C57BL/6 and NFAT2^fl/flCD4-Cre were bred and maintained at the Brazilian National Cancer Institute (INCA) animal facility (Rio de Janeiro, RJ, Brazil). Also, 6- to 12-week-old mice were used in all experiments. NFAT2^fl/flCD4-Cre mice were generated in Dr. Anjana Rao’s laboratory (La Jolla Institute for Allergy and Immunology, San Diego, CA, USA). All animal experiments were performed in accordance with the Brazilian Government’s ethical and animal experimental regulations. The experiments were approved and conducted according to the animal welfare guidelines of the Ethics Committee of Animal Experimentation from INCA (CEUA process no. 004/13).

Primary cells were cultured in DMEM supplemented with 10% FCS, l-glutamine, streptomycin–penicillin, essential and non-essential amino acids, vitamins, HEPES, and 2-ME (all from Gibco) in a humidified environment containing 5% CO₂ at 37°C. P815 cell line was cultured in RPMI supplemented with 10% FCS, l-glutamine, streptomycin–penicillin, sodium pyruvate, and 2-ME (all from Gibco). CD8⁺ T cells were purified from total lymph nodes (inguinal, brachial, and axillary). Cells were isolated with Dynal Mouse CD8 Negative Isolation Kit (Invitrogen). For cytometric purity analysis, cells were stained with anti-CD4-PE and anti-CD8-FITC Abs (both from BD Pharmingen) and analyzed by flow cytometry on a FACScan (BD Biosciences). Cell populations were isolated to >95% purity. Total lymph nodes or purified CD8⁺ T lymphocytes were activated in vitro for indicated times with plate-bound anti-CD3 (1 μg/ml; BD Pharmingen; otherwise indicated) plus soluble anti-CD28 (1 μg/ml; BD Pharmingen). To differentiate CD8⁺ T lymphocytes into cytotoxic CD8⁺T lymphocytes in vitro, 1 × 10⁶ cells were activated for 48 h with plate-bound anti-CD3 (1 μg/ml) plus soluble anti-CD28 (1 μg/ml). Cells were expanded daily for 4 days with 200 U/ml of murine recombinant IL-2 (43). At day 5, cells were analyzed by flow cytometry and in cytotoxicity assay in vitro.

Purified CD8⁺ T lymphocytes (3 × 10⁶) were stimulated or left untreated for indicated times with plate-bound anti-CD3 plus soluble anti-CD28 (both at 1 μg/ml). To extract cytoplasmic fraction, cells were lysed in buffer containing 10 mM Tris-Cl pH 7.5, 10 mM NaCl, 3 mM MgCl₂, 0.5 mM DTT, 0.1 mM EDTA, 0.5% NP-40, 1 mM PMSF, and protease inhibitor. Nuclear extracts were obtained by lysis of the nuclei with buffer containing 40 mM Tris-Cl pH 7.5, 10 mM EDTA, 60 mM sodium pyrophosphate, and 5% SDS followed by incubation at 100°C for 15 min.

Total protein extract from 3 × 10⁶ T lymphocytes was obtained from cells lysis in buffer containing 40 mM Tris pH 7.5, 60 mM sodium pyrophosphate, 10 mM EDTA, and 5% SDS, followed by incubation at 100°C for 15 min. Total cell lysates were resolved by SDS-PAGE, and the separated proteins were transferred onto a nitrocellulose membrane. The antibodies used were GAPDH monoclonal antibody 6C5 (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and NFAT2 monoclonal antibody 7A6 (Santa Cruz Biotechnology, Santa Cruz, CA, USA). The immunodetection was performed with the ECL Western Blotting Detection Kit (GE Healthcare).

Intracellular localization of NFAT2 protein was analyzed in purified CD8⁺ T cells (5 × 10⁵ cells) from C57BL/6 mice by immunofluorescence staining. Briefly, cells were stimulated or left unstimulated for indicated time with plate-bound anti-CD3 plus soluble anti-CD28 (both at 1 μg/ml). Then, cells were fixed in 4% paraformaldehyde, permeabilized with 0.5% Non-idet P-40, and stained with anti-NFAT2 Ab. The cells were photographed with Confocal Laser Scanning Microscope FV10i-O Olympus.

