Optimization of Culture Media for T Cell Expansion: T Cell Expansion Is a Bottle Neck in Adoptive T Cell Therapy

Adoptive T cell therapy is a promising treatment strategy for cancer immunotherapy. The methods used for the expansion of high numbers of T cells are essential steps for adoptive cell therapy. In this study, we evaluated the expansion, proliferation, activation, and anti-tumor response of T lymphocytes, in presence of different concentrations of interleukin-2, phytohemagglutinin, and insulin. Our results showed that supplemented culture media with an optimized concentration of phytohemagglutinin and interleukin-2 increased total fold expansion of T cells up to 500-fold with about 90% cell viability over 7 days. The quantitative assessment of Ki-67 in expanded T cells showed a signicant elevation of this proliferation marker. In addition, the proportion of CD4+ and CD8+ cells were evaluated using ow cytometry, and data showed that both cells were present in the expanded population. Finally, we assessed the activation and tumor cytotoxicity of expanded T cells against target cells. Overexpression of CD107a, as a functional marker of T cell degranulation on expanded T cells and their ability to induce cell death in tumor cells, was observed in the co-cultured experiment. Based on these data we have developed a cost-effective and rapid method to support the ecient expansion of T cells for adoptive cell therapy. integrate local and global cues to discriminate between structurally similar antigens.


Introduction
The potential of Adoptive T cell therapy (ACT) through stimulating the immune system to recognize and destroy tumor cells is emerging as a promising treatment strategy in the eld of cancer immunotherapy.
The results of clinical studies have shown that adoptive cell therapy using T lymphocytes from peripheral blood mononuclear cells, genetically engineered T lymphocytes, and tumor-in ltrating T lymphocytes, is a remarkable strategy for the treatment of various cancers [1,2]. A critical step for cell therapy using T lymphocytes in particular chimeric antigen receptor T cells is isolation and large-scale ex-vivo expansion of T cells with tumor cytotoxic phenotype [3].
In general, the frequency of tumor-in ltrating T lymphocytes and tumor-speci c T cells in peripheral blood is low and leads to a reduction in the e cacy of adoptively transferred T cells. In order to obtain an adequate and su cient number of T lymphocytes for infusion, ex-vivo culture, expansion and activation of T lymphocytes are required. In this approach, T lymphocyte populations are expanded to improve the functional properties of cytotoxic CD8 + cells, CD4 + cells as well as memory T cells pre-infusion [4,5].
Large-scale production of human T cells is a complex process that required ne-tune optimization in order to obtain long-term in-vivo survival of T lymphocytes with desired phenotype and function [6].
Studies in cancer-targeting T cell therapy have often focused on isolation and expansion of cytotoxic T cells to increase the e ciency of tumor cell killing. However, there is growing evidence for the importance of less-differentiated T memory (TM) cells as well as CD4 + T cells in therapeutic T cell products. CD4 + T cells provide a supporting role in the proliferation, activation, cytotoxic function, and recruitment of CD8 + cells to the tumor milieu as well as directly inhibiting tumor growth [7,8].
In recent decades, many efforts have been made to optimize isolation and non-antigen speci c expansion of cytotoxic T cells for adoptive T cell therapy. While antigen-presenting cells, especially dendritic cells, are potent initiators of T-lymphocyte activation and immune response, there are many obstacles to the utilization of these cells at clinical grade due to their long-term and costly culture process [9,10]. Furthermore, the T cell population obtained from these methods comprises a high percentage of cells that are in the nal stage of differentiation, which is associated with low survival rates in vivo. In this regard, proliferation methods should contain substances that contribute to cell growth and activation in the shortest possible time [11,12].
Mitogenic molecules such as phytohemagglutinin (PHA), interleukin-2, and concanavalin A (Con A) are used widely for in vitro T lymphocyte stimulation and expansion [13][14][15]. On the other hand, the mitogenic properties of insulin, insulin receptor signaling, and insulin analogues were documented and insulin has been shown to stimulate in vitro T cell proliferation when supplied exogenously [16].
The main aim of this study is to optimize the expansion and proliferation conditions of T lymphocytes to achieve a simple, cost-effective, and rapid method as well as a su cient number of T lymphocytes with anti-tumor activity for adoptive cell therapy in cancer research.

