B cell-Derived IL35 Drives STAT3-Dependent CD8+ T-cell Exclusion in Pancreatic Cancer

Abstract

Pancreatic ductal adenocarcinoma (PDA) is an aggressive malignancy characterized by a paucity of tumor-proximal CD8⁺ T cells and resistance to immunotherapeutic interventions. Cancer-associated mechanisms that elicit CD8⁺ T-cell exclusion and resistance to immunotherapy are not well-known. Here, using a Kras- and p53-driven model of PDA, we describe a mechanism of action for the protumorigenic cytokine IL35 through STAT3 activation in CD8⁺ T cells. Distinct from its action on CD4⁺ T cells, IL35 signaling in gp130⁺CD8⁺ T cells activated the transcription factor STAT3, which antagonized intratumoral infiltration and effector function of CD8⁺ T cells via suppression of CXCR3, CCR5, and IFNγ expression. Inhibition of STAT3 signaling in tumor-educated CD8⁺ T cells improved PDA growth control upon adoptive transfer to tumor-bearing mice. We showed that activation of STAT3 in CD8⁺ T cells was driven by B cell- but not regulatory T cell-specific production of IL35. We also demonstrated that B cell-specific deletion of IL35 facilitated CD8⁺ T-cell activation independently of effector or regulatory CD4⁺ T cells and was sufficient to phenocopy therapeutic anti-IL35 blockade in overcoming resistance to anti-PD-1 immunotherapy. Finally, we identified a circulating IL35⁺ B-cell subset in patients with PDA and demonstrated that the presence of IL35⁺ cells predicted increased occurrence of phosphorylated (p)Stat3⁺CXCR3^-CD8⁺ T cells in tumors and inversely correlated with a cytotoxic T-cell signature in patients. Together, these data identified B cell-mediated IL35/gp130/STAT3 signaling as an important direct link to CD8⁺ T-cell exclusion and immunotherapy resistance in PDA.

Figure 1.. IL35 mediates activation of STAT3, suppression of IFNɣ, and chemotactic receptors CXCR3 and CCR5 in gp130⁺CD8⁺ T cells.

(A) Representative flow cytometry histograms (left) and quantification (right) of phospho(p)-STAT1, pSTAT3, and pSTAT4 in CD3⁺CD8⁺ T cells, activated with anti-CD3 (1 μg/mL) and anti-CD28 (2 μg/mL) in the presence of OVA (2 μg/mL) with and without recombinant (r)IL35 (50 ng/mL). Proportion of CD8⁺ T cells is indicated. (B) Representative flow cytometry plots and histograms (left) and quantification (right) of IFNɣ production and expression of CXCR3 and CCR5 in CD3⁺CD8⁺ T cells activated as in (A). (C) Representative flow cytometry histograms (left) and quantification (right) of pSTAT1, pSTAT3, and pSTAT in CD3⁺CD8⁺ T cells sorted for IL12Rβ2⁺ or gp130⁺ subsets (single-positive) activated as indicated in (A). Proportion of CD8⁺ T cells is shown. (D) Representative flow cytometry plots and histograms (left) and quantification (right) of IFNɣ production and expression of CXCR3 and CCR5 in CD3⁺CD8⁺ T cells sorted for IL12Rβ2⁺ or gp130⁺ subsets (single-positive) activated as indicated in (A). Proportion of CD8⁺ T cells is shown. Error bars indicate S.E.M.; p-values were calculated using Student’s t-test (unpaired, two-tailed); NS: not significant, *p<0.05; **p<0.01; ***p<0.001. Experiment were conducted using 6–8–week-old C57B6 mice with 6 mice per group in triplicate. Data represents 3 independent experiments.

Figure 2.. STAT3 activation in CD8⁺ T cells is suppressive in vitro and in vivo.