To analyze cell surface expression levels of the various markers, the cell suspensions (1 × 10⁶) were incubated with following mAbs: anti-CD3-APC, anti-B220-FITC, anti-CD4-PE, anti-CD4-FITC, anti-CD8-FITC, anti-NK1.1-PerCP.Cy5.5, anti-PLZF-PE, anti-CD44-FITC, anti-CD69-APC, anti-CD62L-PE, anti-CD62L-PerCP, anti-CD25-APC, anti-CD122-PE, and anti-CD127-PE (all BD Pharmingen) acquired in a FACScan (Becton Dickinson, Mountain View, CA, USA) and analyzed using FlowJo software.

For intracellular cytokine staining, 1 × 10⁶ cells were stimulated in vitro for 6 h with PMA (10 nM) plus ionomycine (1 μM, both from Calbiochem). Brefeldin A (1:1000; BD Pharmingen) was added to the culture for last 2 h. Cells were harvested and stained with anti-CD8-FITC Abs. Then, cells were fixed, permeabilized, and stained with anti-IFN-γ-FITC, anti-IL-2-PE, anti-IL-4-APC, and anti-Granzime B-FITC Abs. For PLZF intracellular staining, cells were harvested and stained with anti-CD3-APC, fixed, permeablized, and stained with anti-PLZF-PE. Samples were analyzed by flow cytometry on a FACScan (BD Biosciences) and FlowJo software.

Purified CD8⁺ T lymphocytes (5 × 10⁶) were stained with CFSE Cell Proliferation Assay (Invitrogen) according to manufacturer’s instructions, then stimulated or not with plate-bound anti-CD3 (0.25 μg/ml) plus soluble anti-CD28 (1 μg/ml) for indicated times in the absence or presence of 200 U/ml of IL-2. Carboxyfluorescein diacetate succinimidyl ester (CFSE) dilution was analyzed by flow cytometry on FACScan (BD Biosciences) and FlowJo Software.

Total RNA from total or sorted into CD44^low and CD44^high naive CD8⁺ T lymphocytes, total thymocytes, CD4⁺ thymocytes, and total splenocytes was extracted using Trizol LS Reagent (Invitrogen), and first-strand cDNA was synthesized using ImProm-II Reverse Transcription System (Promega) or Superscript II kit (Life Technology) for sorted CD44^low and CD44^high CD8⁺ T lymphocytes and CD4⁺ thymocytes. Real-time polymerase chain reactions for NFAT2 and PLZF were performed using SYBR Green master mix (Applied Biosystems) and for Blimp-1, Tbx21, IL-4, Eomes using Taqman probes. Actin or HPRT were used as endogenous control. Sequences of primers used for SyberGreen real-time PCR are NFAT2-F 5′-CCAGAAAATAACATGCGAGCC-3′, NFAT2-R 5′-GTGGGATGTGAACTCGGAAG-3′, PLZF-F 5′-GAGCAGTGCAGCGTGTGT-3′, and PLZF-R 5′-AACCGTTTTCCGCAGAGTT-3′ (44), Actin-F 5′-ATGGTGGGAATGGGTCAGAAG-3′, Actin-R 5′-TTCTCCATGTCGTCCCAGTTG-3′. All procedures were performed according to the manufacturers’ instructions.

Target cells (P815) at 10⁶/ml were labeled with 10-μM Calcein-AM (Molecular Probes) in complete medium for 30 min at 37°C. The assay was performed in V bottom 96-well microtiter plates with E:T ratios ranging from 2:1 to 0.5:1 with wells for spontaneous (only target cells in complete medium) and maximum release (only target cells in medium plus 2% Triton X-100) in the presence of 1 μg/ml of soluble anti-CD3, in triplicate. Each well contained from 1 × 10⁵ to 2.5 × 10⁴ lymphocytes in 100 μl of complete medium and 5 × 10⁴ target cells/50 μl of complete medium. After incubation at 37°C in 5% CO₂ for 4 h, 75 μl of each supernatant was harvested and transferred into new plates. Samples were measured using a Spectramax Gemini dual-scanning microplate spectrofluorimeter (Molecular Devices) (excitation: 485 nm; emission: 530). Data were expressed as arbitrary fluorescent units (AFU). Percent lysis was calculated with the following formula: [(A sample − A spontaneous)/(A max − A spontaneous)] × 100.

For single comparisons, unpaired Student’s t-test was used. Flow cytometry and mean fluorescence intensity analysis was performed using FlowJo 10.1r5 software and analyzed for statistical significance by unpaired two-tailed Student’s t-test using Prism (GraphPad Software). Differences with p < 0.05 were considered statistically significant.

