Intratumoral cytotoxic CD8+ T cells (CTL) enter a dysfunctional state characterized by expression of coinhibitory receptors, loss of effector function, and changes in the transcriptional landscape. Even though several regulators of T-cell exhaustion have been identified, the molecular mechanisms inducing T-cell exhaustion remain unclear. Here, we show that IL18 receptor (IL18R) signaling induces CD8+ T-cell exhaustion in a murine pancreatic cancer model. Adoptive transfer of Il18r−/− OT-1 CD8+ CTLs resulted in enhanced rejection of subcutaneous tumors expressing ovalbumin (OVA) as a model antigen (PancOVA), compared with wild-type OT-1 CTLs. Transferred intratumoral IL18R-deficient CTLs expressed higher levels of effector cytokines TNF and IFNγ and had reduced expression of coinhibitory receptors (PD-1, TIM-3, 2B4, LAG-3) and the transcription factors Eomes and TOX. Lower expression of coinhibitory receptors and TOX on IL18R-deficient versus IL18R-sufficient CD8+ T cells were confirmed in an orthotopic KPC model. IL18R-induced T-cell exhaustion was regulated by IL2/STAT5 and AKT/mTOR pathways, as demonstrated in an in vitro exhaustion assay. Concordantly, mice deficient in NLRP3, the molecular complex activating IL18, had decreased expression of coinhibitory receptors on intratumoral T cells and similar changes in signaling pathways at the transcriptome level. Thus, molecular pathways promoting T-cell exhaustion indicate an involvement of an NLRP3-expressing tumor microenvironment, which mediates IL18 release. The Cancer Genome Atlas analysis of patients with pancreatic carcinoma showed an association between NLRP3-mediated IL18 signaling and shorter survival. These findings indicate NLRP3-mediated IL18R signaling as a regulator of intratumoral T-cell exhaustion and a possible target for immunotherapy.

See related Spotlight by Stromnes, p. 400

With a 5-year overall survival rate of 10% pancreatic carcinoma is one of the deadliest cancer types worldwide (1). Bad prognosis is caused by late diagnosis and drug resistance to established chemotherapy (2) and immunotherapy (3). Higher levels of intratumoral CD8+ T cells correlate with better survival in patients with pancreatic carcinoma (4); however, due to low immunogenicity and a highly immunosuppressive tumor milieu, the clinical course of pancreatic cancer demonstrates that antitumor T-cell responses are strongly impaired. The tumor microenvironment of pancreatic carcinoma is characterized by immunosuppressive cell populations including macrophages (5), myeloid-derived suppressor cells (6), and regulatory T cells (7). In addition, the low mutational burden of pancreatic carcinoma cells results in poor antigenicity and lack of neoantigens (8). Consequently, pancreatic carcinoma is characterized by exhaustion of intratumoral CD8+ T cells.

T-cell exhaustion was first observed in models of viral infection (9), but has also been described in tumor immunology. It is triggered by persistent T-cell receptor activation and characterized by the expression of multiple coinhibitory receptors, such as PD-1, TIM-3, LAG-3, CTLA-4, TIGIT, and 2B4 (10). In this dysfunctional state, transcriptional and epigenetic profiles are changed, causing loss of T-cell effector function (11). Therefore, identifying signaling pathways that suppress and potentially overcome T-cell exhaustion is crucial for breaking the immunosuppressive tumor microenvironment of pancreatic cancer.

Proinflammatory IL18, a member IL1 cytokine family, is released by several cell types in the pancreatic tumor microenvironment, including monocytes, macrophages, dendritic cells, and epithelial cells (12). Together with pro-IL1β, pro-IL18 is processed and biologically activated by the NLRP3 inflammasome (13). IL18 receptor (IL18R) signaling is mediated by a heterodimeric complex; the ligand-binding-chain IL18R1 and the coreceptor chain IL18R2 (14). IL18R1 alone has weak affinity for IL18 binding, but coupling of IL18R1 and IL18R2 results in high affinity of IL18. IL18 signaling can induce multiple pathways that regulate cell proliferation, survival, and functionality (12). Together with IL12, it induces IL2 receptor expression, enabling natural killer (NK) cells to respond to IL2 (15). IL18 promotes MAPK cascade activation, inducing STAT3 activation and IFNγ production (16) and activates NFκB, resulting in proinflammatory cytokine and chemokine production (17). Its activation can be induced by cross-talk with the phosphatidylinositol-kinase/AKT/mTOR pathway, a central mediator of cell proliferation, metabolism, growth, and survival (18, 19).

The role of IL18R signaling in antitumor responses remains ambiguous. IL18 has been shown to inhibit tumor development by inducing production of proinflammatory cytokines such as IFNγ and enhancing cytotoxic effector functions (20). However, correlating survival time and IL18 levels in serum indicates that IL18 is associated with poor prognosis in patients with pancreatic cancer (21). Furthermore, tumor-derived IL18 induces cytotoxic T-cell inhibition by enhancing PD-1 expression on immunosuppressive regulatory B cells in pancreatic cancer (22). IL18R downregulation in exhausted CD8+ T cells might indicate a direct role for IL18R signaling in T-cell exhaustion (23).

In this study, we show that IL18R signaling in CD8+ T cells increased T-cell exhaustion in a murine model of pancreatic carcinoma by enhancing the expression of coinhibitory receptors, suppressed T-cell effector function and decreased tumor rejection. Mechanistically, we found that changes in the IL2/STAT5 and AKT/mTOR pathways induced by IL18R signaling contributed to induction of T-cell exhaustion. Active NLRP3 in the tumor microenvironment induced upregulation of coinhibitory receptors and changed molecular pathways with similarities to IL18R-mediated exhaustion. Thus, we provide data indicating that the NLRP3–IL18R–STAT5–AKT/mTOR pathway is an important mediator of intratumoral T-cell exhaustion and a possible new target for pancreatic cancer immunotherapy.

Cell lines

Murine pancreatic adenocarcinoma (PAAD) cell lines Panc02 (RRID:CVCL_D627) and PancOVA were gifts from Max Schnurr in 2015 (University Munich, Munich, Germany; ref. 24). KPC cancer cell line, derived from KPC mouse tumors, was a gift from Shiv Singh in 2022 (University Göttingen, Göttingen, Germany; ref. 25). The cell lines were not authenticated and were routinely tested for Mycoplasma contamination. Tumor cells were cultured in T75 flasks (Sarstedt) in DMEM (Gibco, catalog no. 41965062) with 10% FCS (Capricorn Scientific, catalog no. FBS-11A). For selection of PancOVA cells, 500 mg/L G418 (Sigma-Aldrich, catalog no. G8168-10ML) was added. Experiments were performed using passages 3 to 20.

Mice

C57BL/6J [wild-type (WT), RRID:IMSR_JAX:000664], CD45.1 C57BL/6J (RRID:IMSR_JAX:002014), OT-1 TCR transgenic (RRID:IMSR_JAX:003831), Il18r−/− (B6.129P2-Il18r1tm1Aki/J, RRID:IMSR_JAX:004131), and Il1r−/− (B6.129S7-Il1r1tm1Imx/J, RRID:IMSR_JAX:003245) mice, all on C57BL/6J CD45.2 background, were purchased from Jackson Laboratories. Il18r−/− mice were provided by Natalio Garbi (University Bonn, Bonn, Germany). Il1r−/− and Il18r−/− mice were mated with OT-1 mice. Nlrp3−/− (B6.129S6-Nlrp3tm1Bhk/J, RRID:IMSR_JAX:021302) mice were a gift from Max Schnurr (University Munich, Munich, Germany). Mouse experiments were approved by the Regional Agency (Regierungspräsidium Gießen).

Murine CTL in vitro differentiation

Lymph nodes and spleen were mashed through 30-μm strainers (Miltenyi Biotec, catalog no. 130-041-407), loaded for 1 hour with 5 μmol/L ovalbumin (OVA)257–264 peptide (Invivogen, catalog no. code vac-sin) and cultured for 2 days in T-cell medium, supplemented with 10 ng/mL recombinant mouse IL12 (rm-IL12; Peprotech, catalog no. 210-12-50UG), followed by 3 days of cultivation with 20 ng/mL rm-IL2 (Peprotech, catalog no. 212-12). As T-cell medium, RPMI (Gibco, catalog no. 21875-091) was used, supplemented with 10% FCS (Capricorn Scientific, catalog no. FBS-11A), 2% P/S (Gibco, catalog no. 15070-063), 2% sodium pyruvate (Gibco, catalog no. 11360-070), 3% HEPES (Gibco, catalog no. 15630-056), and 0.1% β-mercaptoethanol (Gibco, catalog no. 31350010). Cells were stained for CD8, CD25, CD44, CD69, and CD62 L expression for 30 minutes in PBS (Gibco, catalog no. 14190-169, with 3% FCS) at 4°C. Flow cytometric analysis were performed on a Canto II cytometer (BD), CytoFLEX LX6 (Beckman Coulter), or Aria III (BD) and analyzed with FlowJo (BD, RRID:SCR_008520, version 10.5). The gating strategy is depicted in Supplementary Fig. S1A.