(A) Representative flow cytometry plots and histograms of IFNɣ production and expression of CXCR3, CCR5, phospho(p)-STAT1, pSTAT3, and pSTAT4 in CD3⁺CD8⁺ T cells stimulated with anti-CD3 (1 μg/mL) and anti-CD28 (2 μg/mL) in the presence of OVA (2 μg/mL), and rIL35 (50 ng/mL) and/or STA-21 (20μM) inhibitor. Proportion of CD8⁺ T cells is indicated. (B) Quantification of IFNɣ production and expression of CXCR3, CCR5, pSTAT1, pSTAT3, and pSTAT4 in CD3⁺CD8⁺ T cells from (A). (C) Experimental schema used to test the role of Stat3 activation in CD8⁺ T cells. (D) Quantification of tumor weight from CD45.1⁺ mice 3 weeks post-orthotopic adoptive transfer of pre-treated with anti-CD3/CD28 with and without STA-21 CD45.2⁺CD8⁺ T cells and injection with KPC cells (n=6 mice/group). (E) Numbers (no.) of adoptively transferred CD45.2⁺ (left) and endogenous CD45.1⁺ (right) tumor-infiltrating CD3⁺CD8⁺ T cells 3 weeks post-orthotopic adoptive transfer to the CD45.1⁺ recipients indicated in (D). (F) Representative flow cytometry histograms (left) and quantification (right) of pSTAT1, pSTAT3, and pSTAT4 in intratumoral CD45.2⁺CD8⁺ T cells 3 weeks post-orthotopic adoptive transfer to CD45.1⁺ recipients indicated in (D). (G) Representative flow cytometry plots and histograms (left) and quantification (right) of IFNɣ production and expression of CXCR3 and CCR5 in intratumoral CD45.2⁺CD8⁺ T cells 3 weeks post-orthotopic adoptive transfer to CD45.1⁺ recipients indicated in (D). Error bars indicate S.E.M.; p-values were calculated using Student’s t-test (unpaired, two-tailed); NS: not significant, **p<0.01; ***p<0.001. Data represents 3 independent experiments.

Figure 3.. B cell–derived IL35 promotes tumor growth and mediates suppression of IFNɣ and chemotactic receptors CXCR3 and CCR5 in CD8⁺ T cells.

(A) Experimental schema used to generate tumor-bearing mixed bone marrow chimeras containing B cell–specific deletion of p35 (B^p35–/–) and corresponding control B^WT mice. (B) Quantification of tumor weight from B^WT and B^p35–/– mice 3 weeks post-orthotopic injection with KPC4662 cells (n=6 mice/group). (C) Representative flow cytometry histograms (left) and quantification (right) of intracellular p35 expression in intratumoral and intrasplenic CD19⁺CD21^hiCD5⁺CD1d^hi Bregs from mice in (B). Proportion of p35⁺ Bregs is indicated. (D) Representative flow cytometry histograms (left) and quantification (right) of intratumoral CD45⁺CD3⁺CD4⁺CD25^– effector T cells in B^WT and B^p35–/– mice from (B). (E) Representative flow cytometry plots (left) and quantification (right) of intracellular IFNɣ (Isotype; Rat IgG1, κ) and TNFα (Isotype; Rat IgG1, κ) expression by CD3⁺CD4⁺ intratumoral T cells from mice in (B). Proportion of CD4⁺ T cells is indicated. (F) Representative flow cytometry histograms (left) and quantification (right) of intratumoral CD3⁺CD4⁺Foxp3⁺ Tregs from mice in (B). Proportion of CD4⁺ T cells is indicated. (G) Representative flow cytometry histograms (left) and quantification (right) of intracellular p35 (Isotype; Rat IgG2a, ĸ) and IL10 (Isotype; Rat IgG2b, κ) expression by intratumoral CD3⁺CD4⁺ T cells from mice in (B). (H) Quantification of frequency of tumor-infiltrating CD45⁺CD3⁺CD8⁺ T cells from B^WT and B^p35–/– mice from (B) determined by flow cytometry. (I) Representative flow cytometry plots (left) and quantification (right) of IFNγ (Isotype; Rat IgG1, κ) in intratumoral CD45⁺CD3⁺CD8⁺ T cells from mice in (B). (J) Ratio of mean CD3⁺CD4⁺CD25^– effector T cells (Teff) to Tregs (left) and ratio of mean CD3⁺CD8⁺ cytotoxic T cells to Tregs (right) were calculated based on the percent-positive lymphocyte population determined by flow cytometry. Error bars indicate S.E.M.; p-values were calculated using Student’s t-test (unpaired, two-tailed); NS: not significant, **p<0.01; ***p<0.001. Data represents 3 independent experiments.

Figure 4.. Production of IL35 by B cells impedes anti-PD1 therapy.