TCR stimulation results in calcium mobilization and consequent activation of NFAT transcription factors. NFAT is activated by the phosphatase calcineurin, which dephosphorylates multiple phosphoserines in the regulatory domain of NFAT, leading to NFAT nuclear translocation (45).

To evaluate the NFAT2 expression and activation, CD8⁺ T lymphocytes from WT mice were purified, left unstimulated, or stimulated in vitro with anti-CD3 plus anti-CD28 for different times. As expected, TCR stimulation increased both NFAT2 mRNA and protein level (Figures 1A,B). In unstimulated cells, NFAT2 was present in the cytoplasm in an inactive phosphorylated form (Figures 1C,D). At 24 h post-activation, we observed NFAT2 dephosphorylation (Figures 1B,C), and at 48 h post-activation, we detected NFAT2 in the nucleus (Figures 1C,D). These data show that NFAT2 protein is present and functional in CD8⁺ T lymphocytes.

To study how NFAT2 regulates CD8⁺ T cell development and responses, we used mice with T cell-specific deficiency in NFAT2. We used CD4-cre mice crossed to mice with NFAT2 flanked by loxP sites (NFAT2^fl/fl). In this model, the NFAT2 gene is excised at the DP stage of thymocyte development, affecting both SP CD4 and CD8 mature T cells. The absence of NFAT2 protein expression in CD8⁺ T cells from NFAT2^fl/flCD4-Cre mice was confirmed by Western blot (Figure 2A).

NFAT2^fl/flCD4-Cre mice showed only little changes in total cell numbers in the thymus and spleen, but a significant decrease of total cell number in lymph nodes (Figure 2B), and consistently NFAT2^fl/flCD4-Cre mice had reduced size of lymph nodes (data not shown). Thymus of NFAT2^fl/flCD4-Cre mice had similar frequency of DP as well as CD4⁺ single positive (SP) thymocytes as WT (Figure 2C). Interestingly, there was an increase in the frequency of CD8⁺ SP thymocytes (Figure 2C). To examine whether observed difference was thymus-specific or a general characteristic, we then analyzed spleen and lymph nodes of NFAT2^fl/flCD4-Cre mice. The spleen of NFAT2^fl/flCD4-Cre mice had significant reduction of the frequency of total T lymphocytes, in particular CD4⁺, and a slight increase of B lymphocytes, while no change was observed for CD8⁺ T cells (data not shown). Lymph nodes of NFAT2^fl/flCD4-Cre mice showed significantly lower number of CD8⁺ T lymphocytes in NFAT2^fl/flCD4-Cre mice compared to WT (data not shown), however, with only little changes in the frequency of total T and B lymphocytes. Similarly, to the spleen, we observed a decrease in the frequency of CD4⁺ T cells (Figure 2C), and as seen in the thymus, an increase in the frequency of CD8⁺ T lymphocytes (Figure 2C).

A higher proportion of CD8⁺ T cells in the thymus and lymph nodes of NFAT2^fl/flCD4-Cre mice compared to WT was expressing CD44, an activated T cell marker. An increase in the frequency of CD8⁺CD44^high cells was also observed in the spleen of NFAT2^fl/flCD4-Cre mice even without a significant increase in the frequency of CD8⁺ T cells (Figure 3A). As the percentage of CD8⁺CD44^high cells was increased in the thymus, lymph nodes, and spleen, it suggested that this is not dependent on preferential enrichment of these cells in specific tissues.

To exclude the fact that CD8⁺CD44^high T cells are present due to an activation, we examined the expression of two activation markers: CD69 (early activation marker) and CD25 (IL-2 receptor α-chain). CD8⁺CD44^high population from both WT and NFAT2^fl/flCD4-Cre mice expressed low and similar percentage of CD69⁺ and CD25⁺ cells showing that CD8⁺CD44^high population is not pre-activated in vivo (Figure 3B).

Expression of CD122, the IL-2Rβ chain, has been reported to define the population of innate-like CD8⁺ T cells (46), and we found that the CD8⁺CD44^high population in both WT and NFAT2^fl/fl CD4-Cre mice expressed CD122, although the NFAT2^fl/flCD4-Cre mice carried larger percentage of these cells (Figure 3B).

Moreover, NFAT2^fl/flCD4-Cre mice had larger percentage of CD8⁺CD44^high CD127⁺ and CD8⁺CD44^high CD62L⁺ compared to WT (Figure 3B), further showing that CD8⁺CD44^high population display conventional memory T cell characteristics without previous antigen exposure.