T-cell exhaustion model in vitro

OT-1 CD8+ T cells were isolated from spleen and lymph nodes using a negative selection kit (BioLegend, catalog no. 480035) according to the manufacturer's instructions. A total of 1 × 106 CD8+ T cells/well were placed in a 24-well F-bottom plate with plate-bound anti-CD3 (1 μg/mL; BioLegend, catalog no. 100202) and anti-CD28 (2 μg/mL; BioLegend, catalog no. 102102) in T-cell medium containing 20 ng/mL rm-IL2, with or without 10 ng/mL rm-IL18 (BioLegend, catalog no. 767006). Cells were activated for 48 hours at 37°C, and replated in a 96-well R-bottom plate (1 × 105 cells/well) as described above. For repetitive antigen stimulation, 10 ng/mL OVA257–264 was added. After 48 hours, cells were split and treated as described above. After 48 hours, cells were split and all cells were cultivated without OVA257–264, but with or without 10 ng/mL rm-IL18 for 48 hours. Starting from day 2, respective wells were treated with anti-IL2 (50 ng/mL; Peprotech, catalog no. 212-12), STAT5 inhibitor (100 μmol/L; Sigma-Aldrich, catalog no. 573108-10MG), AKT VIII (2 μmol/L; Sigma-Aldrich, catalog no. 124018-5MG), or rapamycin (1 nmol/L; Merck, catalog no. 553210-1MG). After restimulation [RPMI, 100 ng/mL OVA257–264, 1 μg/mL Brefeldin A (Sigma-Aldrich, catalog no. B7651); 4 hours, 37°C], Cytofix/Cytoperm Fixation/Permeabilisation Kit (BD, catalog no. 554714) was used to stain intracellular cytokines. For p-STAT5A staining, cells were fixed (2% paraformaldehyde; AlfaAeser, catalog no. J61899) and permeabilized (98% methanol; Roth, catalog no. 4627.5). p-S6 and p-mTOR were stained using Phosflow Lyse/Fix Buffer I (BD, catalog no. 558049) and Phosflow Perm/Wash Buffer I (BD, catalog no. 557885). Live/dead Zombie Aqua dye (BioLegend, catalog no. 423102) was stained in PBS for 15 minutes. Transcription factor and Ki67 staining was performed using eBioscience Foxp3/Transcription Factor Staining Buffer Set (Thermo Fisher Scientific, catalog no. 00-5523-00). Detailed information about inhibitors and antibodies is provided in Supplementary Tables S1 and S2.

Tumor model and adoptive T-cell transfer

A total of 1 × 106 Panc02 or PancOVA tumor cells were injected subcutaneously into the flank of 8 to 12 weeks old CD45.1 mice. After 7 days, tumor-bearing mice were treated intraperitoneally, with 1 × 106 CTLs. Tumor growth was measured every second day, and tumor volume was calculated (V = length × width2 × 0.5; mm3). A total of 2 × 104 KPC tumor cells were orthotopically implanted into 8 to 12 weeks old CD45.1 mice as described previously (25). Thirteen days after surgery, tumor-bearing mice were treated intraperitoneally with 1 × 106 CTLs. Tumors were monitored weekly using ultrasound (Vevo 2100 Imaging System). Mice were sacrificed at day 18.

Isolation and staining of murine tumor-infiltrating lymphocytes ex vivo

In the subcutaneous tumor model, 3 days after adoptive T-cell transfer, tumors were isolated and digested in Hank's Balanced Salt Solution (Thermo Fisher Scientific, catalog no. 14025092) supplemented with 3 mg/mL collagenase D (Roche, catalog no. 11088882001), 100 U/mL DnaseI (Roche, catalog no. 4536282001) 1 hour at 37°C. For antigen-specific intracellular cytokine stimulation, OVA257–264 loaded splenocytes from CD45. One mice were incubated with tumor lysates 1:1 for 4 hours at 37°C in RPMI with 1 μg/mL Brefeldin A. For polyclonal intracellular cytokine stimulation, cells were restimulated with 50 ng/mL phorbol 12-myristate 13-acetate (Sigma-Aldrich, catalog no. P8139), 1 μg/mL Ionomycin (Sigma-Aldrich, catalog no. I0634), and 1 μg/mL Brefeldin A (Sigma-Aldrich, catalog no. B7651) for 4 hours. Surface, intracellular cytokine, and transcription factor staining was performed as described above.

RNA sequencing analysis and bioinformatics

CD45.2+CD8+ and CD8+ cells were sorted on a FACS Aria III. RNA was purified using RNeasy Plus Micro Kit (Qiagen). For RNA sequencing (RNA-seq) of transferred CTLs, 1 ng total RNA was used for cDNA synthesis and amplification with Takarabio SmartSeqv4 kit. cDNA (1 ng) was used for library preparation using Nextera XT DNA Sample Prep Kit (Illumina). Barcoded RNA-seq libraries were sequenced (150 PE) at Novogene on Illumina NovaSeq. Paired-end reads were mapped to the mouse transcriptome (Ensembl 102) using salmon. Resulting transcript quantifications were imported into R and summarized to gene level using “tximeta.” Differential gene expression analysis was performed using DESeq2 (RRID:SCR_000154).

For endogenous CD8+ T cells, barcoded mRNA-seq cDNA libraries were prepared from 50 ng total RNA using NEBNext Poly(A) mRNA Magnetic Isolation Module and NEBNext Ultra II RNA Library Prep Kit for Illumina. Barcoded RNA-seq libraries were onboard clustered using HiSeq Rapid SR Cluster Kit v2 using 8 pmol/L and 59 bps were sequenced on Illumina HiSeq2500. Sequence reads were trimmed for adapter sequences and further processed using Qiagen's software CLC Genomics Workbench. Reads were aligned to GRCm38 genome. Differential gene expression analysis was performed using DESeq2.

Principal component analysis (PCA) was performed using R and “pcaExplorer.” Gene expression profiles were plotted as heatmaps [color-coded z-scores for the regularized logarithm (rlog) transformed, corrected expression values with R and “gplots”].

Human RNA-seq The Cancer Genome Atlas data analysis

The Cancer Genome Atlas (TCGA) data were used to compare transcriptomic data from patients with PAAD. Survival gene expression profiling and gene expression correlations with pathologic stages in PAAD were performed using GEPIA2. Differential expression between PAAD and normal controls was determined by matching TCGA normal and GTEx data and log2(TPM+1) for log scale. For survival analysis, quartiles were set as a group cutoff and Kaplan–Meier curves were plotted. Log-rank P value and HR with 95% confidence interval were calculated. Pearson correlation was used to calculate the correlation ratio between NLRP3, IL18, and IL1B expression and six T-cell exhaustion genes (HAVCR2, TIGIT, LAG3, PDCD1, CXCL13). TIMER database provided a systematic analysis of infiltrating abundances of NLRP3 and IL18R1 in six types of immune cells (B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells) and their infiltration-relevant clinical outcome.

Pathway enrichment analysis and gene set enrichment analysis

Pathway enrichment analysis was performed using EnrichR [Molecular Signatures Database (MSigDB): Hallmark 2020]. Top five significantly enriched pathways (−log10Padj) were plotted. Gene set enrichment analysis (GSEA; RRID:SCR_003199) was performed to validate differences in expression patterns between transferred WT and Il18r−/− CD8+ OT-1 CTLs and endogenous WT and Nlrp3−/− CD8+ T cells.

Statistical analysis

GraphPad Prism v10 (RRID:SCR_002798) was used for statistical analysis. Data are presented as bar graphs (mean ± SD) or tumor growth curves (mean ± SEM). Datapoints represent biological replicates. Normal distribution and variance homogeneity were evaluated by Shapiro–Wilk and F-Test/Brown-Forsythe tests. Statistical significance between two groups was evaluated using two-tailed, unpaired t tests with or without Welch correction or Mann–Whitney test. For multiple groups and/or multiple condition comparisons, one-way ANOVA with or without Welch correction, nonparametric Kruskal–Wallis test or two-way ANOVA followed by Tukey honestly significant difference (HSD) was performed. A critical value for significance of P < 0.05 (*) was used throughout the study. Statistical thresholds of P < 0.01 (**), P < 0.001 (***), and P < 0.0001 (****) are indicated by asterisks.

Data availability statement

Data generated during and/or analyzed during the current study are available in the main text or the Supplementary Materials and Methods, or are available from the corresponding author on reasonable request. Sequencing data have been deposited in the Gene Expression Omnibus under accession numbers GSE200248 and GSE200570.

IL18R signaling promotes intratumoral T-cell exhaustion

Because IL18 expression is associated with shorter survival in patients with pancreatic cancer, we investigated whether IL18 influences intratumoral CD8+ T cells. We aimed to determine whether IL18R signaling engages in regulation of effector function of tumor-infiltrating CTLs in a pancreatic carcinoma mouse model. We generated Il18r−/− mice of the OT-1 strain. These mice have CD8+ T cells specific for a peptide epitope of the model antigen ovalbumin (OVA257–264). Murine CTLs were generated in vitro by stimulating OVA-specific CD8+ T cells from OT-1 WT or OT-1 Il18r−/− mice with OVA257–264. CD8+ T cells from both mouse strains differentiated into effector CTLs, confirmed by expression of surface markers (Supplementary Fig. S1A and S1B). WT CD8+ T cells showed activation-dependent IL18R upregulation (Supplementary Fig. S1C). Panc02 is a murine PAAD cell line derived from a 3-methylcholanthrene–induced tumor in a C57Bl/6 mouse. Panc02 cells were used untransfected or stably transfected with OVA (PancOVA; ref. 24). CTLs (CD45.2+) were adoptively transferred intraperitoneally into PancOVA tumor-bearing congenic CD45.1+ mice, 7 days after subcutaneous injection of PancOVA cells (Fig. 1A). Tumor growth analysis revealed that Panc02 tumors lacking cognate antigen expression were not rejected by WT OT-1 CTLs. In contrast, WT OT-1 CTLs rejected PancOVA tumors in an antigen-specific manner. Treatment with Il18r−/− OT-1 CTLs caused significantly faster tumor rejection (Fig. 1B). Ex vivo analysis of transferred CTLs was performed 3 days after adoptive transfer when no difference in tumor rejection by WT and Il18r−/− OT-1 CTLs was detectable (Supplementary Fig. S1D). WT OT-1 CTLs showed decreased infiltration into tumor tissue, tumor-draining lymph nodes, and peripheral blood (Fig. 1C). In tumor tissue, WT OT-1 CTLs entered an exhausted state, as evidenced by upregulation of coinhibitory receptors PD-1, TIM-3, 2B4, and LAG-3 (Fig. 1D; Supplementary Fig. S1E). In contrast, deficiency in IL18R signaling restricted the expression of coinhibitory receptors.

