Enhanced metabolic activities for ATP production and elevated metabolic flux via pentose phosphate pathway contribute for better CIK cells expansion

Abstract Objective Ex vivo expansion is an effective way to produce cytokine‐induced killer (CIK) cells needed for clinical trials. Here, ex vivo expansion and metabolism characters of CIK cells in static and dynamic cultures and the relationship between cell expansion and metabolism were investigated. Materials and methods Oxygen transfer efficiency was assessed by computational fluid dynamics technique. Cell phenotype, apoptosis and of transporter expression were determined by flow cytometry and Western blotting. Metabolites and enzyme activities were assessed by biochemical methods. Results Dynamic cultures favoured better CIK cell expansion without impairing their phenotype and cytotoxicity, enhanced oxygen transfer efficiency. The glucose metabolism flux of cells in dynamic cultures was enhanced by upregulating surface glucose transporter 1 expression and phosphofructokinase activity. Moreover, pentose phosphate pathway (PPP) metabolic flux was enhanced through upregulating glucose‐6‐phosphate dehydrogenase activity. Glutaminolysis was also accelerated via boosting glutamine transporters expression, glutaminase (GLS) and glutamate dehydrogenase activities. Together with higher oxygen consumption rate and extracellular acidification rate, it was suggested that cells in dynamic cultures were in a more vigorous metabolic state for ATP production. Conclusion Dynamic cultures accelerated glucose and glutamine metabolic flux to promote ATP production, elevated glucose metabolic flux through PPP to promote biosynthesis for better cell expansion. These findings may provide the basis for ex vivo CIK cell expansion process optimization.

2 weeks. 9 The obtained CIK cells are heterogeneous lymphocytes, in which over 90% are CD3 + cells, and can be divided into two main subsets: CD3 + CD56 + cells and CD3 + CD56 -T cells. [10][11][12][13] Due to the limited initial number, CIK cells need to be expanded ex vivo up to 10 10 cells per infusion to meet the clinical requirements. [14][15][16] Moreover, it was reported that the clinical response was related to injected cell numbers [17][18][19] ; therefore, it is necessary to optimize the ex vivo CIK cells expansion process for better clinical efficacy.
For proliferation, CD3 + cells need to enter into an activation state from a quiescent state, because only cells which were fully activated could undergo cell proliferation. 20 The activation process required the production and surface expression of nutrient transporters, costimulatory molecules, etc, all of which require energy. Besides the large energy burden after CD3 + cells were activated, there was also an increasing demand for biosynthetic precursor molecules, 21 which was also an energy-consuming process. 22 Therefore, it is important to provide sufficient energy and promote the biosynthesis of lipids, proteins, nucleic acids and other carbohydrates for the proliferation of CD3 + cells.
Glucose is one of the most important carbon and energy source for CD3 + cells. The uptake of glucose in CD3 + cells is mainly via glucose transporter 1 (GLUT1). [23][24][25] After entering cells, glucose is phosphorylated to glucose-6-phosphate (G6P). G6P could be converted into pyruvate by phosphofructokinase (PFK), pyruvate kinase, etc through glycolysis, it also could be converted to ribose-5-phosphate by glucose-6-phosphate dehydrogenase (G6PDH), etc through pentose phosphate pathway (PPP). Glucose-derived pyruvate either entered into tricarboxylic acid (TCA) cycle or was converted to lactate by lactic dehydrogenase. High-glucose metabolic flux rate allowed rapid macromolecular synthesis and ATP generation, which were all necessary for cell proliferation. 23,26 It had been reported that an almost 20-fold increase was induced in glucose metabolic flux of thymocytes after activation. 23 In addition, glucose deficiency would inhibit T-cell proliferation and survival. 24 In addition to glucose, glutamine is also an important energy source. In human CD3 + cells, there are several well-characterized transporters, such as ASC amino acid transporter 2 (ASCT2), sodium-coupled neutral amino acid transporter 1/2 (SNAT1/SNAT2) and a heterodimer transporter (CD98/large neutral amino acid transporter, LAT1). [27][28][29] After entering cells, glutamine could be converted to α-ketoglutarate by glutaminase (GLS), glutamate dehydrogenase (GDH), etc for TCA cycle replenishment and oxidative phosphorylation (OXPHOS) to produce metabolic intermediates and ATP for cell proliferation. 30,31 Evidence was obtained that glutamine consumption was enhanced after the activation of CD3 + cells. 32 It had also been indicated that T cells were unable to proliferate in glutamine-free medium. 28 Thus, it was necessary to monitor glucose and glutamine concentrations in culture environment and promote their metabolic rates for better cell proliferation during the ex vivo expansion of CIK cells.
For ATP productionvia OXPHOS, oxygen is one of the indispensable factors. It was illustrated that the oxygen consumption rate of activated CD3 + cells was doubled as that of quiescent CD3 + cells. 26,33 Moreover, cell metabolism patterns were closely associated with oxygen. 34

