NFATc1 Promotes Antitumoral Effector Functions and Memory CD8⁺ T-cell Differentiation during Non–Small Cell Lung Cancer Development

Nuclear factor of activated T cells is a transcription factor family that consists of five members, among which, NFAT1–4 (NFATc1–NFATc4) are regulated by Ca²⁺–calcineurin signaling. NAFTc1 and NFATc2 are the main isoforms expressed in T cells critical in regulating early gene transcription in response to T-cell receptor (TCR)-mediated signals (1–3). In resting T cells, NFAT transcription factors are located in the cytosol in an inactive, phosphorylated state. Upon TCR stimulation, NFAT is activated by calcium flux released from the endoplasmic reticulum and the extracellular environment through store-operated channels. Induced intracytoplasmic calcium promotes calmodulin, a calcium sensor protein, which activates the phosphatase calcineurin. Calcineurin dephosphorylates NFAT, resulting in its translocation into the nucleus, where it cooperates with other transcription factors promoting cell-specific gene transcription (3, 4).

The expression of NFATc1 in T cells is initiated by two distinct promoters, resulting, along with alternative splice/polyadenylation events, in six isoforms: NFATc1αA-C and NFATc1βA-C (5, 6). NFATc1αA has been shown to be strongly induced following TCR stimulation and is maintained by a positive autoregulation, which enables a massive synthesis and facilitates T cells to exert effector functions and escape from apoptosis (5, 7–9). Cytotoxic effector functions are mediated by CD8⁺ T cells via the secretion of inflammatory cytokines (10, 11). A recent study described that NFATc1-deficient cytotoxic T cells showed reduced cytotoxicity against tumor cells. Furthermore, transcriptome analysis demonstrates diminished RNA levels of numerous genes in NFATc1^−/− CD8⁺ T cells, including Il2 (9). IL2 is one of the first molecules that has been shown to be induced by NFATc1 (12) and is important for T-cell activation, proliferation, and survival (13, 14). Furthermore, IL2 influences the generation of memory CD8⁺ T cells classified into central memory (TCM), effector memory (TEM), and tissue-resident memory (TRM) CD8⁺ T cells that differ in their tissue-homing capacity and effector functions (13, 15–17). In most of the established tumors, tumor-infiltrating lymphocytes (TIL) are found to be exhausted, leading to cancer immune evasion. One remarkable feature of functionally exhausted T cells is the expression of inhibitory checkpoint receptors that desensitize TCR signaling, leading to functional impairment of T-cell activation. On the basis of these findings, cancer immunotherapies have been developed that reactivate exhausted TILs by blocking inhibitory checkpoint receptors (18–20). Programmed cell death (PD)-1 is one of the most successful checkpoint target in different cancer types including non–small cell lung cancer (NSCLC). Nevertheless, only approximately 30% of patients are responsive to this therapy, and there is a need to find alternate regulators to improve current approaches (19–23). NFATc1 has been shown to induce the expression of PD-1 upon T-cell activation (24). But there is still a contradiction regarding the influence of αPD-1 antibody treatment on NFATc1 in T cells. Triggering of PD-1 by its ligand PD-L1 expressed on tumor cells induces the inhibition of the PI3K/Akt pathway. Akt normally negatively regulates the glycogen-synthase kinase 3 (GSK3), which inhibits NFATc1 (3, 25, 26). Blocking PD-1 by αPD-1 antibodies restores TCR signaling and ensures the activation of Akt, which inhibits GSK3 and promotes the activation of NFATc1, which induces T-cell activation and effector functions (7, 9, 25). Thus, NFATc1 could be a key regulator in αPD-1 anticancer immunotherapy.

Here we demonstrate an important function of NFATc1 for successful T-cell–mediated antitumoral immune responses in the setting of NSCLC. Targeted deletion of Nfatc1 in T cells (NFATc1^ΔCD4) induced increased lung tumor growth in mice associated with impaired T-cell activation and function. In the absence of NFATc1, reduced IL2 influenced the development of memory CD8⁺ T cells. Consistently, we found a reduction of TEM and CD103⁺ TRM T cells in the lung of tumor-bearing NFATc1^ΔCD4 mice underlining an impaired cytotoxic T-cell response and a reduced TRM tissue-homing capacity. PD-1 was found coexpressed and accumulated with CD4⁺ICOS⁺ T-cells in the draining lymph nodes (dLN) of NFATc1^ΔCD4 mice and associated with lung tumor cell growth. Moreover, targeting PD-1 induced significantly NFATc1 in T cells accompanied by increased antitumor cytotoxic functions. Thus, this study revealed a role of NFATc1 not only in the activation and cytotoxic functions of T cells, but also in the development of memory CD8⁺ T-cell subsets and the regulation of T-cell exhaustion underlining the indispensability of NFATc1 for successful antitumor immune responses in NSCLC.

This study was performed at the Friedrich-Alexander-University of Erlangen in Germany (Erlangen, Germany) and was approved by the ethics review board of the University of Erlangen (Re-No: 56_12B; DRKS-ID: DRKS00005376). Fifty-eight patients that suffered from NSCLC underwent surgery and gave their approval to be enrolled in this study in an informed written consent. The patient studies were conducted in accordance with the ethical guidelines of the Declaration of Helsinki.

The diagnosis of lung cancer was based on pathologic confirmation. The histologic types of lung cancer were classified according to the classification of the World Health Organization (WHO), formulated in 2004. The staging of lung cancer was based on the Cancer TNM Staging Manual, formulated by the International Association for the Study of Lung Cancer (IASLC) in 2010. During surgery, lung tissue samples were taken from the tumoral area (TU: solid tumor tissue), the peritumoral area (PT: up to 2 cm away from the solid tumor) and from the tumor-free control area (CTR: >5 cm away from the solid tumor). This cohort of patients with NSCLC was described previously (27).