Figure 1.

T-cell intrinsic IL18R signaling induces exhaustion and promotes tumor growth. A, Schematic of experimental design. Congenic CD45.1 mice were inoculated with 1 × 106 PancOVA cells subcutaneously. After 7 days, tumor-bearing mice were injected with 1 × 106 WT or Il18r−/− OT-1 CTLs (CD45.2) intraperitoneally. After 72 hours, T-cell analysis was performed. B, Tumor growth curve of subcutaneous tumors is shown (tumor growth normalized to tumor size at day 7 when CTL injection was performed; Panc02+WT n = 3 mice, PancOVA+WT n = 6 mice, and PancOVA+ Il18r−/− n = 6 mice). C, Infiltration rate of CD45.2+CD8+ T cells in blood, tumor and tumor-draining lymph node of tumor-bearing mice (n = 3 mice). D, Expression of coinhibitory receptors PD-1, TIM-3, 2B4, and LAG-3 (n = 7–13 mice). E, Simultaneous expression of multiple coinhibitory receptors as relative proportion (pie chart) and total expression (bar graph; mean of n = 6 mice). F, Intratumoral CTLs were stimulated with CD45.1 OVA257–264-loaded splenocytes for 4 hours with 1 μg/mL Brefeldin A, followed by flow cytometry analysis of IFNγ/TNF production (n = 7 mice). G, Expression of transcription factors Eomes, Tbet, and TOX (n = 5 mice), shown as fold change of MFI. Tumor growth in A shows mean ± SEM. Statistical analysis was performed using one-way ANOVA with Welch correction. All other graphs show mean ± SD and statistics were evaluated by two-tailed, unpaired t tests. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Figure 1.

T-cell intrinsic IL18R signaling induces exhaustion and promotes tumor growth. A, Schematic of experimental design. Congenic CD45.1 mice were inoculated with 1 × 106 PancOVA cells subcutaneously. After 7 days, tumor-bearing mice were injected with 1 × 106 WT or Il18r−/− OT-1 CTLs (CD45.2) intraperitoneally. After 72 hours, T-cell analysis was performed. B, Tumor growth curve of subcutaneous tumors is shown (tumor growth normalized to tumor size at day 7 when CTL injection was performed; Panc02+WT n = 3 mice, PancOVA+WT n = 6 mice, and PancOVA+ Il18r−/− n = 6 mice). C, Infiltration rate of CD45.2+CD8+ T cells in blood, tumor and tumor-draining lymph node of tumor-bearing mice (n = 3 mice). D, Expression of coinhibitory receptors PD-1, TIM-3, 2B4, and LAG-3 (n = 7–13 mice). E, Simultaneous expression of multiple coinhibitory receptors as relative proportion (pie chart) and total expression (bar graph; mean of n = 6 mice). F, Intratumoral CTLs were stimulated with CD45.1 OVA257–264-loaded splenocytes for 4 hours with 1 μg/mL Brefeldin A, followed by flow cytometry analysis of IFNγ/TNF production (n = 7 mice). G, Expression of transcription factors Eomes, Tbet, and TOX (n = 5 mice), shown as fold change of MFI. Tumor growth in A shows mean ± SEM. Statistical analysis was performed using one-way ANOVA with Welch correction. All other graphs show mean ± SD and statistics were evaluated by two-tailed, unpaired t tests. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001.

Close modal

In addition, WT OT-1 CTLs infiltrating PancOVA tumors were more likely to express multiple coinhibitory receptors, correlating with a state of exhaustion, as reported previously (ref. 26; Fig. 1E). Consistent with these data, Il18r−/− OT-1 CTLs were able to produce more effector cytokines IFNγ/TNF (Fig. 1F). Concordantly, levels of Eomes and TOX, both transcription factors involved in the induction of exhaustion (27), were increased in WT tumor-infiltrating OT-1 CTLs, whereas Tbet expression, important for CD8+ T-cell effector function, was decreased (Fig. 1G).

Having shown that IL18R promoted T-cell exhaustion in a subcutaneous PancOVA model, we implemented an orthotopic KPC model to determine whether this effect also applies to polyclonal T cells in the specific microenvironment of pancreatic tissue. KPC tumor cells were implanted orthotopically into CD45.1+ mice, as described before (25). Polyclonal CTLs (CD45.2+) from WT C57BL/6 or Il18r−/− mice were adoptively transferred intraperitoneally into tumor-bearing mice 13 days postsurgery. Ex vivo analysis was performed on day 18, when no difference in tumor rejection by WT and Il18r−/− CTLs was detectable (Supplementary Fig. S1F and S1G). Consistent with the findings in the subcutaneous PancOVA model, intratumoral Il18r−/− CTLs in the KPC orthotopic model expressed significantly lower levels of coinhibitory receptors compared with WT CTLs (Supplementary Fig. S1H) and displayed lower TOX levels (Supplementary Fig. S1I).

Similarly, signaling through the IL1 receptor (IL1R), belonging to the same family as IL18R, had an impact on tumor growth (Supplementary Fig. S2A and S2B). Whereas CTL infiltration into the tumor, tumor-draining lymph nodes and blood (Supplementary Fig. S2C) and IFNγ production (Supplementary Fig. S2D) was not significantly different, expression of coinhibitory receptors (Supplementary Fig. S2E) was decreased in IL1R-deficient CTLs, indicating an involvement of IL1R in the induction of T-cell exhaustion.

Together, less effective tumor rejection accompanied by decreased T-cell tumor infiltration, increased T-cell expression of coinhibitory receptors, loss of effector function, and changes in the expression of transcription factors regulating exhaustion suggest that IL18R signaling in intratumoral CTLs promotes exhaustion in mice with PAADs.

IL18R signaling controls the transcriptional profile associated with T-cell exhaustion

To understand the mechanisms underlying IL18R-promoted exhaustion, we performed RNA-seq of sorted WT or Il18r−/− OT-1 CTLs isolated from PancOVA tumors after adoptive transfer (Fig. 2A). PCA highlighted a distinct transcriptional program for WT and Il18r−/− CTLs (Fig. 2B). Comparing the WT and Il18r−/− CTL transcriptomes defined 1,448 differentially expressed genes, 717 upregulated and 731 downregulated in WT compared with Il18r−/− (log2FC 0.75, Padj < 0.01; Fig. 2C and D). Genes associated with T-cell exhaustion (Havcr2, Pdcd1, Tigit, Il2ra, Irf4) were upregulated in WT CTLs. Genes associated with T-cell plasticity (28), like Tcf7, Mapk8, and Lef1 were upregulated in Il18r−/− CTLs.

Figure 2.

IL18R signaling alters transcriptional profile of intratumoral CTLs. A, Schematic of experimental design. Congenic CD45.1 mice were inoculated with 1 × 106 PancOVA cells subcutaneously. After 7 days, tumor-bearing mice were injected with 1 × 106 CD45.2+ WT or Il18r−/− OT-1 CTLs (CD45.2) intraperitoneally, which were sorted 3 days later. RNA-seq was performed. B, PCA of CTLs obtained from transferred WT and Il18r−/− CD8+ CTLs from PancOVA tumors (n = 4 mice). C, Heat map of differentially expressed genes (z-score normalized, logFC > 0.75, Padj ≤ 0.01) by intratumoral transferred WT or Il18r−/− CD8+ CTLs (n = 4, biological replicates). D, Volcano plot depicting differentially expressed genes in WT and Il18r−/− CD8+ CTLs. Red dots represent genes expressed at higher levels in Il18r−/− CD8+ CTLs, whereas gray dots represent genes with higher expression levels in WT CD8+ CTLs. E, GSEA comparing the relative expression of genes in WT and Il18r−/− CD8+ CTLs from PancOVA tumors examining the distribution of genes involved in T-cell exhaustion according to Bengsch and colleagues (11) and to GSE9650. F, Pathway enrichment analysis for MSigDB (MSigDB: Hallmark 2020). Bar graph displays the five most significantly enriched pathways (upregulated in WT = gray; upregulated in Il18r−/− = red) by –log10 value (Padj) for differentially expressed genes with Padj ≤ 0.01 and logFC > 0.75. FWER, family-wise error rate.

Figure 2.

IL18R signaling alters transcriptional profile of intratumoral CTLs. A, Schematic of experimental design. Congenic CD45.1 mice were inoculated with 1 × 106 PancOVA cells subcutaneously. After 7 days, tumor-bearing mice were injected with 1 × 106 CD45.2+ WT or Il18r−/− OT-1 CTLs (CD45.2) intraperitoneally, which were sorted 3 days later. RNA-seq was performed. B, PCA of CTLs obtained from transferred WT and Il18r−/− CD8+ CTLs from PancOVA tumors (n = 4 mice). C, Heat map of differentially expressed genes (z-score normalized, logFC > 0.75, Padj ≤ 0.01) by intratumoral transferred WT or Il18r−/− CD8+ CTLs (n = 4, biological replicates). D, Volcano plot depicting differentially expressed genes in WT and Il18r−/− CD8+ CTLs. Red dots represent genes expressed at higher levels in Il18r−/− CD8+ CTLs, whereas gray dots represent genes with higher expression levels in WT CD8+ CTLs. E, GSEA comparing the relative expression of genes in WT and Il18r−/− CD8+ CTLs from PancOVA tumors examining the distribution of genes involved in T-cell exhaustion according to Bengsch and colleagues (11) and to GSE9650. F, Pathway enrichment analysis for MSigDB (MSigDB: Hallmark 2020). Bar graph displays the five most significantly enriched pathways (upregulated in WT = gray; upregulated in Il18r−/− = red) by –log10 value (Padj) for differentially expressed genes with Padj ≤ 0.01 and logFC > 0.75. FWER, family-wise error rate.