Another characteristic of innate-like CD8⁺ T cells is the rapid production of large amount of IFN-γ upon stimulation (16). To determine if these innate-like CD8⁺ cells exhibit similar functional behavior, we analyzed their ability to rapidly secrete IFN-γ upon stimulation. Comparison of IFN-γ production by PMA plus ionomycin stimulated CD8⁺CD44^high T cells in vitro indicated that NFAT2-deficient cells behave similarly to those from WT mice in rapid (within 6 h) secretion of this cytokine. A similar percentage of the CD8⁺CD44^high T cells responded from both WT and NFAT2^fl/flCD4-Cre mice (Figure 3C). By contrast, CD8⁺CD44^low T cells did not secrete any appreciable IFN-γ in this time period (data not shown).

Further analysis of CD8⁺CD44^high population revealed that cells from both NFAT2^fl/flCD4-Cre and WT mice expressed elevated Tbx21 and Eomes mRNA level compared to respective CD8⁺CD44^low counterparts. Eomes expression remained similar (Figure 3D), while Tbx21 mRNA was reduced without a significant difference in CD8⁺CD44^high cells from NFAT2^fl/flCD4-Cre animals compared to WT.

Taken together, these results demonstrate that NFAT2 deficiency leads to the development of innate-like CD8⁺ T cells without affecting theirs characteristics, such as surface markers and transcription factors expression and IFN-γ production.

Interleukin-4 is a cytokine classically associated with CD4⁺ T helper type 2 differentiation (47) but has been recently shown in different mice models to be also required for the development of innate-like CD8⁺ T lymphocytes through Eomes upregulation (9–11, 44, 48–50). To compare the IL-4 production between NFAT2^fl/flCD4-Cre and WT mice, we stimulated WT and NFAT2-deficient thymocytes, splenocytes, and total lymph nodes cells ex vivo with PMA plus ionomycine for 6 h. We analyzed the production of IL-4 by CD3⁺ cells. We observed that NFAT2^fl/flCD4-Cre mice had higher percentage of CD3⁺IL-4⁺ cells in the thymus and lymph nodes and only small increase in the spleen (Figure 4A). Consistently, IL-4 mRNA was clearly more abundant in the thymus, lymph nodes, and even the spleen of NFAT2^fl/flCD4-Cre mice (Figure 4B).

In several gene-deficiency models, IL-4 is produced by an expanded population of T cells expressing the transcription factor PLZF. First, we tested whether NFAT2 absence affects the PLZF expression, by measuring the PLZF mRNA expression in the thymus by real-time PCR. Indeed, we found much higher PLZF mRNA level in NFAT2-deficient CD4⁺ thymocytes compared to WT (Figure 4C). Interestingly, in NFAT2^fl/flCD4-Cre mice, we observed the reduction of CD3⁺ NK1.1⁺ as compared to WT. Further analysis showed that PLZF⁺ T cell population in NFAT2^fl/flCD4-Cre mice was NK1.1⁻, CD4⁺, or CD4⁻ (Figure 4D). Thus, the absence of NFAT2 results in the expansion of PLZF⁺, NK1.1⁻, CD4⁺, or CD4⁻ expressing T cells and increases the IL-4 production in the thymus and the periphery, which, in consequence, may lead to the higher frequency of innate-like CD8⁺ T cells.

Nuclear factor of activated T cell transcription factors are ubiquitous regulators of gene expression during cellular activation; however, most reports are focused mainly on CD4⁺ T lymphocytes [for review, see Ref. (23)]. To evaluate the specific role of NFAT2 in CD8⁺ T cell activation, we purified CD8⁺ T lymphocytes from lymph nodes of WT and NFAT2^fl/flCD4-Cre mice, and we left them either unstimulated or stimulated for 24 and 48 h with different doses of anti-CD3 plus 1 μg/ml of anti-CD28. As expected, more than 85% of unstimulated cells were positive for CD62L, a protein highly expressed on naive cells, and only few expressed CD69 and CD25 activation markers (data not shown). TCR stimulation with 1, 0.5, and 0.25 μg/ml of anti-CD3 for 24 h led to a reduction of both CD69 and CD25 expression in NFAT2-deficient cells compared to WT (Figure 5, upper panel). The dose of 0.1 μg/ml of anti-CD3 was not sufficient to activate CD8⁺ T cells (Figure 5). Interestingly, TCR stimulation with 1 and 0.5 μg/ml of anti-CD3 for 48 h resulted in a similar cell activation as shown by CD69 and CD25 expression between WT and NFAT2-deficient cells (Figure 5, lower panel), indicating that NFAT2 absence was compensated by another transcription factor, likely NFAT1. Only at lower anti-CD3 concentration (0.25 μg/ml), the cell activation of NFAT2-deficient cells remained significantly reduced in comparison to WT (Figure 5, lower panel), indicating that the absence of NFAT2 increases the threshold of CD8⁺ T cell activation at suboptimal TCR stimulation.