For protein extraction, lung tissue samples were lysed in RIPA buffer (Thermo Fisher Scientific) with added inhibitor cocktail (Roche Diagnostics), followed by homogenization using the SpeedMill PLUS (Analytik Jena) and innuSPEED lysis Tube P (Analytik Jena). After centrifugation (5 minutes, 3,000 rpm, 4°C), supernatants were incubated on ice for 45 minutes, followed by centrifugation (1 × 5 minutes, 3,000 rpm, 4°C; 1 × 45 minutes, full speed, 4°C). Finally, protein concentration was calculated after using Bradford Assay (Protein Assay Dye Reagent Concentration, Bio-Rad). Western blot analysis to detect NFATc1 (1:250; sc-1149, Santa Cruz Biotechnology), pNFATc1 (1:250; sc-32978, Santa Cruz Biotechnology), and β-actin (1:500, sc-1616, Santa Cruz Biotechnology) was performed as described previously (28) with 50 μg of total lung protein. Quantification of total NATc1 and pNFATc1 was performed using the AlphaView Software for FluorChem Systems (Biozym Scientific).

IHC was performed on paraffin-embedded sections. Before staining, paraffin was removed from the slides by incubation at 72°C for 30 minutes and treatment with Roti-Histol (Carl Roth) two times for 10 minutes. The tissue sections were then rehydrated by immersion in ethanol series with descending concentrations (100%, 95%, 70%) for 3 minutes each and in deionised water for 1 minute, followed by blocking endogeneous peroxidase in 3% H₂O₂ (in methanol) for 20 minutes. Heat-induced antigen retrieval was performed as described previously (27) using 1 mmol/L Tris-EDTA wash buffer. Afterwards, slides were incubated with primary antibody to CD3 (1:100, RBK024, Zytomed Systems GmbH) overnight at 4°C. After different washing steps and detection of the primary antibody following the manufacturer's instructions of the ZytoChem-Plus AP Polymer Kit (POLAP006, Zytomed Systems GmbH), the second antibody to NFATc1 (1:50, sc-7294, Santa Cruz Biotechnology) was applied for 2 hours at room temperature. The second antibody was detected following the manufacturer's instructions of the Dako EnVision Detection System Kit (K4065). Slides were covered with coverslips using Aquatex (108562, Merck). Negative controls were not treated with the primary antibodies and the other steps remain the same. Stained slides were scanned using the digital slide scanner (Scan 150, 3D Histech Ltd) and NFATc1⁺CD3⁺ cells were quantified using the Definiens Tissue Studio 4.1 software (Definiens) at the Institute of Pathology, Friedrich-Alexander-Universität Erlangen-Nürnberg (FAU; Erlangen, Germany). Whole-slide images were visualized by the CaseViewer (Version 2.0, 3D Histech Ltd).

Frozen human tissue samples were cut into small pieces using scalpels (1–3 mm²) followed by the preparation of single-cell suspensions using the Tumor Dissociation Kit, human (Miltenyi Biotec) and the gentleMACS Dissociator (Miltenyi Biotec) according to the manufacturer's instructions. The resulting single-cell suspension was washed and ammonium-chloride-potassium (ACK) lysis buffer was applied as described previously (27). For flow cytometry analyzes, total cells were incubated with the respective mix of surface antibodies dissolved in PBS and incubated for 30 minutes at 4°C. For the intracellular staining of NFATc1, cells were fixed and permeabilized with fix/perm solution, consisting of 1% paraformaldehyde in 70% ethanol in accordance to the manufacturer's protocol (Biolegend). Antibodies used for flow cytometry are shown in Supplementary Table S1. Flow cytometric analyses were performed by using FACS Canto II (BD Biosciences). Datasets were analyzed by Cell Quest Pro version 4.02 (BD Biosciences) and Flow-Jo v10.2 (FlowJo, LLC).

The human A549 cell line was purchased authenticated from the ATCC bank. ATCC authenticates cell lines by Mycoplasma, bacterial, and fungal contamination testing, PCR and sequencing of selected genes, human virus testing, COI assay, and STR DNA profiling. We detected Mycoplasma contamination using the Mycoplasma Detection Kit (Absource Diagnostics GmbH), according to the manufacturer's protocol (latest date: Aug 9, 2016). On average, we perform ten cell passages between thawing and use in the described experiments. Cells were cultured in F-12K Nut mix medium (Gibco, Thermo Fisher Scientific), supplemented with 10 % of FCS (PAA Laboratories), 1% of the antibiotics penicillin and streptavidin (pen/strep; PAA Laboratories) and 1% of l-glutamine (l-Glu; Gibco) at 37°C and 5% of CO₂.

The murine LL/2-luc-M38 (LL/2) cell line was purchased and authenticated from Caliper Life Sciences (Bioware cell line, Caliper Life Sciences). All Caliper Life Sciences cell lines were confirmed to be pathogen-free by the IMPACT profile I (PCR) at the University of Missouri Research Animal Diagnostic and Investigative Laboratory. Furthermore, the luciferase expression is coupled to a neomycin-resistant gene, which renders the cell resistant to Geneticin (G418). Therefore, we treated LL/2 cells with G-418 solution (500 μg/mL, Sigma-Aldrich) to select luciferase-expressing cells. Three cell passages between thawing and usage in the described experiments were performed. LL/2 cells were cultured in DMEM (Thermo Fisher Scientific), supplemented with 10% FCS, 1% pen/strep, and 1% l-Glu at 37°C and 5% of CO₂. Mycoplasma contamination was detected using the Mycoplasma Detection Kit (Absource Diagnostics GmbH), according to the manufacturer's protocol (latest date: August 9, 2016).

The murine CTLL2 cell line was kindly provided to us by PD Dr. Ulrike Schleicher from the Institute of Microbiology, University Hospital Erlangen (Erlangen, Germany). CTTL2 cells were cultured in RPMI1640 medium (Gibco, Thermo Fisher Scientific), supplemented with 10% of FCS, 1% pen/strep, 1% of l-Glu, and 4 ng/mL IL2 (PeproTech GmBH) at 37°C and 5% of CO₂.

Human lung tissue samples were homogenized using Precellys Lysing Kits (Bertin Technologies) and the benchtop homogenizer Minilys (Bertin Technologies) as described in the manufacturer's protocol. Total RNA was isolated using peqGold RNA Pure (Peqlab) according to manufacturer's instructions. RNA was reverse-transcribed into cDNA using the RevertAid First Strand cDNA Synthesis Kit (Fermentas) in accordance to the manufacturer's instructions.