Close modal

GSEA using published gene signatures (11), GSE650 and GSE41867, revealed that the WT CTL transcriptome was enriched for genes expressed by exhausted CD8+ T cells (Fig. 2E; Supplementary Fig. S3A–S3D). To identify changes in their biological pathways, enrichment pathway analysis for hallmark genes (MSigDB: Hallmark 2020) was performed (Fig. 2F). Genes upregulated in WT CTLs were involved in IL2/STAT5 signaling (Iftim, Nrp1, Ctla4, Cd44, Il2ra), myc targets, mTORC1 signaling (Slc7a5, Map2k3), unfolded protein response, and TNFα signaling via NFκB (Stat5a, Relb, Nfkb2). Among downregulated genes, those involved in oxidative phosphorylation were prominent (Cox6b1, Gpx4, Ndufb8).

Thus, IL18R signaling sufficiency imprints intratumoral CTLs cells with an exhaustion-specific transcriptional profile characterized by changes in the expression of signaling molecules and metabolic genes.

IL18 induces T-cell exhaustion by activating the IL2/STAT5 pathway

To evaluate the signaling pathways that might contribute to IL18R signaling–mediated exhaustion, GSEA was performed. It revealed significant enrichment of genes associated with the IL2/STAT5 pathway in WT OT-1 CTLs (Fig. 3A; Supplementary Fig. S4A; MSigDB: Hallmark 2020, dataset from Fig. 2). Therefore, we further explored the involvement of the IL2/STAT5 pathway using a previously published in vitro T-cell exhaustion model (29). CD8+ T cells from OT-1 mice were purified and activated for 2 days by anti-CD3. Thereafter, exhaustion was induced by repetitive stimulation with OVA257–264 in the presence or absence of IL18 (Fig. 3B). Repetitive stimulation resulted in upregulation of PD-1/TIM-3. Likewise, production of IFNγ/TNF was downregulated, mostly in PD-1+TIM-3+ double-positive cells, compared with activated CD8+ OT-1 T-cells lacking repetitive OVA257–264 treatment (Supplementary Fig. S4B–S4D).

Figure 3.

IL18 controls T-cell exhaustion in vitro via the IL2/STAT5 pathway. A, GSEA comparing the relative expression of genes in WT and Il18r−/− CD8+ CTLs examining the distribution of genes involved in IL2/STAT5 signaling based on MSigDBv7.5.1 (hallmark dataset, dataset from Fig. 2). B, Schematic of experimental design of the in vitro exhaustion model. CD8+ T cells were isolated from spleen and lymph nodes of OT-1 mice and differentiated with 1 μg/mL anti-CD3, 2 μg/mL anti-CD28 and 20 ng/mL IL2 ± 10 ng/mL IL18 for 2 days (= activation). After activation, T-cell exhaustion was induced for 6 days by repetitive addition of 10 ng/mL OVA257–264 (= exhaustion). C, Expression of CD25 6 days after induction of exhaustion under IL18 treatment conditions (n = 6). Expression of coinhibitory receptors PD-1 and TIM-3 (D) and production of IFNγ and TNF after 4 hours restimulation with OVA257–264 in the presence of 1 μg/mL Brefeldin A (E) under IL18 (10 ng/mL) and anti-IL2 (50 ng/mL) treatment 6 days after induction of exhaustion is shown (n = 8–19). F, Expression of p-STAT5 6 days after induction of exhaustion under IL18 treatment conditions (n = 6). Expression of PD-1 and TIM-3 (G) and production of IFNγ and TNF (H) under IL18 (10 ng/mL) and STAT5 inhibitor (100 μmol/L) treatment 6 days after induction of exhaustion (n = 7–19). Fold change was normalized to treatment conditions without repetitive OVA257–264 stimulation and IL18. All graphs show mean ± SD. Statistics were evaluated by two-tailed, unpaired t test with Welch correction (C and F), by two-way ANOVA followed by a Tukey HSD (D, E, G, and H). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant. FWER, family-wise error rate.

Figure 3.

IL18 controls T-cell exhaustion in vitro via the IL2/STAT5 pathway. A, GSEA comparing the relative expression of genes in WT and Il18r−/− CD8+ CTLs examining the distribution of genes involved in IL2/STAT5 signaling based on MSigDBv7.5.1 (hallmark dataset, dataset from Fig. 2). B, Schematic of experimental design of the in vitro exhaustion model. CD8+ T cells were isolated from spleen and lymph nodes of OT-1 mice and differentiated with 1 μg/mL anti-CD3, 2 μg/mL anti-CD28 and 20 ng/mL IL2 ± 10 ng/mL IL18 for 2 days (= activation). After activation, T-cell exhaustion was induced for 6 days by repetitive addition of 10 ng/mL OVA257–264 (= exhaustion). C, Expression of CD25 6 days after induction of exhaustion under IL18 treatment conditions (n = 6). Expression of coinhibitory receptors PD-1 and TIM-3 (D) and production of IFNγ and TNF after 4 hours restimulation with OVA257–264 in the presence of 1 μg/mL Brefeldin A (E) under IL18 (10 ng/mL) and anti-IL2 (50 ng/mL) treatment 6 days after induction of exhaustion is shown (n = 8–19). F, Expression of p-STAT5 6 days after induction of exhaustion under IL18 treatment conditions (n = 6). Expression of PD-1 and TIM-3 (G) and production of IFNγ and TNF (H) under IL18 (10 ng/mL) and STAT5 inhibitor (100 μmol/L) treatment 6 days after induction of exhaustion (n = 7–19). Fold change was normalized to treatment conditions without repetitive OVA257–264 stimulation and IL18. All graphs show mean ± SD. Statistics were evaluated by two-tailed, unpaired t test with Welch correction (C and F), by two-way ANOVA followed by a Tukey HSD (D, E, G, and H). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant. FWER, family-wise error rate.

Close modal

In the ex vivo experiments, in vitro IL18 supplementation increased the PD-1/TIM-3 expression (Fig. 3D) and reduced cell number and cell-cycle progression (Supplementary Fig. S4E). In addition, IL18 decreased the production of IFNγ/TNF but not IL2 (Fig. 3E; Supplementary Fig. S4F). On the basis of CD62 L and CD44 expression, no changes in T-cell memory formation were detectable, further indicating a specific IL18 effect on the induction of exhaustion (Supplementary Fig. S4G).

In this system, we investigated the influence of IL18 on IL2 signaling during exhaustion. Consistent with the enriched IL2/STAT5 signaling in IL18R-sufficient CTLs, CD25 expression was significantly upregulated in exhausted CD8+ T cells supplemented with IL18 (Fig. 3C). Anti-IL2 reduced PD-1/TIM-3 expression and blocked IL18-induced upregulation of PD-1/TIM-3 (Fig. 3D). Anti-IL2 also partially blocked IFNγ/TNF production and inhibited the IL18-induced decrease of IFNγ/TNF production in a setting of T-cell exhaustion (Fig. 3E). Similar effects were detectable when experiments were performed without IL2 supplementation during the repetitive stimulation phase (Supplementary Fig. S5B–S5D).

As a downstream target of IL2, we investigated the role of STAT5 by measuring the expression of p-STAT5A (Fig. 3F). Inhibition of STAT5 did not reverse IL18-mediated upregulation of PD-1/TIM-3 (Fig. 3G), but it did restore the loss of effector function by increasing IFNγ/TNF production (Fig. 3H) and it abolished the IL18-mediated reduction in cell number and cell-cycle progression (Supplementary Fig. S5E). In summary, these results confirmed that IL18 promotes T-cell exhaustion in vitro as characterized by upregulation of PD-1/TIM-3 and decreased production of effector cytokines. This resulted in the activation of multiple pathways, partially involving IL2/STAT5.

IL18 induces T-cell exhaustion by activating the AKT/mTOR pathway

Because anti-IL2 and STAT5 inhibition only partially blocked IL18-induced T-cell exhaustion in the in vitro model, we hypothesized that further pathways are involved. IL2 signaling can also activate the Akt/mTOR pathway (30). Consistent with this hypothesis, GSEA revealed significant enrichment of genes associated with the mTORC1 pathway in intratumoral WT compared with Il18r−/− OT-1CTLs (MSigDB: Hallmark 2020, dataset from Fig. 2; Fig. 4A; Supplementary Fig. S5F). These data, together with the increased expression of p-mTOR (Fig. 4B) and its downstream target p-S6 after IL18 treatment (Fig. 4C) in the in vitro T-cell exhaustion model indicates involvement of mTOR signaling in IL18R-mediated T-cell exhaustion. To gain mechanistic insight, we blocked the AKT/mTOR signaling pathway using AKT inhibitor VIII (31) and rapamycin in an in vitro setting of IL18-induced T-cell exhaustion. AKT inhibition abrogated IL18-induced expression of PD-1/TIM-3 (Fig. 4D), reversed the decrease in IFNγ/TNF production found in IL18-treated exhausted T cells (Fig. 4E) and diminished the IL18-mediated decrease in cell number and cell-cycle progression (Supplementary Fig. S5G), indicating an AKT-mediated mechanism of IL18-induced T-cell exhaustion.

Figure 4.