Isolation of PBMC and T-cell culture
Informed consent was obtained from ve healthy donors prior to blood collection and the study and the use of human material were approved by the institutional review boards at the Isfahan University of Medical Science (IR.MUI.RESEARCH.REC.1398.788). Peripheral blood mononuclear cells were isolated from heparinized blood using the Ficoll density gradient. Cells were incubated for 24 hours in complete RPMI-1640 medium (BioIdea-Iran) containing 10% FBS (BioIdea-Iran) at 37 ° C and 5% CO 2 in order to isolate monocytes in a T-75 cell culture ask.

Activation and expansion of T lymphocytes
Monocyte-depleted PBMC were centrifuged at 500 × g for 5 minutes and lymphocytes were collected and suspended in complete RPMI-1640 medium containing 1 µg / ml phytohemagglutinin (Sigma-Aldrich-USA), then cells were incubated for 24 h at 37 ° C and 5% CO 2 . The cells were incubated with anti-CD3 antibody (BD Bioscience-USA) at 3.3 μg/mL nal concentration for 48 h in order to isolate untouched mature T cells.
The cells were also treated using a combination of different concentrations of these reagents to examine their combined effect. From day 2 the cells were counted daily, and cell viability was assessed by Trypan Blue (Gibco, Invitrogen, CA) staining. Cell morphology and T-cell cluster formation were examined by phase-contrast microscopy for 7 days after treatment in different groups.
After 7 days of treatment, MTT assay was performed to determine the cell toxicity in each treated group compared to untreated cells as a control group.
For this purpose, MTT (3-(4,5-dimethylthiazolyl-2)-2,5-diphenyl2H-tetrazolium bromide, Sigma-Aldrich, 5mg MTT/ml PBS) solution (10 μL/well) was added to the cells and incubated for 4 h in a 37˚C incubator protected from light, giving rise to insoluble (purple) formazan crystals in living cells. Next, 100 μL/well of dimethylsulfoxide (DMSO) was added to solubilize the crystals. Absorbance (570 nm with 650 nm background correction) was measured using Microplate Reader (Eppendorf, Germany). All experiments were performed in triplicate. The cell viability (%) was calculated using the following equation: Cell proliferation assay using Quantitative real-time RT-PCR Total RNA isolation from 1×10 6 cells was performed by RNA extraction kit according to the manufacturer's instructions (Yektatajhiz-Iran) and RNA was eluted with 100 μl elution buffer and then stored at −80 °C until use. The concentration of total RNA was assessed using a spectrophotometer and the quality was checked by agarose gel electrophoresis. First-strand cDNA synthesis was carried out following the manufacturer's protocol (Biofact-south Korea). Quantitative real-time RT-PCR reaction was performed to evaluate the expression level of cell proliferation marker Ki-67 and β-actin housekeeping gene using the following speci c primers: Ki-67 forward, 5'-TCTGACCCTGATGAGAAAGCTC-3', Ki-67 reverse, 5'-TTGAGTCATCTGCGGTACTG-3' and β-actin forward, 5'-ATGTGTGACGAAGAAGCAT-CAGCC-3', and β-actin reverse, 5'-TCATCCCAGTTGGTGATAATGCCG -3'.
All qRT-PCR reactions were prepared in triplicate by mixing 10 μL of SYBR green master mix (Ampliqon kit, Denmark), 0.5 μL of each primer (10 μM), 2 μL of synthesized cDNA, and 7 μL RNase free water. ABI Applied Biosystems™ Thermal cycler (Thermo Scienti c, USA) was used with the cycling conditions comprised an initial 10 min incubation at 95 °C, followed by 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Relative expression was determined using the 2 -ΔΔCT method.

Flow cytometry
The composition of CD8+ and CD4+ cells in isolated T cells was evaluated by ow cytometry. The uorochrome-conjugated mouse anti-human antibodies: Fluorescein isothiocyanate (FITC)-CD8 and phycoerythrin (PE)-CD4 were purchased from BioLegend, Inc. (San Diego, CA, USA). For surface staining, cells were washed once with PBS, incubation was conducted with antibodies in cell staining buffer (3% FBS in PBS) in dark for 25 min at 4°C. Appropriates isotype antibodies were used as negative controls.
Cell uorescence analyzed by the ow cytometry with a BD ow cytometer (BD Bioscience, USA). FlowJo (version 10, TreeStar) software was used for ow cytometric data analysis.