Quantitative real-time PCR (qPCR) of synthesized cDNA was performed by using iTaq Universal SYBR Green Supermix (Bio-Rad) in a total volume of 20 μL. Primers were purchased from Eurofins-MWG-Operon. Primer sequences for murine and human qPCR analyzes are depicted in Supplementary Table S2. Reactions (50 cycles, initial activation 98°C, 2 minutes, denaturation 95°C, 5 minutes, hybridization/elongation 60°C, 10 minutes) were performed using the CFX-96 Real-Time PCR Detection System (Bio-Rad), and analyzed by the CFX Manager Software (Bio-Rad). Relative quantification was performed using the 2^−ΔΔC_t method, hypoxanthine-guanine-phosphoribosyltransferase (Hprt) was used as internal standard.

For siRNA-mediated silencing of NFATc1, 3 × 10⁵ cells were incubated overnight in 6-well plates in antibiotic-free F-12K Nut mix medium containing 10 % FCS. For transfection, 25 nmol/L NFATc1-siRNA (GE Dharmacon) or 25 nmol/L nontargeting control siRNA (siNT; GE Dharmacon) were applied together with 4 μL DharmaFECT Transfection reagent 1 (GE Dharmacon) in 2-mL antibiotic-free medium, supplemented with 10 % FCS according to the manufacturer's instructions. After 24-hour incubation, transfected cells were washed with PBS and then cultured in 2-mL antibiotic-free medium, supplemented with 10% FCS, and LEAF purified anti-human PD-L1 antibody (5 μg/mL; Biolegend) or the respective LEAF purified mouse IgG2bκ isotype control (5 μg/mL; Biolegend) for 48 hours. For the induction of PD-L1 on A549 cells, 3 × 10⁵ cells were incubated overnight ± IFNγ (50 ng/mL; Biolegend) followed by siRNA-mediated silencing of NFATc1 as described above and subsequent treatment with above described anti-human PD-L1 antibody or IgG2bκ isotype control antibody for 24 hours. RNA was isolated and NFATC1-mRNA expression was analyzed via qPCR. Apoptosis assay was performed according to the manufacturer's instructions by staining the cells with Annexin V (BD Biosciences) and propidium iodide (BD Biosciences), followed by flow cytometry analyses.

For luminescence analysis, 7 × 10³ LL/2-luc-M38 (LL/2) cells were cultured in a 96-well plate and incubated for 24 hours at 37°C, 5% CO₂ in DMEM supplemented with 10% FCS, 1% pen/strep, and 1% l-Glu. After 24 hours, supernatants were removed, cells were washed in PBS, and incubated with either 20% lymph node–conditioned medium (LNCM) or lung-conditioned medium (LUCM). For preparation of LNCM or LUCM, total cells of dLNs or lungs were cultured for 24 hours in the presence of αCD3 (10 μg/mL; BD Biosciences) and αCD28 (1 μg/mL; Biolegend) antibodies or in the presence of αPD-1 mAb RMP1-14 (10 μg/mL; Hölzel Diagnostika) or the respective rat IgG2a mAb 2A3 isotype control (10 μg/mL; Hölzel Diagnostika). Resulting supernatant was used in LL/2 luminescence assay. After 24-hour incubation, LNCM or LUCM was removed and LL/2 cells were treated with 15 μg/mL luciferin (Promega) to detect the luminescence intensity using the Centro XS³ LB 960 Microplate Luminometer (Berthold Technologies). Respective cell numbers were calculated by using a standard curve.

For apoptosis analysis, 7 × 10³ LL/2-luc-M38 (LL/2) cells were seeded in 96-well plates in DMEM supplemented with 10% FCS, 1% pen/strep, and 1% l-Glu. IncuCyte Caspase-3/7 Reagent (Essen BioScience) was used at a final concentration of 1 μmol/L (1:5,000) and added directly to LL/2 cells in addition to 30 ng/mL rmTNFα (ImmunoTools). Apoptotic cells (green object count, 1/mm²) were detected using the IncuCyte Live-Cell Analysis System (Essen BioScience) over a period of 72 hours, and the images were captured every 4 hours. Apoptosis has been quantified as the number of green fluorescent caspase-3/7–active objects for each time point.

The ELISA technique was utilized to analyze the cytokine concentration in cell culture supernatants. ELISA was performed in accordance with the manufacturer's instructions. Murine TNFα, IL2, IL10, and IFNγ ELISA Sets were obtained from BD Biosciences, VEGFA, and IL7 ELISA Sets from R&D Systems.

NFATc1^ΔCD4 and NFATc1^fl/fl littermate control mice are on a C57BL/6 genetic background. These mice were together with C57BL/6 wild-type mice kept in-house at the local animal care facility of the Friedrich-Alexander-University Erlangen-Nürnberg (Erlangen, Germany) under specific pathogen-free conditions. All experiments were performed in accordance with the German and European laws for animal protection and were approved by the local ethics committees of the Regierung Unterfranken (Az 55.2-2532.1-36/13).

For tumor induction, LL/2-luc-M38 cells were cultured in DMEM supplemented with 10% FCS, 1% pen/strep, and 1% l-Glu. A total of 1 × 10⁶ cells suspended in 200-μL DMEM (without supplements) were injected into the tail vein of 6- to 8-week-old female mice. At the indicated time points, mice were weighted and injected intraperitoneally (i.p.) with luciferin (0.15 mg/g body weight; Promega). After 20 minutes, luciferase activity was measured by the IVIS Spectrum In Vivo Imaging System (PerkinElmer). Luciferase activity was measured by detecting luminescence intensity (photons per second). Mice were anesthetized with isoflurane during the measurements. Lung tumor load analysis was performed in a logarithmic scale mode and the total flux (photons per second) was determined as described previously (29). Blocking of PD-1 in vivo was performed by injecting αPD-1 mAb RMP1-14 intraperitoneally (Hölzel Diagnostika) or the respective rat IgG2a mAb 2A3 isotype control (Hölzel Diagnostika) at 150 μg/mouse every 3 days for maximal four injections starting at day 9 postintravenous injection of LL/2 cells. At day 20 post intravenous injection, total lung cells were isolated as described previously (30) and in vitro rechallenged with αPD-1 mAb RMP1-14 (10 μg/mL; Hölzel Diagnostika) or the respective rat IgG2a mAb 2A3 isotype control (10 μg/mL; Hölzel Diagnostika) for 24 hours followed by flow cytometry analyses.