IL18 controls T-cell exhaustion in vitro via AKT/mTOR pathway. A, GSEA comparing the relative expression of genes in WT and Il18r−/− CD8+ CTLs examining the distribution of genes involved in mTORC1 signaling based on MSigDBv7.5.1 (hallmark dataset, dataset from Fig. 2). Expression of p-mTOR (B) and p-S6 (C) 6 days after induction of exhaustion under IL18 treatment conditions (n = 9). Expression of PD-1 and TIM-3 (D) and production of IFNγ and TNF (E) under IL18 (10 ng/mL) and AKT inhibitor VIII (2 μmol/L) treatment 6 days after induction of exhaustion (n = 5–19). Expression of PD-1 and TIM-3 (F) and production of IFNγ and TNF (G) in the presence of IL18 (10 ng/mL) and rapamycin (100 nmol/L) 6 days after induction of exhaustion (n = 7–19). H, Schematic influence of IL18R signaling on CD25/STAT5 and AKT/mTOR pathway leading to T-cell exhaustion. Fold change was normalized to treatment conditions without repetitive OVA257–264 stimulation and IL18. All graphs show mean ± SD. Statistics were evaluated by two-tailed, unpaired t tests with (C) or without Welch correction (B) or by two-way ANOVA followed by a Tukey HSD (D–G). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant. FWER, family-wise error rate.

Figure 4.

IL18 controls T-cell exhaustion in vitro via AKT/mTOR pathway. A, GSEA comparing the relative expression of genes in WT and Il18r−/− CD8+ CTLs examining the distribution of genes involved in mTORC1 signaling based on MSigDBv7.5.1 (hallmark dataset, dataset from Fig. 2). Expression of p-mTOR (B) and p-S6 (C) 6 days after induction of exhaustion under IL18 treatment conditions (n = 9). Expression of PD-1 and TIM-3 (D) and production of IFNγ and TNF (E) under IL18 (10 ng/mL) and AKT inhibitor VIII (2 μmol/L) treatment 6 days after induction of exhaustion (n = 5–19). Expression of PD-1 and TIM-3 (F) and production of IFNγ and TNF (G) in the presence of IL18 (10 ng/mL) and rapamycin (100 nmol/L) 6 days after induction of exhaustion (n = 7–19). H, Schematic influence of IL18R signaling on CD25/STAT5 and AKT/mTOR pathway leading to T-cell exhaustion. Fold change was normalized to treatment conditions without repetitive OVA257–264 stimulation and IL18. All graphs show mean ± SD. Statistics were evaluated by two-tailed, unpaired t tests with (C) or without Welch correction (B) or by two-way ANOVA followed by a Tukey HSD (D–G). *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001; ns, not significant. FWER, family-wise error rate.

Close modal

Rapamycin-induced mTORC1 inhibition demonstrated a pattern similar to that observed using AKT inhibitor VIII, which was characterized by significant downregulation of PD-1/TIM-3 (Fig. 4F). Rapamycin was able to restore IFNγ/TNF in the setting of IL18-induced effector cytokine suppression (Fig. 4G) and inhibited the IL18-mediated decrease in cell number and cell-cycle progression (Supplementary Fig. S5H). In addition, it downregulated IL18-induced phosphorylation of mTOR and S6 (Supplementary Fig. S5I and S5J).

In summary, inhibition of IL2 and its downstream targets mTOR, and to a lesser degree AKT and STAT5, reversed exhaustion-promoting effects of IL18, underlining involvement of the IL2/STAT5 and AKT/mTOR pathways (Fig. 4H).

IL18 correlates with poor survival and tumor progression in patients with pancreatic carcinoma

To determine the relevance of IL18 signaling in patients with pancreatic carcinoma, we analyzed RNA-seq data from patients with PAAD from TCGA database. The expression of IL18, IL1B, and NLRP3, was significantly increased in PAAD tumor tissue compared with normal tissue (Fig. 5A). Of note, only expression of IL18, but not of IL1B and NLRP3, correlated with the pathologic staging of PAAD tumors (Fig. 5B). A similar correlation was observed between gene expression and patient survival (Fig. 5C). Although a different expression ratio in IL1B and NLRP3 did not alter survival for patients with PAAD, high expression of IL18 was associated with poor prognosis on overall survival, indicating preeminent significance of IL18 compared with NLRP3 and IL1B. Furthermore, correlation analysis of IL18R1, IL1R1, and NLRP3 with T-cell exhaustion signature genes (HAVCR2, TIGIT, LAG3, PDCD1, CXCL13, LAYN; ref. 32) showed a significant correlation in patients with PAAD (Supplementary Fig. S6A). Correlation analysis of NLRP3 expression with different infiltrating immune cell populations showed that a positive correlation between NLRP3 expression and tumor-infiltrating subpopulations was highest among antigen-presenting cells, such as macrophages, dendritic cells, and neutrophils (Fig. 5D).

Figure 5.

NLRP3-mediated IL18 signaling associates with shorter survival in patients with PAAD. TCGA database RNA-seq analysis of patients with PAAD. A, Box plots showing the expression of IL18, IL1B, and NLRP3 in cancer tissues of patients with PAAD was analyzed by GEPIA2 [tumor tissue (purple) n = 179, healthy tissue (gray) n = 171]. Transcriptional levels were log-normalized by the log2(TPM+1) method. t test was used to compare differences in expression between tumor and normal tissues. *, P < 0.05. B, Association of mRNA expression of IL18, IL1B, and NLRP3 with pathologic tumor stages in patients with PAAD. Violin plots were created using GEPIA2 platform. F-value indicates statistical value of the F-test. Pr(> F) indicates P value. C, Overall survival analyses in patients with PAAD related to IL18, IL1B, and NLRP3 mRNA expression (n = 44–45). Patient survival was compared between two groups divided at top (red) and bottom (blue) quartile value of the expressed gene. Kaplan–Meier plots were created using GEPIA2 platform. D, Correlation between NLRP3 expression and abundance of immune cells (B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells) in PAAD. The study was performed in the TIMER database. Spearman correlation was used to evaluate the correlation between NLRP3 expression and abundance of immune cells.

Figure 5.

NLRP3-mediated IL18 signaling associates with shorter survival in patients with PAAD. TCGA database RNA-seq analysis of patients with PAAD. A, Box plots showing the expression of IL18, IL1B, and NLRP3 in cancer tissues of patients with PAAD was analyzed by GEPIA2 [tumor tissue (purple) n = 179, healthy tissue (gray) n = 171]. Transcriptional levels were log-normalized by the log2(TPM+1) method. t test was used to compare differences in expression between tumor and normal tissues. *, P < 0.05. B, Association of mRNA expression of IL18, IL1B, and NLRP3 with pathologic tumor stages in patients with PAAD. Violin plots were created using GEPIA2 platform. F-value indicates statistical value of the F-test. Pr(> F) indicates P value. C, Overall survival analyses in patients with PAAD related to IL18, IL1B, and NLRP3 mRNA expression (n = 44–45). Patient survival was compared between two groups divided at top (red) and bottom (blue) quartile value of the expressed gene. Kaplan–Meier plots were created using GEPIA2 platform. D, Correlation between NLRP3 expression and abundance of immune cells (B cells, CD8+ T cells, CD4+ T cells, macrophages, neutrophils, and dendritic cells) in PAAD. The study was performed in the TIMER database. Spearman correlation was used to evaluate the correlation between NLRP3 expression and abundance of immune cells.

Close modal

Taken together, expression of the genes encoding the NLRP3 inflammasome and its activation products, IL1β and IL18, were found to be increased in tissues of patients with PAAD, but only IL18 expression had negative effects on tumor staging and survival, suggesting that other proteins might be involved in IL18 processing. Because NLRP3 expression correlates with the expression of T-cell exhaustion genes and is mainly expressed by antigen-presenting tumor infiltrates, we hypothesized that IL18 signaling inducing T-cell exhaustion in patients with PAAD is probably NLRP3 mediated.

Similarities in transcriptional profiles between intratumoral CD8+ T cells from Il18r−/− and Nlrp3−/− mice

To investigate the contribution of NLRP3 signaling to T-cell exhaustion, we injected subcutaneously PancOVA tumor cells into WT and Nlrp3−/− mice. After 10 days, tumor weight was not significantly different (Supplementary Fig. S6B). Tumor-infiltrating CD8+ T cells were isolated and analyzed by flow cytometry (Fig. 6A). Like for IL18R-sufficient CTLs, CD8+ T cells obtained from mice with intact NLRP3 signaling in the tumor microenvironment showed increased expression of PD-1, TIM-3, 2B4, and LAG-3 compared with CD8+ T cells from PancOVA-bearing Nlrp3−/− mice (Fig. 6B). However, IFNγ/TNF production by CD8+ T cells from PancOVA-bearing WT versus Nlrp3−/− mice, although showing a tendency, was not significantly different (Supplementary Fig. S6C). To gain more insight into the transcriptional changes caused by NLPR3-mediated signaling, we performed RNA-seq of endogenous CD8+ T cells derived from NLRP3-deficient versus WT mice. PCA revealed a distinct transcriptional program for WT and Nlrp3−/− CD8+ T cells (Fig. 6C). Comparison of WT and Nlrp3−/− CD8+ T-cell transcriptomes identified 978 differentially expressed genes, 535 upregulated and 443 downregulated in WT compared with Nlrp3−/− (log2FC = 0.75; Fig. 6D). Like in the RNA-seq results from Il18r−/− CTLs, genes associated with T-cell exhaustion, (Havcr2, Pdcd1, CD244, Ctla4, Il2ra) were upregulated in WT CD8+ T cells. Genes associated with T-cell plasticity, such as Tcf7 and Lef1, were upregulated in Nlrp3−/− CD8+ T cells.

Figure 6.