As NFAT2 deficiency in CD8⁺ T cells increased the threshold of T cell activation, we next tested the effect of NFAT2 absence on CD8⁺ T cell proliferation. Purified CD8⁺ T cells were stained with CFSE and then activated for different times with 0.25 μg/ml anti-CD3 plus 1 μg/ml of anti-CD28. At indicated times, cell proliferation was monitored by flow cytometric measurement of CFSE dye dilution. Low concentration of anti-CD3 led to a reduced proliferation of NFAT2-deficient CD8⁺ T cells with division indexes of 3.19 for WT and 2.64 for NFAT2-deficient cells after 72 h, indicating that NFAT2 absence increases the threshold of triggering CD8⁺ T cell into division (Figure 6A).

Consistent with defects in cell activation and proliferation, NFAT2-deficient CD8⁺ T cells were also weak producers of both IFN-γ and IL-2. The slight difference was observed even at strong TCR stimulation, but clear decrease of the percentage of IFN-γ- and IL-2-producing cells was observed at suboptimal dose of anti-CD3 (Figure 6B). Next, we decided to determine whether defect of IL-2 production in NFAT2-deficient CD8⁺ T cells results in delayed cell proliferation. Purified CD8⁺ T cells were stained with CFSE and then activated for different times with 0.25 μg/ml anti-CD3 plus 1 μg/ml of anti-CD28 in the absence or presence of 200 U/ml of IL-2. At indicated times, cell proliferation was monitored by flow cytometric measurement of CFSE dye dilution. Addition of IL-2 restored the proliferation of NFAT2-deficient CD8⁺ T cells at 48 h (data not shown) and 72 h post-activation (Figure 6C) demonstrating that reduced proliferation in the absence of NFAT2 is due to decreased IL-2 secretion.

Antigen-induced CD8⁺ T cells differentiate into cytotoxic CD8⁺ T cells (CTLs) that kill infected cells and release cytokines, such as IFN-γ (1). To analyze if NFAT2-deficient CD8⁺ T cells have any other intrinsic defects besides increased threshold of activation, proliferation, and cytokine secretion, we determined if differentiation and function of CTLs in vitro were intact. To differentiate CTLs in vitro, we followed as reported (43). Briefly, purified CD8⁺ T cells from WT and NFAT2^fl/flCD4-Cre mice were stimulated for 48 h with 1 μg/ml of anti-CD3 plus anti-CD28 and expanded daily with murine recombinant IL-2 (200 U/ml) (Figure 7A). At day 5, cells were collected and analyzed. The WT cytotoxic CD8⁺ T cells differentiated in vitro showed a downregulation of NFAT2 mRNA expression compared to naive CD8⁺ T cells (Figure 7B). No significant difference in cell viability and proliferation between WT and NFAT2-deficient CTLs was observed using the Trypan blue exclusion method (Figure 7C). The expression of activation markers, CD44 and CD25, was also comparable between WT and NFAT2-deficient CTLs (Figure 7D), indicating that NFAT2 does not influence either proliferation or activation of cytotoxic CD8⁺ T cells.

CD8⁺ T cytotoxic activity was measured as the ability to effectively lyse P815 cells. NFAT2-deficient CTLs had slightly reduced potential to lyse P815 cells at 1:1 and 1:2 effector:target ratios (Figure 7E). Consistently, granzyme B expression was slightly decreased without statistical significance in NFAT2-deficient CTLs (Figure 7F). Remarkably, NFAT2-deficient CTLs produced significantly less IFN-γ compared to their proficient counterparts (Figure 7F).

Transcription factors T-bet and Blimp-1 are reported to play a crucial role in promoting cytotoxic CD8⁺ T cell differentiation (51, 52); therefore, we tested what is the effect of NFAT2 absence in CTLs on the expression of T-bet and Blimp-1. NFAT2-deficient CTLs had lower expression of both T-bet and Blimp-1, although in both cases, the difference remained insignificant (Figure 7G).

Taken together, these data demonstrate that NFAT2 does not significantly affect the differentiation of CD8⁺ T cells into cytotoxic T cells but positively controls the IFN-γ production.