Lungs were removed, fixed in 10% formalin–PBS solution, dehydrated, and embedded in paraffin. Five-micrometer–thick lung sections from paraffin blocks were stained with hematoxylin and eosin for visualization of lung tumors.

Single-cell suspensions from murine lungs were prepared as described previously (30). For total cell isolation from spleen and dLNs, the same protocol was used without collagenase/DNAseI digestion. Total cells were incubated with the respective mix of surface antibodies dissolved in PBS and incubated for 30 minutes at 4°C. For intracellular staining, cells were fixed and permeabilized with Fixation/Permeabilization concentrate/diluent in accordance to the manufacturer's protocol (eBioscience). Intracellular staining of NFATc1 was performed as described in a previous section. For cytokine immunofluorescence analysis, cells were stimulated for 4 hours with ionomycin (1 μmol/L, Sigma-Aldrich) and PMA (50 ng/mL, Sigma-Aldrich) in the presence of the Golgi inhibitor monensin (2 μmol/L, eBioscience). Antibodies for intracellular staining were dissolved in permeabilization buffer (eBioscience) and incubated for 30 minutes at 4°C. Antibodies used for flow cytometry are shown in Supplementary Table S1. Flow cytometric analyses were performed by using FACSCalibur and FACSCanto II (BD Biosciences). Datasets were analyzed by Cell Quest Pro version 4.02 (BD Biosciences) and Flow-Jo v10.2 (FlowJo, LLC).

Single-cell suspensions from murine lungs were prepared as described previously (30). CD8⁺ T cells were isolated from the lungs of naïve and tumor-bearing mice by magnetic cell separation using the CD8a (Ly-2) MicroBeads Mouse Kit (Miltenyi Biotec) in accordance with the manufacturer's instructions. The purity of the isolated CD8⁺ T cells was confirmed by FACS analysis. CD8⁺ T cells were then cultured in RPMI1640 medium (Gibco) containing 10% FCS, 1% pen/strep, and 1% l-Glu and stimulated with αCD3 (10 μg/mL; BD Biosciences) and αCD28 (1 μg/mL; Biolegend) antibodies. After 48 hours, CD8⁺ T cells were harvested and resuspended in 500 μL 75% cold EtOH added dropwise. Cells were incubated for 24 hours at −20°C. Fixed cells were then washed twice with 1-mL wash buffer (PBS with 1% FCS, 0.09% NaN₃ pH 7.2) and stained with PI (BD Biosciences) staining solution (1:85 in wash buffer), followed by flow cytometric analyses using FACSCanto II (BD Biosciences).

Statistics were computed with GraphPad Prism7. The unpaired t test was performed for parametric data containing no more than two groups. Data are presented as mean ± SEM and significance levels indicated as follows: *, P < 0.05; **, P < 0.01; ***, P < 0.001.

In this study, we started to investigate the role of NFATc1 in NSCLC by analyzing its mRNA expression in the tumoral (TU, solid tumor), peritumoral (PT, 2 cm around the solid tumor) and control lung region (CTR, tumor-free control area) of our cohort of patients with lung adenocarcinoma (ADC) and lung squamous cell carcinoma (SCC), collectively grouped as NSCLC (Table 1). Here we found a significantly decreased expression of NFATC1 mRNA in the TU region of patients with both ADC and SCC as compared with the respective PT and CTR region (Fig. 1A). Accordingly, higher amounts of total NFATc1/β-actin protein levels were found in the CTR and PT region of patients with ADC, whereas in the TU region, total NFATc1/β-actin protein levels decreased (Fig. 1B; Supplementary Fig. S1A). In addition, we analyzed the phosphorylation status of NFATc1 (pNFATc1) as dephosphorylation of NFAT proteins leads to their activation in terms of nuclear translocation (3). As a result, we found no differences in pNFATc1/β-Actin and pNFATc1/NFATc1 protein levels in the CTR, PT, and TU regions (Fig. 1B; Supplementary Fig. S1A). We next analyzed the NFATC1 mRNA expression in different disease stages according to the TNM classification describing the size of the primary tumor, the degree of spread to regional lymph nodes, and the presence of distant metastasis. We found that, in the tumoral area, NFATC1 mRNA expression decreased as the tumor stadium progressed (Fig. 1C). Moreover, using IHC double staining for NFATc1 and CD3 on lung tissue arrays obtained from our cohort of patients, we demonstrated a downregulation of NFATc1 in CD3⁺ T cells in the tumoral region of advanced NSCLC stages (Fig. 1D; Supplementary Fig. S1B). Furthermore, flow cytometry revealed an increase of NFATc1 in CD8⁺ T cells in the PT region as compared with the respective TU region of patients with ADC (Fig. 1E), whereas in CD4⁺ T cells, no differences of NFATc1 could be observed (Supplementary Fig. S1C). In addition, NFATc1 was found to be increased in CD8⁺PD-1⁺ T cells (Fig. 1F) in the CTR and PT region as compared with the respective TU region. No regional regulation was observed in the distribution of NFATc1 in CD4⁺PD-1⁺ T cells (Supplementary Fig. S1D). Regarding CD4⁺ T cells, we observed a strong direct correlation of NFATC1 and CD4 expression in the PT region (Fig. 1G), whereas no direct correlation was found in the CTR and TU region (Supplementary Fig. S1E). Moreover, NFATC1 directly correlates with TBX21 expression (Fig. 1H), a gene that codes for Tbet (T-box transcription factor TBX21), the main transcription factor controlling IFNγ production in T cells (31). In general, these results indicate a downregulation of NFATc1 in T cells in the presence of NSCLC and at the advanced disease stage. Furthermore, NFATc1⁺PD1⁺CD8⁺ T cells were found induced in the CTR and PT region of the lung of patients with ADC, confirming recent findings that T cells in these patients are exhausted (32) and thus represent the target for the αPD-1-tumor immunotherapy. Because NFATc1 controls the IL2 promoter, we next correlated NFATC1 with IL2 and its receptor subunits IL2Rα (CD25), IL2Rβ (CD122) and IL2Rγ (common cytokine receptor γ, CD132; ref. 33). We found that in the TU region, NFATC1 expression directly correlates with IL2 as well as CD122 and CD132 (Fig. 1I) but not with CD25 (Supplementary Fig. S1F). The IL2Rα chain is present on T effector as well as T regulatory cells. In contrast, the IL2Rβ and IL2Rγ chain are characteristic for memory T cells (34, 35), indicating a potential role of NFATc1 in T-cell memory expansion. In accordance to this, we found a direct correlation of NFATC1 with the T memory cell markers IL7 in the CTR region as well as CD127 (IL7Rα; ref. 36) in both the CTR as well as TU region of patients with NSCLC (Fig. 1J).