NLRP3 deficiency prevents intratumoral T-cell exhaustion. A, Schematic of experimental design. WT or Nlrp3−/− mice were injected with 1 × 106 PancOVA cells subcutaneously. On day 10 after tumor inoculation, endogenous CD8+ T cells were isolated from the tumor and were analyzed ex vivo. B, Expression of the coinhibitory receptors PD-1, TIM-3, 2B4, and LAG-3 (n = 4–7). Graph shows mean ± SD and statistics were evaluated by two-tailed, unpaired t tests. *, P < 0.05; ****, P < 0.0001; ns, not significant. C, PCA of CTLs obtained from endogenous WT and Nlrp3−/− CD8+ T cells from PancOVA tumors (n = 4). D, Heat map of differentially expressed genes (z-score normalized, logFC > 0.75) between intratumoral WT or Nlrp3−/− CD8+ T cells (n = 4, biological replicates). E, Venn diagram summarizing the overlap between differentially expressed genes in Il18r−/− (dataset from Fig. 2) and Nlrp3−/− CD8+ T cells based on RNA-seq (logFC > 0.75). F, Scatter plot overlapping gene regulation in Il18r−/− (dataset from Fig. 2) and Nlrp3−/− CD8+ T cells. G, GSEA comparing the relative expression of genes in endogenous WT and Nlrp3−/− CD8+ T cells from PancOVA tumors examining the distribution of genes involved in T-cell exhaustion according to Bengsch and colleagues (11). H, GSEA comparing relative expression of genes in WT and Nlrp3−/− CD8+ T cells examining distribution of genes involved in IL2–STAT5 signaling based on MSigDBv7.5.1 (hallmark dataset). I, Summary of the proposed mechanism of IL8R signaling resulting in intratumoral T-cell exhaustion. NLRP3-mediated IL18 release from intratumoral antigen-presenting cells enhances T-cell exhaustion. Induction of exhaustion via signaling through T-cell IL18R can be enhanced by IL2, resulting in enhanced activation of the STAT5 and AKT/mTOR pathway. This in turn results in transcriptional changes of genes encoding coinhibitory receptors and molecules involved in T-cell metabolism and self-renewal, resulting in an exhausted T-cell phenotype. FWER, family-wise error rate.

Figure 6.

NLRP3 deficiency prevents intratumoral T-cell exhaustion. A, Schematic of experimental design. WT or Nlrp3−/− mice were injected with 1 × 106 PancOVA cells subcutaneously. On day 10 after tumor inoculation, endogenous CD8+ T cells were isolated from the tumor and were analyzed ex vivo. B, Expression of the coinhibitory receptors PD-1, TIM-3, 2B4, and LAG-3 (n = 4–7). Graph shows mean ± SD and statistics were evaluated by two-tailed, unpaired t tests. *, P < 0.05; ****, P < 0.0001; ns, not significant. C, PCA of CTLs obtained from endogenous WT and Nlrp3−/− CD8+ T cells from PancOVA tumors (n = 4). D, Heat map of differentially expressed genes (z-score normalized, logFC > 0.75) between intratumoral WT or Nlrp3−/− CD8+ T cells (n = 4, biological replicates). E, Venn diagram summarizing the overlap between differentially expressed genes in Il18r−/− (dataset from Fig. 2) and Nlrp3−/− CD8+ T cells based on RNA-seq (logFC > 0.75). F, Scatter plot overlapping gene regulation in Il18r−/− (dataset from Fig. 2) and Nlrp3−/− CD8+ T cells. G, GSEA comparing the relative expression of genes in endogenous WT and Nlrp3−/− CD8+ T cells from PancOVA tumors examining the distribution of genes involved in T-cell exhaustion according to Bengsch and colleagues (11). H, GSEA comparing relative expression of genes in WT and Nlrp3−/− CD8+ T cells examining distribution of genes involved in IL2–STAT5 signaling based on MSigDBv7.5.1 (hallmark dataset). I, Summary of the proposed mechanism of IL8R signaling resulting in intratumoral T-cell exhaustion. NLRP3-mediated IL18 release from intratumoral antigen-presenting cells enhances T-cell exhaustion. Induction of exhaustion via signaling through T-cell IL18R can be enhanced by IL2, resulting in enhanced activation of the STAT5 and AKT/mTOR pathway. This in turn results in transcriptional changes of genes encoding coinhibitory receptors and molecules involved in T-cell metabolism and self-renewal, resulting in an exhausted T-cell phenotype. FWER, family-wise error rate.

Close modal

A total of 307 transcripts expressed in Il18r−/− CTLs and Nlrp3−/− CD8+ T cells overlapped (Fig. 6E; Supplementary Table S3). Within the concordantly upregulated genes (Fig. 6F, top right quadrant), we found transcripts associated with developmental stage (Sox4, Drc1), whereas T-cell exhaustion signature genes (Havcr2, Cd44) were downregulated (Fig. 6F, bottom left quadrant). GSEA using published gene signatures (11), revealed that the WT CD8+ transcriptome was enriched in genes expressed by exhausted CD8+ T cells (Fig. 6G; Supplementary Fig. S6D) and in genes associated with the CD25/STAT5 pathway (Fig. 6H; Supplementary Fig. S6E; MSigDB: Hallmark 2020).

Together, these results indicate that cytokines processed by the NLRP3 inflammasome in the tumor microenvironment regulate similar pathways to IL18R-mediated exhaustion in CD8+ T cells. In turn, the increased IL18 abundance in the tumor microenvironment induced T-cell exhaustion in intratumoral CD8+ T cells via the IL2/STAT5/mTOR pathway (Fig. 6I).

Our findings highlight IL18R signaling in intratumoral CD8+ T cells as a possible target for improving current T cell–based immunotherapies in patients with pancreatic cancer. This study uncovers a molecular mechanism contributing to already extensively studied features of intratumoral T-cell exhaustion.

Mechanistically, IL18R signaling may affect CD8+ T-cell plasticity in different ways. Recently, an IL18Rαhigh subset of CD8+ T cells was shown to mediate the immune response after adoptive immunotherapy, indicating a role of IL18 signaling in the cytotoxic strength of intratumoral CD8+ T cells (33). IL18, an IFNγ-inducing cytokine, may initially promote T cell–mediated antitumor immunity (34). However, multiple mechanisms may lead to IL18-promoting tumor progression. First, IL18 might suppress antitumor responses by upregulating PD-L1 in the tumor microenvironment causing a negative feedback loop, resulting in T-cell exhaustion (35).

Second, IL18R signaling, together with IL12, can induce exhaustion by interfering with CD25 expression and IL2 signaling (15). Recently, high IL2 levels were described to promote exhaustion by activating STAT5 in CD8+ T cells (36). As IL18 has been reported to synergize with common γ-chain family cytokines, such as IL2, to drive IL2R expression, CD8+ T cells could become dysfunctional in the tumor microenvironment by IL18-mediated activation of the IL2/CD25/STAT5 pathway. Our data on STAT5 expression and inhibition confirm interplay between IL18R and IL2R signaling, resulting in T-cell exhaustion.

Examining IL18R-induced changes in the molecular characteristics of intratumoral T cells by RNA-seq revealed that IL18R-mediated T-cell exhaustion is also regulated by changes in the AKT/mTOR signaling pathway. We showed that inhibition of IL18-induced AKT signaling reversed T-cell exhaustion. Fully exhausted T cells derive from self-renewing “preexhausted” T cells (37). This corresponds to AKT activity driving CTL differentiation to a “memory-naïve” phenotype. AKT inhibition improves antitumor immunity of exhausted T cells by increasing memory-like characteristics (38). Acquiring transcriptional and metabolic properties of memory-like T cells during continuous T-cell receptor stimulation enables these cells to escape T-cell exhaustion resulting in more effective tumor immune response (39). T-cell pool modulation is clinically relevant, as “preexhausted” CD8+ T cells represent cells that react to checkpoint inhibitor therapy (28). These “preexhausted” T cells are characterized by intermediate PD-1 expression, high TCF-1 and CXCR5 expression and lower levels of NR4a and TOX. In our case, defective IL18R signaling might prevent cytotoxic T cells from reaching a point of no return toward terminal exhaustion. IL18R signaling pushes T cells toward highly effective but short-lived cytotoxic effector cells, ultimately resulting in terminal exhaustion. Absence of IL18R signaling sustains T-cell plasticity and can maintain an effective antitumor immune response.

Early targeting of mTOR signaling improves T-cell immune responses by preserving the activity of “preexhausted” T cells (40). mTOR signaling might affect T-cell effector plasticity and effector function mainly by regulating T-cell metabolism. mTORC1 controls T-cell effector function by converting T-cell metabolism to glycolysis. mTOR inhibition preserves the cellular metabolism of “preexhausted” T cells and sustains long-term effector T-cell immune responses (40). IL18R signaling interferes with and impairs mitochondrial oxidative phosphorylation via activation of mTOR and limits self-renewal and effector function of intratumoral T cells under persistent antigen exposure (41). As oxidative phosphorylation is characteristic for memory T-cell pool formation after successful antigen clearance (42), IL18R-induced impairment of mitochondrial oxidative phosphorylation via activation of mTOR might push T cells further into a state of T-cell exhaustion. Furthermore, mTOR-mediated effects of IL18 may be regulated via synergism between IL18 and IL12. IL18 works synergistically with IL12 to induce IL2R expression (15). mTOR inhibition in IL12-conditioned cytotoxic T cells enhances memory T-cell pool formation and increases their effector function (43). IL18R signaling might exert its effect in a similar manner.