Cytotoxicity assay
To evaluate the antitumor activity of expanded T lymphocytes, MCF-7 breast cancer cell lines as target cells were co-cultured with isolated T lymphocytes in different effector-target (E:T) ratio of 1:1, 3:1, 5:1, 10:1, in 96-well cell culture plate in triplicates and incubated at 37 ° C and 5% CO 2 for 24 hours.
A quantitative assessment of apoptosis was performed using FITC Annexin V Apoptosis Detection Kit (BD Biosciences, San Jose, CA, USA). Brie y, T lymphocytes were removed from the co-culture medium, and the supernatant was collected for further assays. Target cells were detached using 0.025% trypsin, washed twice with cold PBS, and resuspended in binding buffer. Cell staining was performed according to the manufacturer's instructions. Cell uorescence immediately analyzed by the ow cytometry with a BD ow cytometer (BD Bioscience, USA) by accumulating up to 100,000 cells per tube and the obtained data were analyzed. The percentage cytotoxicity in the MCF-7 cells was measured in appropriate control without effector cells. Cellular debris was omitted from the analysis. In addition, MTT assay was performed to con rm the cytotoxicity power of expanded T cells.

Evaluation of T-cell degranulation
The expression level of lysosomal-associated membrane protein-1 (LAMP-1 or CD107a), as a sensitive marker for the cytotoxic activity and T-cell degranulation following stimulation, was accessed using ow cytometry. For this purpose, expanded T cells co-cultured with MCF-7 cells as target cells in 10:1 E/T ratio for 4 h. After incubation, T lymphocytes were removed from the co-culture medium and stained with PECy5-conjugated anti-CD107a (BD Bioscience, San Jose, CA) and cell uorescence was measured by ow cytometry. FlowJo (version 10, TreeStar) software was used for ow cytometric data analysis.
Also, monocyte-depleted untreated PBMC were co-cultured with target cells in the same E:T ratio as control.

Statistical analysis
Statistical analyzes were performed with GraphPad Prism 5.0 (GraphPad Software, Inc., San Diego, CA). Results are shown as mean ± SD. Comparison of results was carried out using the two-tailed unpaired ttest and one-way ANOVA. P < 0.05 was considered statistically signi cant.

Cell viability assay
To quantify the viability of T cells in different culture conditions MTT assay was performed and the results indicated no signi cant difference in cell viability amongst the different groups (all data not shown) (Fig. 1).

Expansion of T cells by combination of optimized concentration of IL-2 and PHA
We compared the effect of different concentrations of insulin, PHA, and IL-2 for their ability on in vitro T cell expansion in 7 days. The cells treated with insulin showed a low expansion rate (total fold expansion range from 0.58 ±0.3413 to 6.2±0.1794) after 7 days (Fig 2.a).
In contrast, T cells expanded higher using PHA and IL-2 independent or in combination with each other (Fig 2.b, c). Isolated T cells showed the most prominent expansion rate with 200 and 300 IU/mL of IL-2 in combination with 1.5 µg/ml of PHA. As shown in gure 2.d, the total fold expansion of T cells in these groups was 176±0.8613 and 592±0.0321 respectively.
As shown in Figure 3, the addition of 300 IU/mL of IL-2 and 1.5 µg/ml of PHA to culture media e ciently activates T cells and enhances T-cell clusters formation.

Relative quantitation of proliferation marker
We evaluate the expression of proliferation marker, Ki-67 at day 7, on cultured T cells by quantitative RT-PCR. The mean value of Ki-67 relative quantitation illustrated in Fig. 4. Signi cantly, the level of Ki-67 mRNA expression was elevated to 79.17 ± 0.1202 and 89.16 ± 0.1732 fold in the cells expanded with IL-2 (200 IU/mL) + 1.5 µg/ml of PHA and IL-2 (300 IU/mL) + 1.5 µg/ml of PHA, respectively, compared to the untreated cells (*P < 0.001 and **P < 0.001).
These results indicated that the addition of 300 IU/mL of IL-2and 1.5 µg/ml of PHA to culture media could expand T cells effectively in a short time without any cytotoxic effect. For this reason, the aforementioned culture condition was chosen to expand T cells for further analysis in this study.