Likewise PD-1, which is highly expressed on TILs in many cancer types, its ligand PD-L1 is commonly upregulated on the tumor cell surface of different human tumors including lung cancer (37, 38). On the basis of this notion, we analyzed whether there is a relation between NFATc1 and PDL-1 in NSCLC development. We found a direct correlation between NFATc1 and PD-L1 in the TU region of patients suffering from NSCLC (Fig. 2A). Furthermore, lung cancer is of epithelial cell origin and thus, we analyzed NFATc1 in epithelial cells of the lung of these patients by using an epithelial-specific antibody (anti-EpCAM). This analysis showed decreased NFATc1⁺PDL-1⁺EpCAM⁺ cells in the TU region of patients with ADC, whereas in the CTR region, a modest proportion of EpCAM⁺ cells expressing both NFATc1 and PD-L1 could be detected (Fig. 2B). We next analyzed the influence of NFATc1 on the response to PD-L1 inhibitors in lung tumor cells. To this aim, we knocked down the expression of NFATC1 in the human lung adenocarcinoma cell line A549 using NFATC1-directed siRNA (siNFATc1) followed by treatment with an αPD-L1 antibody (αPD-L1; Fig. 2C). In this experimental setting, we achieved a specific NFATC1-knockdown as compared with nontargeting siRNA (siNT) after both αPD-1 and IgG2b isotype control treatment (Fig. 2D). Moreover, αPD-L1 antibody treatment of A549 successfully blocked PD-L1, indicated by a decreased proportion of PD-L1⁺PD-1⁻ A549 cells, whereas siRNA-mediated NFATC1 knockdown did not influence PD-L1 on A549 (Fig. 2E, left). Notably, we only detected a low proportion of PD-1⁺PD-L1⁻ A549 cells (Fig. 2E, right). To get a first indication whether the loss of NFATC1 in tumor cells effects their growth, we counted A549 cells and found that NFATC1 knockdown with or without αPD-L1 antibody treatment did not affect the overall numbers of living A549 (Fig. 2F). Furthermore, analysis of apoptosis induction using Annexin (Ann)V/PI staining revealed a slight increase of AnnV⁺PI⁺ late apoptotic A549 cells after siRNA-mediated NFATC1-knockdown in combination with PD-L1 blockade compared with A549 cells transfected with siNT and treated with IgG2b (Fig. 2G). In addition, αPD-L1 antibody treatment of A549 cells, in which NFATC1 was knocked down, induced a slight expansion of AnnV⁻PI⁺ dead A549 cells as compared with the IgG2b treatment (Fig. 2G). As we detected only a modest expression of PD-L1 by A549 cells (Fig. 2E), we subsequently stimulated them with IFNγ followed by siRNA-mediated NFATC1-knockdown and αPD-1 antibody treatment (Fig. 2H). IFNγ has been shown to induce PD-L1 in human airway epithelial cells (39). Furthermore, stimulation of A549 cells with IFNγ mimics host immune responses as this cytokine is predominantly produced by T helper 1 (Th1) cells as well as cytotoxic CD8⁺ T cells to mediate antitumoral immune responses (40). In this experimental setting, downregulation of NFATc1 mRNA was observed by siRNA-mediated NFATC1 knockdown and maintained after αPDL-1 antibody treatment (Fig. 2I). While stimulation of A549 cells with IFNγ induced a strong increase in PD-L1⁺PD-1⁻ A549 cells (Fig. 2J), siRNA-mediated NFATC1 knockdown did not influence the growth of A549 tumor cells neither in combination with αPD-L1 nor with IgG2b (Fig. 2K). Regarding PD-1 expression, again we only detected a low proportion of PD-1⁺PD-L1⁻ A549 cells (Fig. 2J). Altogether, NFATc1 expressed in tumor cells has no influence, neither on tumor cell growth, nor on their response to αPD-L1 antibody treatment.

As we demonstrated that NFATc1 expression in tumor cells has no influence on the tumor cell growth and response to αPD-L1 antibody treatment (Fig. 2), we started to investigate the role of NFATc1 in T cells in a murine model of lung adenocarcinoma, a condition that closely resembles the human condition in NSCLC. Lung tumor development was induced by intravenous injection of LL/2-luc-M38 (LL/2) cells followed by the analysis of tumor growth via an in vivo bioluminescence-based imaging system (Fig. 3A). Here we observed a downregulation of NFATc1α in CD4⁺ T cells isolated from the lung of mice suffering from lung adenocarcinoma (LL/2) as compared with those isolated from naïve control mice (naïve; Fig. 3B), indicating that NFATc1α mediates anti-tumoral effector functions in T cells. On the basis of this observation, we went on and analyzed lung adenocarcinoma development in mice with a conditional inactivation of NFATc1 in CD4⁺ and CD8⁺ T cells (NFATc1^ΔCD4) compared with control littermates (NFATc1^fl/fl). Here, we found that targeted deletion of Nfatc1 in T cells resulted in increased lung tumor growth as shown by bioluminescence in vivo imaging (Fig. 3C) and histologic analysis indicated increased infiltration of tumor cells into the lung (Fig. 3D). Furthermore, we could demonstrate that exon 3 of Nfatc1 was successfully deleted in T cells of NFATc1^ΔCD4 mice as shown by decreased Nfatc1ex3 mRNA expression in isolated lung CD8⁺ T cells (Fig. 3E) and splenic CD4⁺ T cells (Fig. 3F).