IL18R belongs to the IL1R family (14). As for IL18R signal transduction, IL1R signaling influenced the exhaustion state of intratumoral T cells indicating an involvement of the IL1R family. Although IL18R and IL1R signaling contributed to T-cell exhaustion in our analysis, differences in effect size were detectable. Loss of IL18R signaling decreased T-cell exhaustion to a higher extent. This may be explained by different downstream targets. Whereas, IL18R signaling enhanced IL2/STAT5 and AKT/mTOR pathway, IL1R activation enhances NFκB and MAPK activation (14, 44). Signaling through the T-cell receptor results primarily in activation and nuclear translocation of the transcription factors NFAT, AP-1, and NFκB. Being located directly downstream of T-cell receptor activation IL1R-mediated NFκB might be excluded from synergistic activity toward T-cell exhaustion (45). Instead, activation of STAT5 and AKT/mTOR pathways might exert synergistic effects in a setting of repetitive T-cell receptor triggering as revealed in our in vitro exhaustion model.

Interestingly, deletion of the IL18 processing inflammasome NLRP3 led to reduced expression of coinhibitory receptors on endogenous CD8+ T cells, but not to enhanced cytokine production, although showing a tendency. NLRP3 deletion in the whole tumor microenvironment has an impact on several cell types and does not only influence CD8+ T cell–mediated immune responses. Although NLRP3 promotes expansion of immune-suppressive macrophages in PDAC (46), it is also important for stimulation of NK and Th1 cells (47). Comparing our Nlrp3−/− and Il18r−/− results revealed regulation of similar transcripts and signaling pathways in intratumoral CD8+ T cells. Thus, NLRP3 deficiency partially resembles the phenotype of IL18R deficiency in adoptively transferred T cells, which suggests an involvement of NLRP3 in IL18-mediated exhaustion.

Our findings identified IL18R signaling as an immune checkpoint and hold promise for patients with pancreatic carcinoma. Although systemic IL18 application had no effect on the overall survival of mice with orthotopic pancreatic cancer, probably because of the cytokines modulating function on multiple cell types, focusing on the role of IL18R expression by CD8+ T cells might be a new potential clinical therapy target (48). More recent data found connections between IL18 and checkpoint inhibitor expression indicating that IL18 is associated with bad prognosis in various cancer types. Terme and colleagues reported IL18-induced PD-1–dependent immunosuppression in cancer (49), corroborating the hypothesis that during cancer progression protumorigenic effects of IL18 prevail over initial T-cell stimulatory effects. Blocking IL18R signaling might exert beneficial efficacy as an adjuvant treatment when combined with established immunotherapies. Increasing the efficacy of PD-1 inhibitors using concomitant therapeutic strategies, such as IL18 blockade, might drastically change the clinical algorithm for pancreatic cancer therapy. Furthermore, our data indicate that counter-intuitive strategies such as blocking T cell–activating cytokines might have unforeseen effects when applied in tumor models and possibly in patients with cancer. Beyond IL18R signaling, careful consideration of unwanted effects of targeting mTOR and IL2-mediated pathways on T-cell exhaustion is warranted whenever new chemoimmunotherapies are applied.

P. Duewell is cofounder and employee of Stealth Biotech. C. Bauer reports grants from German Research Foundation (DFG) and Behring-Röntgen-Foundation during the conduct of the study. No disclosures were reported by the other authors.

V. Lutz: Conceptualization, formal analysis, investigation, methodology, writing–original draft. V.M. Hellmund: Formal analysis, investigation, methodology, writing–review and editing. F.S.R. Picard: Investigation, methodology. H. Raifer: Formal analysis, investigation. T. Ruckenbrod: Investigation, methodology. M. Klein: Formal analysis, methodology. T. Bopp: Methodology. R. Savai: Methodology. P. Duewell: Resources. C.U. Keber: Methodology. A. Weigert: Methodology. H.-R. Chung: Formal analysis, investigation, methodology. M. Buchholz: Supervision. A. Menke: Funding acquisition, methodology. T.M. Gress: Supervision, funding acquisition. M. Huber: Conceptualization, supervision, funding acquisition, writing–review and editing. C. Bauer: Conceptualization, funding acquisition, validation.

This work was supported by DFG grants KFO325 BA 3824/3-1, BA 3824/3-2, HU 1824/5-1, HU 1824/5-2, GRK 2573/1 RP2, HU1824/7-1 and the Behring-Röntgen-Foundation (project number 63-0010).

We thank Julia Menne for technical advice on the in vitro exhaustion model. We thank the Institute for Laboratory Medicine and Pathobiochemistry from the Philipps University Marburg for the authorization to use the FACS Canto and the FACS Core Facility Marburg for the authorization to use the CytoFlex-Lx (funded by Deutsche Forschungsgemeinschaft grant 443291381). We thank Prof. Dr. Natalio Garbi and Ann-Christine Henneke for providing IL18R-deficient mice. This work was supported by DFG grants KFO325 BA 3824/3-1, BA 3824/3-2, HU 1824/5-1, HU 1824/5-2, GRK 2573/1 RP2, HU1824/7-1 and the Behring-Röntgen-Foundation (project number 63-0010).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

Note: Supplementary data for this article are available at Cancer Immunology Research Online (http://cancerimmunolres.aacrjournals.org/).