| Cell preparation
The experiments conducted in this study were approved by the Science Ethics committee of the State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology and were in accordance with the guidelines for cellular products research and preparation, China (2016). CIK cells were generated from cord blood mononuclear cells (CBMNCs) of full-term healthy delivery with informed consent. CBMNCs were enriched using density gradient centrifugation on Ficoll/Histopaque (density: 1.077 g/mL; GE Healthcare, New York, NY, USA) and cultured in static and dynamic cultures for the generation of CIK cells as our previous study. 45 Culture supernatants were mixed sufficiently before sampling. Cell numbers were counted every day.
The purity of fresh isolated CD3 + cells was over 90% as assessed by flow cytometry.
Cell preparations were analysed on a flow cytometer (FACS Aria I; BD Bioscience, San Jose, CA, USA) to determine the proportions of CD3 + cells, CD3 + CD56 + cells, CD3 + CD8 + cells and CD3 + CD4 + cells in total cell population. All antibodies were purchases from BD bioscience.
After incubation, the mean fluorescence intensity (MFI) of intracellular 2-NBDG was immediately measured using a ImageStream X Mark II imaging flow cytometer on FITC channel (Merck, Darmstadt, Germany).

| Analysis of glutamine and ammonia concentrations in culture supernatant
Glutamine and ammonia concentrations in culture supernatants were analysed using a Nova BioProfile 400 (Nova Biomedical, Waltham, MA, USA). The kinetics were calculated based on the following equations: Specific glutamine consumption rate: Specific ammonia production rate: where Q gln and q NH4 + was the specific glutamine consumption rate and the specific ammonia production rate of cells, S 1 and P 1 were the concentration of glutamine and ammonia at the time point of t 1 , S 2 and P 2 were the concentration of glutamine and ammonia at the time Nf(t)dt was the time integral of viable cell number and f(t) was fitted to the cell density.

| Enzyme activity
Expanded CIK cells for 7 days or fresh isolated CD3 + cells were analysed for the enzyme activities of PFK, G6PDH, GLS and GDH according to the manufacturers' instruction. All enzyme activity detection assays were purchased from Comin Biotechnology (Suzhou, China). (1)

| CFD modelling
Oxygen mass transfer coefficient (k L ) modelling and numerical strategies were according to previous work of Li et al. 47 The flow field

| Statistics
Values were presented as mean ± standard error. Student's t test or one-way ANOVA was applied to evaluate the significance of differences. P < 0.05 was considered as statistically significant.

Ex vivo expansion characters of cells in static and dynamic cultures
were analysed and shown in Figure 1. After a 14-day culture, the expansion folds of total cells in dynamic cultures were 48.14 ± 9.47 folds, significantly higher than the 9.29 ± 1.69 folds in static cultures ( Figure 1A, P < 0.05). On day 14, the percentages of total apoptotic cells in two cultures were lower than 5%, no significant difference was found between the two cultures ( Figure 1B

| Dynamic cultures provide a better mass transfer environment
Oxygen in the culture environment is an important factor for cell growth and proliferation. Through CFD modelling of k L , the mass transfer differences could be easily observed. As shown in Figure 2, k L was obviously higher in dynamic cultures, illustrating that dynamic cultures enhanced oxygen transfer efficiency and could supply more oxygen into the microenvironment which were beneficial for CIK cell proliferation.

| Enhanced surface GLUT1 expression and glycolytic enzyme activity results in high-glucose uptake activity of cells in dynamic cultures
Different metabolism patterns of CD3 + cells were adopted before and after cytokines stimulation, and cell metabolism patterns were closely associated with cell proliferation 31 ; hence, the glucose metabolism was first investigated to understand the relationship be-

| Upregulated glutamine consumption also contributes for better cell expansion in dynamic cultures
In addition to glucose, glutamine is another important carbon and energy source for cell proliferation. Glutamine also depends on transporters to enter cells. Therefore, four important glutamine transporters of cells in static and dynamic cultures were analysed. Furthermore, the Q gln and q NH4 + of cells in static and dynamic cultures were calculated by Equations (1) and (2) and shown in Figure 4I and J. The Q gln and q NH4 + of cells in dynamic cultures were both significantly higher than those of cells in static cultures (P < 0.05), illustrating the upregulation of glutaminolysis metabolic rate. These results demonstrated that cells in dynamic cultures enhanced glutamine consumption through upregulating the expression of glutamine transporters and activities of GLS and GDH for producing more intermediates that are consumed by biosynthetic processes, supporting the better cell expansion.