NFATc1 is highly expressed in peripheral T cells and important for their activation and cytotoxic function (5, 7). Thus, we next analyzed whether the increased tumor growth in NFATc1^ΔCD4 mice could possibly be explained by reduced antitumoral T-cell–mediated immune responses. As effector CD8⁺ T cells have antitumor functions (10, 11), we next analyzed the number of CD8⁺ TILs in the lungs of NFATc1^fl/fl and NFATc1^ΔCD4 mice. We found a reduced proportion CD8⁺ T cells in the lungs of NFATc1^ΔCD4 mice (Fig. 4A). In NFATc1^ΔCD4 mice bearing lung tumor, these CD8⁺ T cells are characterized by a decreased expression of CD25 (IL2Rα; Fig. 4B) as well as by a diminished proliferative capacity indicated by a reduced proportion of CD8⁺ T cells in the G₂–M phase of the cell cycle (Fig. 4C). Furthermore, while we found no differences in the IL2 production by lung CD8⁺ T cells, we could show a strong decrease in IL2 produced by CD8⁺ T cells in the spleen of NFATc1^ΔCD4 tumor-bearing mice (Fig. 4D). A reduced production of IL2 by CD8⁺ T cells is a characteristic feature of T-cell exhaustion (41). Moreover, CD8⁺ memory T cells relay on autocrine IL2 production, which enables optimal secondary population expansion and promotes a robust response in advanced tumor-bearing states (13). Memory CD8⁺ T cells are heterogeneous with respect to phenotypic markers, effector functions, and tissue-homing capabilities and are classified into central memory T cells (TCM), effector memory T cells (TEM), and tissue-resident memory T cells (TRM). TCM cells reside in secondary lymphoid organs and provide protection from a systemic challenge while TEM cells are highly reactive for immediate protection and circulate between lymphoid organs and peripheral tissues or inflammatory sites (15). In the spleen of tumor-bearing NFATc1^ΔCD4 mice, we found an increase in TCM cell populations and a trend toward reduction of TEM cells (Fig. 4E). Furthermore, in the tumor-bearing lung of NFATc1^ΔCD4 mice TEM cells were significantly decreased (Fig. 4F) underlining an impaired cytotoxic T-cell response in these mice. TRM cells occupy tissues and mucosal sites such as the lung and are characterized by the lack of recirculation via the bloodstream, which is important for long-term regional immunity. The surface molecule integrin αE CD103 is linked to TRM cells and promotes their localization to epithelia (16). Furthermore, NFATc1 has been shown to induce the expression of the Itgae gene encoding CD103 (9). Consistently, we found a decreased proportion of CD103⁺ TRM cells in the lung of tumor-bearing NFATc1^ΔCD4 mice (Fig. 4G), indicating an impaired tissue-homing capacity by reduced CD103. Furthermore, in a recent study, a greater density of CD103⁺ TRM cells has been linked to an enhanced cytotoxicity of CD8⁺ T cells in NSCLC (42). Thus, a reduction in CD103⁺ TRM cell populations is accompanied by decreased cytotoxicity of CD8⁺ T cells, a situation that we found in the lungs of tumor-bearing NFATc1^ΔCD4 mice.

In general, IL2 signals during different phases of an immune response are key in optimizing CD8⁺ effector as well as memory T-cell functions. A loss of NFATc1 affects the production of IL2 as well as of CD103, which influences antitumor effector functions of CD8⁺ T cells and the tissue-homing capacity of TRM cells.

As we found a decreased expression of NFATc1α in lung CD4⁺ T cells of tumor-bearing wild-type mice (Fig. 3B), we analyzed the influence of the conditional inactivation of NFATc1 in T cells on CD4⁺ T-cell properties in NFATc1^ΔCD4 mice. While we could not detect a difference in the proportion of lung CD4⁺ T-cell populations neither in naïve nor in tumor-bearing NFATc1^ΔCD4 mice (Supplementary Fig. S2A), we found that in the absence of NFATc1, lung CD4⁺ T cells express a lower amount of the activation marker CD25 (IL2Rα; Fig. 5A). Furthermore, lung CD4⁺ T cells of tumor-bearing NFATc1^ΔCD4 mice, but not of naïve mice, are characterized by a reduced production of IL2 (Fig. 5B; Supplementary Fig. S2B), important for T-cell proliferation, survival, and activation (14). The observed reduction of IL2 production by lung CD4⁺ T cells was even more dramatically reduced by CD4⁺ T cells in the spleen of tumor-bearing NFATc1^ΔCD4 mice (Fig. 5B). In addition, reduced CD4⁺ T-cell proportions in the spleen and the draining lymph nodes (dLN) of naïve and tumor-bearing NFATc1^ΔCD4 mice (Fig. 5C; Supplementary Fig. S2C) indicate an impaired CD4⁺ T-cell development in secondary lymphoid organs. Moreover, we detected an increase of PD1 on ICOS⁺CD4⁺ T cells in the dLNs of tumor-bearing NFATc1^ΔCD4 mice (Fig. 5D). We hypothesized that the PD1⁺ICOS⁺CD4⁺ T-cell population represents lymphocytes that are inhibited in their cytotoxic properties by T-cell exhaustion, which dampened antitumoral immune responses and promotes lung tumor growth. Therefore, we analyzed whether soluble factors present in the supernatants of cells isolated from the dLNs of tumor-bearing NFATc1^ΔCD4 mice (=lymph node-conditioned medium, LNCM) would influence the survival of LL/2 lung tumor cells (Fig. 5E). We could show that, LNCM of tumor-bearing NFATc1^ΔCD4 mice significantly increased LL/2 cell numbers as compared with LNCM of tumor-bearing NFATc1^fl/fl control littermates (Fig. 5F). In addition, we analyzed the influence of supernatants of total lung cells (=lung-conditioned medium, LUCM), but could not observe any effect on LL/2 cell proliferation (Supplementary Fig. S2D and S2E). Thus, LNCM but not LUCM, of NFATc1^ΔCD4 mice contain soluble factors that promote LL/2 lung tumor cell survival. To start to identify this factor, we analyzed the presence of different cytokines and growth factors. While VEGFA, IFNγ, and IL10 were not found to be differentially regulated (Supplementary Fig. S2F–S2H), we observed a decrease of TNFα, IL2, and IL7 in LNCM of NFATc1^ΔCD4 mice (Fig. 5G–I). Moreover, we could show that TNFα, an antitumoral cytotoxic molecule, induces apoptosis in LL/2 cells (Fig. 5J). In summary, during lung adenocarcinoma development, the loss of NFATc1 in CD4⁺ T cells resulted in decreased CD4⁺ T-cell activation and IL2 production in both the lung and secondary lymphoid organs. Furthermore, we could show that NFATc1 upregulates antitumor cytokines in the dLNs of tumor-bearing NFATc1^ΔCD4 mice.