Phenotype of expanded T cells
Recent studies have shown that a combination of CD8+ and CD4+ T cell subpopulations can increase the effectiveness of adoptive cell therapy, particularly in CAR T cell therapy. According to this, we examined the composition of CD8+ and CD4+ T cell subpopulations in the isolated T cell to investigate the effect of this short in vitro expansion time on the phenotype of expanded T cells.
The expanded cell papulation possessed similar phenotypes and both CD4+ and CD8+ cells were detected under the optimized conditions and the mean percentage of CD4+ and CD8+ cells at day 7 were 24.96±1.812% and 74.94±2.319 % respectively (Table.1).

Cytotoxic activity of isolated T cells
Annexin V/PI double staining and MTT assay were performed to analyze the impact of optimized culture media on the functional characteristics of the expanded T cells.
The cytotoxic activity was assessed against MCF-7 breast cancer target cells by ow cytometry and the induced apoptosis was analyzed in MCF-7 cells as shown in Fig. 5.A.
The percentages of apoptotic cells in case of co-culture of expanded T cells in 300 IU/mL of IL-2 and 1.5 µg/ml of PHA containing media with MCF-7 in 10:1 E/T ratio was found to be 68.5% while, no signi cant apoptosis was observed in control cells (0.28%). These results were well correlated with MTT data (Fig.  5.B).

CD107a Degranulation Assay
Degranulation assay detecting surface expression of CD107a as a marker for stimulation-induced granule exocytosis was performed to evaluate the degranulation and cytotoxic power of expanded T cells after the co-culture experiment. The percentages of CD107a positive cells were 75.3% in compare to monocyte-depleted untreated PBMC (0.19%). As ow cytometry histograms show in gure 6, target cell co-incubation led to a high CD107a expression in the T cell and expanded T cells can effectively stimulate in the presence of tumor cells.