1.
Siegel
RL
,
Miller
KD
,
Fuchs
HE
,
Jemal
A
.
Cancer statistics, 2021
.
CA Cancer J Clin
2021
;
71
:
7
33
.
2.
Saung
MT
,
Zheng
L
.
Current standards of chemotherapy for pancreatic cancer
.
Clin Ther
2017
;
39
:
2125
34
.
3.
Macherla
S
,
Laks
S
,
Naqash
A
,
Bulumulle
A
,
Zervos
E
,
Muzaffar
M
.
Emerging role of immune checkpoint blockade in pancreatic cancer
.
Int J Mol Sci
2018
;
19
:
3505
.
4.
Ino
Y
,
Yamazaki-Itoh
R
,
Shimada
K
,
Iwasaki
M
,
Kosuge
T
,
Kanai
Y
, et al
.
Immune cell infiltration as an indicator of the immune microenvironment of pancreatic cancer
.
Br J Cancer
2013
;
108
:
914
23
.
5.
Zhu
Y
,
Herndon
JM
,
Sojka
DK
,
Kim
K-W
,
Knolhoff
BL
,
Zuo
C
, et al
.
Tissue-resident macrophages in pancreatic ductal adenocarcinoma originate from embryonic hematopoiesis and promote tumor progression
.
Immunity
2017
;
47
:
323
38
.
6.
Trovato
R
,
Fiore
A
,
Sartori
S
,
Canè
S
,
Giugno
R
,
Cascione
L
, et al
.
Immunosuppression by monocytic myeloid-derived suppressor cells in patients with pancreatic ductal carcinoma is orchestrated by STAT3
.
J Immunother Cancer
2019
;
7
:
255
.
7.
Liyanage
UK
,
Moore
TT
,
Joo
H-G
,
Tanaka
Y
,
Herrmann
V
,
Doherty
G
, et al
.
Prevalence of regulatory T cells is increased in peripheral blood and tumor microenvironment of patients with pancreas or breast adenocarcinoma
.
J Immunol
2002
;
169
:
2756
61
.
8.
Knudsen
ES
,
Vail
P
,
Balaji
U
,
Ngo
H
,
Botros
IW
,
Makarov
V
, et al
.
Stratification of pancreatic ductal adenocarcinoma: combinatorial genetic, stromal, and immunologic markers
.
Clin Cancer Res
2017
;
23
:
4429
40
.
9.
Gallimore
A
,
Glithero
A
,
Godkin
A
,
Tissot
AC
,
Plückthun
A
,
Elliott
T
, et al
.
Induction and exhaustion of lymphocytic choriomeningitis virus–specific cytotoxic T lymphocytes visualized using soluble tetrameric major histocompatibility complex class I–peptide complexes
.
J Exp Med
1998
;
187
:
1383
93
.
10.
Wherry
EJ
.
T cell exhaustion
.
Nat Immunol
2011
;
12
:
492
9
.
11.
Bengsch
B
,
Ohtani
T
,
Khan
O
,
Setty
M
,
Manne
S
,
O'Brien
S
, et al
.
Epigenomic-guided mass cytometry profiling reveals disease-specific features of exhausted CD8 T cells
.
Immunity
2018
;
48
:
1029
45
.
12.
Dinarello
CA
,
Novick
D
,
Kim
S
,
Kaplanski
G
.
Interleukin-18 and IL-18 binding protein
.
Front Immunol
2013
;
4
:
289
.
13.
Schroder
K
,
Tschopp
J
.
The inflammasomes
.
Cell
2010
;
140
:
821
32
.
14.
Tsutsumi
N
,
Kimura
T
,
Arita
K
,
Ariyoshi
M
,
Ohnishi
H
,
Yamamoto
T
, et al
.
The structural basis for receptor recognition of human interleukin-18
.
Nat Commun
2014
;
5
:
5340
.
15.
Lee
S-H
,
Fragoso
MF
,
Biron
CA
.
Cutting edge: A novel mechanism bridging innate and adaptive immunity: IL-12 induction of CD25 to form high-affinity IL-2 receptors on NK cells
.
J Immunol
2012
;
189
:
2712
6
.
16.
Kalina
U
,
Kauschat
D
,
Koyama
N
,
Nuernberger
H
,
Ballas
K
,
Koschmieder
S
, et al
.
IL-18 activates STAT3 in the natural killer cell Line 92, augments cytotoxic activity, and mediates IFN-γ production by the stress kinase p38 and by the extracellular regulated kinases p44 erk-1 and p42 erk-2 1
.
J Immunol
2000
;
165
:
1307
13
.
17.
Chandrasekar
B
,
Valente
AJ
,
Freeman
GL
,
Mahimainathan
L
,
Mummidi
S
.
Interleukin-18 induces human cardiac endothelial cell death via a novel signaling pathway involving NF-κB-dependent PTEN activation
.
Biochem Biophys Res Commun
2006
;
339
:
956
63
.
18.
Deason
K
,
Troutman
TD
,
Jain
A
,
Challa
DK
,
Mandraju
R
,
Brewer
T
, et al
.
BCAP links IL-1R to the PI3K–mTOR pathway and regulates pathogenic Th17 cell differentiation
.
J Exp Med
2018
;
215
:
2413
28
.
19.
Almutairi
SM
,
Ali
AK
,
He
W
,
Yang
D-S
,
Ghorbani
P
,
Wang
L
, et al
.
Interleukin-18 up-regulates amino acid transporters and facilitates amino acid–induced mTORC1 activation in natural killer cells
.
J Biol Chem
2019
;
294
:
4644
55
.
20.
Coughlin
CM
,
Salhany
KE
,
Wysocka
M
,
Aruga
E
,
Kurzawa
H
,
Chang
AE
, et al
.
Interleukin-12 and interleukin-18 synergistically induce murine tumor regression which involves inhibition of angiogenesis
.
J Clin Invest
1998
;
101
:
1441
52
.
21.
Yao
Z
,
Zhao
M
,
Gao
G
,
Yang
J
,
Wang
Z
,
Liu
Y
.
Prognostic role of IL-18 in various human cancers and radiation injuries: a meta-analysis
.
Dose Response
2020
;
18
:
1559325820931360
.
22.
Zhao
Y
,
Shen
M
,
Feng
Y
,
He
R
,
Xu
X
,
Xie
Y
, et al
.
Regulatory B cells induced by pancreatic cancer cell-derived interleukin-18 promote immune tolerance via the PD-1/PD-L1 pathway
.
Oncotarget
2018
;
9
:
14803
14
.
23.
Ingram
JT
,
Yi
JS
,
Zajac
AJ
.
Exhausted CD8 T cells downregulate the IL-18 receptor and become unresponsive to inflammatory cytokines and bacterial co-infections
.
PLoS Pathog
2011
;
7
:
e1002273
.
24.
Jacobs
C
,
Duewell
P
,
Heckelsmiller
K
,
Wei
J
,
Bauernfeind
F
,
Ellermeier
J
, et al
.
An ISCOM vaccine combined with a TLR9 agonist breaks immune evasion mediated by regulatory T cells in an orthotopic model of pancreatic carcinoma
.
Int J Cancer
2011
;
128
:
897
907
.
25.
Tu
M
,
Klein
L
,
Espinet
E
,
Georgomanolis
T
,
Wegwitz
F
,
Li
X
, et al
.
TNF-α-producing macrophages determine subtype identity and prognosis via AP1 enhancer reprogramming in pancreatic cancer
.
Nat Cancer
2021
;
2
:
1185
203
.
26.
Blackburn
SD
,
Shin
H
,
Haining
WN
,
Zou
T
,
Workman
CJ
,
Polley
A
, et al
.
Coregulation of CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral infection
.
Nat Immunol
2009
;
10
:
29
37
.
27.
Beltra
J-C
,
Manne
S
,
Abdel-Hakeem
MS
,
Kurachi
M
,
Giles
JR
,
Chen
Z
, et al
.
Developmental relationships of four exhausted CD8+ T cell subsets reveals underlying transcriptional and epigenetic landscape control mechanisms
.
Immunity
2020
;
52
:
825
41
.
28.
Siddiqui
I
,
Schaeuble
K
,
Chennupati
V
,
Fuertes Marraco
SA
,
Calderon-Copete
S
,
Pais Ferreira
D
, et al
.
Intratumoral Tcf1+PD-1+CD8+ T cells with stem-like properties promote tumor control in response to vaccination and checkpoint blockade immunotherapy
.
Immunity
2019
;
50
:
195
211
.
29.
Liu
C
,
Somasundaram
A
,
Manne
S
,
Gocher
AM
,
Szymczak-Workman
AL
,
Vignali
KM
, et al
.
Neuropilin-1 is a T cell memory checkpoint limiting long-term antitumor immunity
.
Nat Immunol
2020
;
21
:
1010
21
.
30.
Ross
SH
,
Cantrell
DA
.
Signaling and function of interleukin-2 in T lymphocytes
.
Annu Rev Immunol
2018
;
36
:
411
33
.
31.
Klebanoff
CA
,
Crompton
JG
,
Leonardi
AJ
,
Yamamoto
TN
,
Chandran
SS
,
Eil
RL
, et al
.
Inhibition of AKT signaling uncouples T cell differentiation from expansion for receptor-engineered adoptive immunotherapy
.
JCI Insight
2017
;
2
:
e95103
.
32.
Tang
Z
,
Kang
B
,
Li
C
,
Chen
T
,
Zhang
Z
.
GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis
.
Nucleic Acids Res
2019
;
47
:
W556
60
.
33.
Kim
S-H
,
Cho
E
,
Kim
YI
,
Han
C
,
Choi
BK
,
Kwon
BS
.
Adoptive immunotherapy with transient anti-CD4 treatment enhances anti-tumor response by increasing IL-18Rαhi CD8+ T cells
.
Nat Commun
2021
;
12
:
5314
.
34.
Zhou
T
,
Damsky
W
,
Weizman
O-E
,
McGeary
MK
,
Hartmann
KP
,
Rosen
CE
, et al
.
IL-18BP is a secreted immune checkpoint and barrier to IL-18 immunotherapy
.
Nature
2020
;
583
:
609
14
.
35.
Spranger
S
,
Spaapen
RM
,
Zha
Y
,
Williams
J
,
Meng
Y
,
Ha
TT
, et al
.
Up-regulation of PD-L1, IDO, and T(regs) in the melanoma tumor microenvironment is driven by CD8 + T cells
.
Sci Transl Med
2013
;
5
:
200ra116
.
36.
Liu
Y
,
Zhou
N
,
Zhou
L
,
Wang
J
,
Zhou
Y
,
Zhang
T
, et al
.
IL-2 regulates tumor-reactive CD8+ T cell exhaustion by activating the aryl hydrocarbon receptor
.
Nat Immunol
2021
;
22
:
358
69
.
37.
Utzschneider
DT
,
Gabriel
SS
,
Chisanga
D
,
Gloury
R
,
Gubser
PM
,
Vasanthakumar
A
, et al
.
Early precursor T cells establish and propagate T cell exhaustion in chronic infection
.
Nat Immunol
2020
;
21
:
1256
66
.
38.
Crompton
JG
,
Sukumar
M
,
Roychoudhuri
R
,
Clever
D
,
Gros
A
,
Eil
RL
, et al
.
Akt Inhibition enhances expansion of potent tumor-specific lymphocytes with memory cell characteristics
.
Cancer Res
2015
;
75
:
296
305
.
39.
Man
K
,
Gabriel
SS
,
Liao
Y
,
Gloury
R
,
Preston
S
,
Henstridge
DC
, et al
.
Transcription factor IRF4 Promotes CD8+ T cell exhaustion and limits the development of memory-like t cells during chronic infection
.
Immunity
2017
;
47
:
1129
41
.
40.
Gabriel
SS
,
Tsui
C
,
Chisanga
D
,
Weber
F
,
Llano-León
M
,
Gubser
PM
, et al
.
Transforming growth factor-β-regulated mTOR activity preserves cellular metabolism to maintain long-term T cell responses in chronic infection
.
Immunity
2021
;
54
:
1698
714
.
41.
Vardhana
SA
,
Hwee
MA
,
Berisa
M
,
Wells
DK
,
Yost
KE
,
King
B
, et al
.
Impaired mitochondrial oxidative phosphorylation limits the self-renewal of T cells exposed to persistent antigen
.
Nat Immunol
2020
;
21
:
1022
33
.
42.
Pearce
EL
,
Poffenberger
MC
,
Chang
C-H
,
Jones
RG
.
Fueling immunity: insights into metabolism and lymphocyte function
.
Science
2013
;
342
:
1242454
.
43.
Rao
RR
,
Li
Q
,
Odunsi
K
,
Shrikant
PA
.
The mTOR kinase determines effector versus memory CD8 + T cell fate by regulating the expression of transcription factors T-bet and Eomesodermin
.
Immunity
2010
;
32
:
67
78
.
44.
Weber
A
,
Wasiliew
P
,
Kracht
M
.
Interleukin-1 (IL-1) pathway
.
Sci Signal
2010
;
3
:
7
.
45.
Barnes
SE
,
Wang
Y
,
Chen
L
,
Molinero
LL
,
Gajewski
TF
,
Evaristo
C
, et al
.
T cell-NF-κB activation is required for tumor control in vivo
.
J Immunother Cancer
2015
;
3
:
1
.
46.
Daley
D
,
Mani
VR
,
Mohan
N
,
Akkad
N
,
Pandian
GSDB
,
Savadkar
S
, et al
.
NLRP3 signaling drives macrophage-induced adaptive immune suppression in pancreatic carcinoma
.
J Exp Med
2017
;
214
:
1711
24
.
47.
Ghiringhelli
F
,
Apetoh
L
,
Tesniere
A
,
Aymeric
L
,
Ma
Y
,
Ortiz
C
, et al
.
Activation of the NLRP3 inflammasome in dendritic cells induces IL-1β–dependent adaptive immunity against tumors
.
Nat Med
2009
;
15
:
1170
8
.
48.
Guo
X
,
Zheng
L
,
Jiang
J
,
Zhao
Y
,
Wang
X
,
Shen
M
, et al
.
Blocking NF-κB is essential for the immunotherapeutic effect of recombinant IL18 in pancreatic cancer
.
Clin Cancer Res
2016
;
22
:
5939
50
.
49.
Terme
M
,
Ullrich
E
,
Aymeric
L
,
Meinhardt
K
,
Desbois
M
,
Delahaye
N
, et al
.
IL-18 induces PD-1–dependent immunosuppression in cancer
.
Cancer Res
2011
;
71
:
5393
9
.