| Improved ATP production ability accounts for high cell proliferation ability in dynamic cultures
For proliferation, large amounts of ATP would be needed for the synthesize of biomass, while glucose and glutamine are two major substances of energy source. Considering both glucose and glutamine consumption were enhanced, intracellular ATP content was measured to investigate the energy metabolism of cells in two cultures.
As shown in Figure 5, ATP content in cells of both cultures ascended first then descended, surprisingly, it was found that ATP content in cells of dynamic cultures were similar or a bit lower than those of cells in static cultures (P < 0.05).
Since intracellular ATP level was a result of generation and consumption, we measured ECAR (an indicator of aerobic glycolysis) and OCR (an indicator of OXPHOS) to investigate the ATP production ability of cells in two cultures ( Figure 6A-D). The results showed that compared with fresh isolated CD3 + cells, all ECAR and OCR values were significantly higher in expanded CIK cells ( Figure 6E-G, P < 0.05), demonstrating expanded CIK cells went into a more vigorous metabolic state ( Figure 6H).
In addition, all ECAR and OCR values in expanded CIK cells from dynamic cultures were found significantly higher than those from static cultures ( Figure 6E-G, P < 0.05). The higher ECAR and OCR values indicated that expanded CIK cells in dynamic cultures were in a metabolically higher energy state ( Figure 6H). The ratio of OCR/ECAR was higher in dynamic cultures than in static cultures ( Figure 6I, P < 0.05), indicating that expanded CIK cells were in a higher energy state in dynamic cultures through mechanisms that relied more on mitochondrial metabolism than on glycolysis.
Collectively, these results showed that the two major ATP-generating metabolic ways, glycolysis and OXPHOS, were both enhanced in cells of dynamic cultures. More importantly, cells in dynamic cultures were more oxidative to generate ATP which was needed for the production of biomass.
In addition, NADPH was another important cofactor in T-cell activation, differentiation and proliferation. 48

| D ISCUSS I ON
Immune effector cells-based immunotherapy has become a reality for many diseases. 19 Many strategies had been developed to opti- Dynamic cultures not only provided a more homogeneous environment to increase the cell-to-cell and cell-to-cytokines contact, but also improved oxygen mass transfer to increase the oxygen tension in the microenvironment, which was beneficial for cell activation.
Therefore, it was reasonable to infer that dynamic cultures improved cell expansion via regulating cell activation and cell activation-induced metabolism reprogramming.
Further, it had been noted that metabolism patterns of lymphocytes after cell activation was correlated with the cell proliferation ability. 33 Our results showed that cells in dynamic cultures elevated both glycolysis metabolic rate via upregulating GLUT1expression and PFK activities, which was further demonstrated by 2-NBDG assays and higher ECAR values. This results were consistent with the results of our previous study. 45 Meanwhile, the PPP metabolic rate was also elevated via upregulating G6PDH activities, resulting higher intracellular NADPH level. In addition to glucose, higher F I G U R E 8 Elevated overall metabolic activities of cells in dynamic cultures. Upregulated transporters, enzyme activities, or metabolites were indicated with red arrows glutamine consumption rate of cells was observed in dynamic cultures, which could support T cells for the synthesis of protein, nucleotides and amino sugars, all of which were important for proliferating T cells. 28,55 Moreover, the oxidation of glutamine in TCA cycle could produce ATP, potentially allowing activated T cells to divert glucose to other biosynthetic pathways.
ATP is a key donor that provides energy for cellular processes of CD3 + cells, and it can be produced mainly via two pathways, glycolysis and TCA cycle/OXPHOS. 30 In this study, lower ATP content was found in cells of dynamic cultures. The ATP production ability of cells in dynamic cultures was demonstrated to be improved by the enhanced ECAR and OCR of cells, which were indicators of glycolysis and OXPHOS, respectively. So, the decreased cellular ATP content may result from more consumption for cell proliferation.
In addition, cells in dynamic cultures were more energetic and oxidative with a higher basal OCR value and OCR/ECAR ratio. More importantly, the oxidative ATP turnover was found much higher in

ACK N OWLED G EM ENTS
This work was supported by the Science and Technology Innovation Action Plan of Basic Research, Shanghai, China (15JC1401402).

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflicts of interest with the contents of this article.