells exposed to free DOX reduced the intracellular levels of Gly, NAD and Asp, which was consistent with our findings in 2D Hela cells [66]. Previous studies have also shown that reduced levels of Gly (2D) and Glu (3D) were associated with reduced glycolysis [67], and Gly in particular has been considered as a biomarker for cancer prognosis and treatment response [68]. To corroborate this, extracellular metabolite data revealed that glucose uptake and lactate production were inhibited in both 2D and 3D cells after exposure to DOX, which also indicated a reduced glycolysis (Figure 10 ). Interestingly, the intracellular Gln pool showed different trends in 2D and 3D cells after dosing with DOX. Glutamine was an important source of carbon in cells. Glutaminolysis was necessary for cancer cells to satisfy the need to replenish TCA cycle intermediates and to produce NADPH [69]. For 2D cells, the addition of DOX reduced the glycolytic flux and the intracellular glutamine concentration, indicating that the cells may be insufficient to replenish the TCA cycle through glutamine, resulting in the death of 2D cells; while for 3D cells, intracellular glutamine concentration was observed increased, indicating that glutamine was sufficient to maintain the replenishment requirements of the TCA cycle when the cells were reduced in glycolysis.
As shown in Figure 12 , pathway enrichment analysis using targeted metabolomics data showed that differential pathways for 2D and 3D cells after doses of DOX were associated with redox stress, such as glutathione metabolism (2D: Glycine/NADP+/L-Cysteine/Acetyl-CoA/L-Glutamate down-regulated, 3D: NADP+/L-Glutamate down-regulated, L-Cysteine up-regulated), purine metabolism (2D: D-Ribose 5-phosphate/L-Glutamine down-regulated, 3D: ADP/AMP/ATP down-regulated, L-Glutamine up-regulated), nicotinate and nicotinamide metabolism (2D: NADP+/ L-Aspartate down-regulated, 3D: NADP+/NAD+ down-regulated). Oxidative stress signifies the imbalance between the production of reactive oxygen species (ROS) and antioxidant defenses, which could lead to apoptosis and is accompanied by a lot of metabolic alterations. Metabolic changes associated with redox stress, including downregulation of Gly (2D) and ATP (3D), have been implicated in oxidative stress-induced aging [70]. In addition, previous study has also showed that there appeared to be a link between the reduction of NAD levels and the induction of oxidative stress [71]. To verify this, we also measured ROS of Hela cells (2D cells and 3D MTSs) before and after dosing with DOX (Figure S8 ). Our results were consistent with the putative DOX targeting mechanism of action and the increase in redox stress where ROS has been confirmed by previous studies as the main cause of cytotoxicity [62]. In addition to this, the results also showed that DOX exposure would significantly affect amino acid metabolism-related pathways (Glycine, serine and threonine metabolism, cysteine and methionine metabolism, arginine and proline metabolism, and alanine, aspartate and glutamate metabolism) in both 2D and 3D cells. Triba et al. observed the simultaneous decrease of glutamine and alanine after DOX-treated in B16-F10 cells which was interpreted as a drug-induced transient switch from glycolysis to oxidative phosphorylation to supply ATP, and further initiate apoptosis [72]. Previous studies have also reported that apoptosis might be associated with increased levels of branched-chain amino acids (valine, leucine, and isoleucine) and decreased levels of alanine [66]. Finally, the metabolomics data of this study also indicated the common pathway enrichment, which had been reported on other drugs, such as purine and pyrimidine metabolism. For example, Rusz et al. found that exposure to oxaliplatin affected purine metabolism and pyrimidine synthesis, which was in accordance with the ribosome biogenesis stress recently proposed as the primary cause for the cytotoxic effects [51].