In our cohort of patients, we observed an increase of NFATc1 in CD8⁺PD-1⁺ T cells in the CTR and PT region (Fig. 1F) confirming recent findings that T cells in these patients are exhausted (32) probably because of chronic activation via the tumor antigen. They thus represent the target for αPD1-tumor immunotherapy. To analyze whether a blockade of PD-1 influences NFATc1 in T cells, we continued treating lung tumor–bearing B6 wild-type (WT) mice with αPD-1 antibodies (αPD-1; Fig. 6A). In terms of survival (Fig. 6B) and weight change (Fig. 6C), we could not observe any differences between WT mice treated with αPD-1 or the respective IgG2a isotype control (IgG2a). Analysis of lung tumor load revealed a trend toward reduced tumor development in αPD-1–treated WT mice at day 15 postinjection (p.i.), which was declined at day 18 p.i. (Fig. 6D). As we could not observe a difference in lung cancer development after αPD-1 antibody treatment, we rechallenged total cells from the lungs of these mice with αPD-1 or IgG2a in vitro for 24 hours (Fig. 6A). Rechallenging the lung cells in vitro with αPD-1 antibodies induced a strong increase of NFATc1 in both CD4⁺ and CD8⁺ T cells from about 70% up to 90% (Fig. 6E and F; Supplementary Fig. S3A and S3B). Afterwards, we applied supernatants of the rechallenged total lung cells (LUCM) in a cytotoxicity assay with LL/2 tumor cells used to induce lung tumor growth in mice (Fig. 6G). As a result, we observed a reduction in LL/2 cell growth induced by LUCM of total lung cells obtained from mice bearing tumor and in vitro rechallenged with αPD-1 as compared with IgG2a (Fig. 6H). Collectively, these data demonstrate a strong induction of NFATc1 in T cells by αPD-1 antibody treatment, which seems to inhibit LL/2 lung tumor growth via a soluble released factor. These are very supportive and challenging results in T cells as a new therapeutic avenue for NSCLC.

In our current study, we demonstrated an important function of NFATc1 for successful T-cell–mediated antitumoral immune responses in the setting of NSCLC. In our human patient cohort, we found a downregulation of NFATc1 in T cells in the presence of NSCLC and at advanced disease stages, indicating that a loss of NFATc1 in T cells is associated with poor prognosis. Upon T-cell activation, NFATc1 regulates the expression of numerous genes controlling the activity and fate of T cells. Among them IL2, which has been shown to be important for T-cell activation, proliferation, and survival as well as for CD8⁺ effector and memory T-cell generation (9, 14, 17). Accordingly, we found a direct correlation between NFATc1 and IL2 as well as its receptor subunits CD122 and CD132, but not CD25. While immunosuppressive Treg cells are characterized by the expression of all three IL2-R subunits, CD122 and CD132 are characteristic for memory T cells and NK cells (34, 35). Among fate markers for memory T cells, the IL7R genes are NFATc1 targets (9). According to that, NFATc1 correlates with the IL7Rα chain as well as with IL7 implicated in CD8⁺ memory T-cell differentiation and homeostasis (9, 36). In addition, NFATc1 directly correlates with Tbet, the main transcription factor controlling IFNγ that is relevant for the cytotoxic antitumor activity of TIL (31), which was found downregulated in the CD8⁺ T cells isolated from the tumoral region of our cohort of patients with NSCLC (27). Thus, in the tumoral and control region of patients with NSCLC, NFATc1 correlates with factors that are characteristic for cytotoxic and memory T-cell responses and seems to be important for antitumoral immunity.

In our murine model of lung adenocarcinoma, we found a decreased expression of NFATc1α in lung CD4⁺ T cells of tumor-bearing wild-type mice. Moreover, targeted deletion of NFATc1 in T cells (NFATc1^ΔCD4) resulted in increased lung tumor growth, which was associated with an impairment of CD4⁺ and CD8⁺ T cells consistent with the importance of NFATc1 for T-cell activation and cytotoxic T-cell functions (5, 7). We further detected reduced CD8⁺ T-cell proportions as well as decreased proliferation of CD8⁺ T cells in the tumor-bearing lung of NFATc1^ΔCD4 mice. The production of IL2 by lung CD8⁺ T cells was not influenced, while splenic CD8⁺ T cells were characterized by a strongly decreased IL2 secretion. IL2 contributes to the primary as well as secondary expansion of CD8⁺ T cells. As the magnitude of T-cell expansion defines the number of memory CD8⁺ T cells, IL2 influences memory cell generation. Furthermore, at the memory state, CD8⁺ T-cell frequencies can be strengthened by administration of IL2 (13, 17). Consistently, we found a reduction of TEM cell populations in the spleen and the lung of tumor-bearing NFATc1^ΔCD4 mice underlining an impaired cytotoxic T-cell response in these mice. Interestingly, splenic TCM cells of tumor-bearing NFATc1^ΔCD4 mice were significantly increased. The capacity to secrete IL2 has been associated with TCM cells rather than TEM cells (15). Thus, even though we detected an increased TCM cell population in the spleen of tumor-bearing NFATc1^ΔCD4 mice, we suggest that these cells are ineffective in secreting IL2 as indicated by a reduced IL2 production by splenic CD8⁺ T cells. This reduced IL2 production could inhibit the development of splenic TEM cells as IL2 signals are able to rescue CD8⁺ T cells from cell death and provide an increase in CD8⁺ memory T-cell counts (17). Furthermore, as TEM cells circulate between lymphoid organs and peripheral tissue, a decreased splenic TEM cell population is accompanied by a reduction in lung TEM cells in tumor-bearing NFATc1^ΔCD4. A recent study described increased TCM and decreased TEM cell populations in the spleen of Listeria monocytogenes–infected NFATc1^ΔCD4 (9), which is similar to the results we obtained in the murine lung adenocarcinoma model. Beside TEM and TCM cell populations, we also analyzed the proportion of TRM cells, a unique tissue-resident memory T-cell subset important for enhanced regional immunity. TRM cells in various nonlymphoid tissues are characterized by the integrin αE CD103, which promotes the localization to epithelia (16). In the tumor-bearing lung of NFATc1^ΔCD4 mice, we found a strong decrease of CD103⁺ TRM cells consistent with the finding that NFATc1 has been shown to induce the expression of the Itgae gene encoding CD103 (9). These results indicate that the tissue-homing capacity of lung TRM cells of tumor-bearing NFATc1^ΔCD4 mice is impaired. In addition, in a recent study a greater density of CD103⁺ TRM cells has been linked to an enhanced cytotoxicity of CD8⁺ T cells and was predictive for a better survival outcome of patients with lung cancer (42). Thus, a reduction in CD103⁺ TRM cell populations is accompanied by a decreased function and cytotoxicity of CD8⁺ T cells, a situation that we found in the lungs of tumor-bearing NFATc1^ΔCD4 mice mainly promoted by decreased IL2.