Discussion
There have been remarkable recent advances in adoptive immunotherapy based on ex-vivo expanded autologous or antigen-speci c T cells, in particular, gene-modi ed T-cells expressing chimeric antigen receptors (CARs), to treat advanced cancers [17]. Adoptive T cell therapy is a rapidly progressing arena and there is evidence for the potential of this approach as an in uential treatment strategy for cancer.
Since this type of treatment requires a large number of expanded T cells for re-infusion to the patient, over the last decade, the multi-step process for T-cell therapies has been re ned to decrease the time of expansion and improve the quality of isolated cells [18]. The optimized protocol for the large-scale expansion of patient-derived T cells is essential to facilitate the possibilities of the therapeutic potential of adoptive T cell therapy [19]. In addition, the immune-phenotype of isolated cells in uences the effectiveness of treatment, and the expanded cells must be able to ght against the tumor cells [11,20,21], so the optimization of expansion procedure in order to obtain high yield while retaining the desired phenotype and function is an emerging need in T cell-based therapies. In clinical studies and commercial procedures of CAR T cells, the CD3 + T cell population is isolated from the peripheral blood mononuclear cell population by anti-CD3 monoclonal antibody. This is a fast, low cost and simple strategy in producing CAR T cells [3].
In this report, we demonstrate that the addition of PHA and IL-2 to the cell culture medium after CD3 + T cell isolation can improve T cell growth and decrease the time of expansion.
According to our results, adding insulin for 7 days had no signi cant effect on cell proliferation and this may be caused by the lack of insulin receptor expression on T cells. However, signi cant up-regulation of insulin receptor expression is observed on activated T cells, which suggests an important role of insulin signaling during T cell activation [16]. Therefore, although insulin is a suitable mitogen its use in the expansion step does not seem to have much effect.
Also, insulin in combination with PHA and IL-2 did not have a signi cant effect on the rate of cell proliferation induced by these reagents. Therefore, cells treated with different doses of insulin were excluded from our study.
Isolated T cells showed the most prominent expansion rate with optimized culture media. Similar results previously reported T cell expansion in the Quantum system up to 500-fold. The Quantum® Cell Expansion System is a functionally-closed, automated, hollow-ber bioreactor system that is able to expand T cells [22].
Another study demonstrated 1300 mean fold expansions of T cells using the Quantum system. Given that this system conforms to the Good Manufacturing Practice (GMP), its effectiveness may be more appropriate in clinical use, however, it requires a special environment, technique, and tools [23].
Recent studies showed that the evaluation of proliferation marker at mRNA level would provide more de ned information about the proliferation state of cells [24,25]. Follow these observations, we conducted a quantitative assessment of Ki-67 as a proliferation marker using quantitative real-time PCR. Signi cantly, the level of Ki-67 mRNA expression was elevated in comparison with untreated cells.
On the other hand, given that studies have shown that the use of T cell populations composed of CD4 + and CD8 + cells has a synergistic effect on the response against tumor cells, optimal methods of T cell expansion should ultimately lead to the proper composition of these cells [26].
The phenotypic composition obtained in our study is appropriate for the construction of chimeric antigen T cells using CD3 + T cell population, although, may not be suitable for other purposes.
On the other hand, pre-clinical animal models have con rmed that memory T cells are pivotal for in-vivo anti-tumor e cacy. Kaartinen and et al shown that the number of early memory T cells can be increased by reducing the amount of IL-2 as well as limiting the length of T cell expansion [27]. Since the memory phenotype in expanded cells has not been considered, it is necessary to determine the proportion of these cells in the population of expanded cell composition in future studies.
In addition, the anti-tumor e ciency of expanded T cell population was assessed using a co-cultured experiment. Our results showed that isolated cells have the ability to induce cell death in tumor cells when expanded T cells exposed to them at 10:1 E/T ratio. This antitumor activity of the expanded cells was con rmed by increasing the effector cell activation marker, CD107a.
Our results are con rmed by recent studies that have shown the effect of IL-2 on T cell cytotoxic function [28,29], but the effect of this substance in combination with PHA on cell proliferation has not been determined. Recent work has shown that IL-2 can activate a distinctive pattern of the signaling pathway in the antigen-independent manner which associated with strong lymphocyte-speci c protein tyrosine kinase/JAK3-dependent activation of the PI3K/AKT pathway. These signaling pathways induce rapid proliferation of IL-2 activated naïve CD8 + T cells and upregulate expression of eomesodermin (Eomes), signifying the differentiation fate of primary naïve effectors into long-lived memory cells. In addition, results from previous studies indicated that IL-2-stimulated CD8 cells showed strong effector function after adoptive transfer in-vivo [30,31].

Conclusion
Overall, we report on our efforts to optimize the T cell expansion protocol with regard to increase the total expansion fold in the shortest time. We also identi ed the phenotype of the cell population to determine the CD8 + and CD4 + composition of expanded T cells.
The activation and tumor cytotoxicity of expanded T cells against target cells show that the optimized method in this study can be used as a fast, low cost, and effective method in producing cell population required in adoptive T cell therapy studies.    Optimization of culture conditions for T cells expansion. Aiming to optimize the culture condition for su cient T cell expansion, isolated T cells were treated with different mitogen reagents for 7 days (a-d).
Cell counts were determined to re ect the e ciency of T cell expansion in different culture conditions and the highest total fold expansion was observed with combination of 300 IU/mL of IL-2 and 1.5 µg/ml of PHA. Data are expressed as mean ± SD (n=3).
Page 15/18 Figure 3 T-cell clusters formation. T cells effectively activated with addition of 300 IU/mL of IL-2 and 1.5 µg/ml of PHA to culture media and T-cell clusters were formed on day 2 (Magni cation × 400).

Figure 4
Quantitative real-time polymerase chain reaction results. The Ki-67 expression level was signi cantly increased in T cells when a combination of 300 IU/mL of IL-2and 1.5 µg/ml of PHA was added in culture media as illustrated in the graph. The results are presented as the means ± SD, n = 3 (*P < 0.001).  Flowcytometry analysis of CD107a expression on target cell co-cultured expanded T cells. Degranulation assay was conducted using ow cytometry and degranulation marker CD107a was released by expanded