Furthermore, we found decreased CD4⁺ T-cell proportions that produced reduced amounts of IL2 in secondary lymphoid organs of tumor-bearing NFATc1^ΔCD4 mice, indicating impaired T-cell activation and effector functions also in terms of CD4⁺ T cells. Interestingly, in the dLN of tumor-bearing NFATc1^ΔCD4 mice, we found a strongly increased proportion of PD-1 on ICOS⁺CD4⁺ T cells. High expression of PD-1 is a key feature of T-cell exhaustion, a state of T-cell dysfunction defined by inhibited cytotoxic properties (18, 41). Furthermore, ICOS has been described to be crucial for the expansion and suppressive activity of Foxp3⁺ Treg cells as well as for IL10 production (43). Thus, we hypothesized that the PD1⁺ICOS⁺CD4⁺ T-cell population represents highly suppressive lymphocytes affecting T-cell effector functions and inhibiting antitumoral immune responses while promoting tumor growth. This was supported by the fact that lymph node–conditioned medium (LNCM) of tumor-bearing NFATc1^ΔCD4 mice induced LL/2 tumor cell survival. Consistently, we found a decreased production of TNFα. TNFα is a pleiotropic cytokine with pro- and antitumoral functions as it regulates diverse events like proliferation, invasion, and apoptosis (44). We could show that TNFα induced apoptosis in LL/2 lung tumor cells, underlining its ability to suppress tumor cell proliferation and induce tumor regression in our murine model of lung adenocarcinoma. Moreover, reduced proportion of IL7 in LNCM confirms impaired memory T-cell responses in the absence of NFATc1.

Immune checkpoint blockade with αPD-1 or αPD-L1 has proven to be a highly promising treatment of human cancers, including lung cancer (19, 22, 23). As lung cancer is of epithelial cell origin and NFATc1 has been described to be expressed in mouse lung epithelial cells (45), we analyzed whether there is a correlation of NFATc1 and PD-L1 in EpCAM⁺ epithelial cells in the different regions of patients with NSCLC. We found a direct correlation between NFATc1 and PDL-1 in the TU region of patients with NSCLC, whereas NFATc1⁺PDL-1⁺EpCAM⁺ cells decreased, indicating that epithelial cells expressing NFATc1 together with PD-L1 are not present in the TU region of the lung. Furthermore, using NFATC1-directed siRNA followed by αPD-L1 antibody treatment of the human lung adenocarcinoma cell line A549, we could show that, NFATc1 expressed in tumor cells has no influence on tumor cell growth and their response to αPD-L1 antibody treatment. After this observation, we focused on NFATc1 and PD-1 expression in T cells in the setting of NSCLC. We found that NFATc1 is induced in CD8⁺PD-1⁺ T cells of the CTR and PT region of the lung of patients with ADC. In the tumor microenvironment, tumor antigens induce a durative activation of T cells, which probably contributes to T-cell exhaustion via PD-1 and promotes the impairment of effector functions (41, 46). As NFATc1 is important for T-cell activation (3, 9), and has been shown to induce the expression of PD-1, (24) the NFATc1⁺PD-1⁺CD8⁺ T cells could represent exhausted T cells that are targets for αPD1-tumor immunotherapy. Indeed, we could show that rechallenged total lung cells of tumor-bearing wild-type mice with αPD-1 antibodies induced a strong increase of NFATc1 from about 70% up to 90% in T cells. This strong increase could be explained by the PD-1 signaling pathway: triggering of PD-1 by its ligand PD-L1 expressed on tumor cells induces the inhibition of the PI3K/Akt pathway. Akt normally negatively regulates the GSK3, which inhibits NFATc1 (3, 25, 26). Thus, the blockade of PD-1 restores TCR signaling, ensured the activation of Akt, which inhibits GSK3 and promotes the activation of NFATc1, which induces T-cell activation and effector functions (7, 9, 25). Although preliminary, this is a very promising avenue: to challenge T cells of patients with α-PD-1 antibodies ex vivo to induce NFATc1 and promote T-cell activation followed by the transfer of these cells back into the patients to boost antitumoral T-cell responses. In fact, supernatants of lung cells rechallenged with α-PD-1 antibodies inhibited LL/2 lung tumor cell growth supporting the importance of NFATc1 for T-cell–mediated cytotoxic immune responses.

No potential conflicts of interest were disclosed.

Conception and design: L. Heim, S. Finotto

Development of methodology: L. Heim, J. Friedrich, M. Engelhardt, S. Finotto

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): L. Heim, J. Friedrich, D.I. Trufa, C.I. Geppert, R.J. Rieker, H. Sirbu, S. Finotto

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): L. Heim, S. Finotto

Writing, review, and/or revision of the manuscript: L. Heim, S. Finotto

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): L. Heim, D.I. Trufa, C.I. Geppert, R.J. Rieker, S. Finotto

Study supervision: S. Finotto

Other (collection of human samples and cell isolation): D.I. Trufa

This work was supported by the SFB643 (Project B12) in Erlangen and by the Department of Molecular Pneumology. We thank the whole team at the Department of Molecular Pneumology, especially Sonja Trump, Susanne Mittler, and Adriana Geiger for their technical help, as well as the team at the Department of Thoracic Surgery and the Institute of Pathology at the University Hospital, Friedrich-Alexander-Universität Erlangen-Nürnberg, for their help and support to this work. Furthermore, we thank Dr. Paolo Ceppi at the Interdisciplinary Center for Clinical Research at the Friedrich-Alexander-Universität Erlangen-Nürnberg for giving us the opportunity to perform apoptosis analysis using the IncuCyte live-cell analysis system.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.