(A) Experimental schema used to generate tumor-bearing mice that contain B cell–specific deletion of Ebi3 (B^Ebi3–/–) and corresponding control B^Ebi3+/– mice. (B) Quantification of tumor weight from anti–PD-1–treated (200 μg) and IgG-treated (200 μg) B^Ebi3+/– and B^Ebi3–/– mice 3 weeks post-orthotopic injection with KPC cells (n=6 mice/group). (C) Quantification of tumor growth by ultrasound from B^Ebi3+/− and B^Ebi3−/− mice from (B). (D) Quantification of frequency of tumor-infiltrating CD45⁺CD3⁺CD8⁺ T cells from B^Ebi3+/− and B^Ebi3−/− mice from (B) determined by flow cytometry. (E) Representative flow cytometry plots of intratumoral CD45⁺CD3⁺CD8⁺ T cells stained for IFNγ (Isotype; Rat IgG1, κ) from mice in (B). (F) Quantification of intratumoral CD8⁺IFNγ⁺ T cells from all mice in (B). (G) Representative flow cytometry histograms (left) and quantification (right) of expression of CXCR3, CCR5, phospho(p)-STAT1, pSTAT3, and pSTAT4 in intratumoral CD3⁺CD8⁺ T cells isolated from B^Ebi3+/– and B^Ebi3–/– mice from (B). (H) Representative flow cytometry plots and histograms (left) and quantification (right) of IFNɣ production and expression of CXCR3, CCR5, pSTAT1, pSTAT3, and pSTAT4 in CD3⁺CD8⁺ T cells 48 hours after coculture with Breg cells in 1:1 ratio derived from the indicated mice. Experiment were done in triplicate with 6 mice each group. Error bars indicate S.E.M.; p-values were calculated using Student’s t-test (unpaired, two-tailed); NS: not significant, *p<0.05; **p<0.01; ***p<0.001. Data represents 3 independent experiments.

Figure 5.. IL35 blockade relieves immunosuppression of CD8⁺ T cells and synergizes with anti–PD-1.

(A) Schematic of the antibody treatment regimen. Anti-IL35 (200 μg first dose followed by 100 μg per week) or control IgG antibody was administered in therapeutic schedule (1 week after tumor cell injection on days 7, 11, and 15). Administration of anti–PD-1 (200 μg) was initiated on day 7 after tumors reached approximately 4–5 mm in diameter. Two more doses of anti–PD-1 were administered on days 9 and 11. Mice were sacrificed 3 weeks post-tumor cell injection or assessed for survival. (B) Quantification of tumor weight from WT mice treated with therapeutic anti-IL35, anti–PD-1, or combination (as in (A)) 3 weeks post-orthotopic injection with KPC cells (n=6 mice/group). (C) Quantification of tumor growth by ultrasound from WT mice in (B). (D) Survival plot of orthotopically injected WT mice from (B) treated with the indicated therapeutic antibodies. (E) Quantification of tumor growth by ultrasound from spontaneous KPC mice treated with therapeutic anti-IL35, anti–PD-1, or combination as in (A). Treatment was initiated when tumor measuring ~5mm was visualized by ultrasound. (n=6 mice/group). Data represents 3 independent experiments. (F) Quantification of frequency of orthotopic tumor-infiltrating CD45⁺CD3⁺CD8⁺ T cells from mice in (B). (G) Representative flow cytometry plots of intracellular IFNγ in intratumoral CD45⁺CD3⁺CD8⁺ T cells from the mice in (B). Proportion of CD8⁺ T cells is indicated. (H) Quantification of intratumoral CD8⁺IFNγ⁺ T cells from the mice in (B). (I) Quantification of CXCR3, CCR5, pSTAT3, and pSTAT4 expression in intratumoral CD3⁺CD8⁺ T cells from mice in (B). Error bars indicate S.E.M.; p-values were calculated using Student’s t-test (unpaired, two-tailed); NS: not significant, **p<0.01; ***p<0.001. Data represents 3 independent experiments.

Figure 6.. Identification of IL35-producing B cells in patients with PDA.

(A) Quantification of CD19⁺CD24^hiCD38^hi B cells in peripheral blood of healthy volunteers (n=30) and treatment-naïve PDA patients (n=30). Proportion of CD19⁺ cells is indicated. (B) Fold-change in Il10, p35, and Ebi3 from sorted CD19⁺CD24^hiCD38^hi Bregs or CD19⁺CD24^loCD38^lo Bcon cells from healthy volunteers or PDA patients (n=5 samples/B-cell group) determined by qPCR. Fold-change determined by comparing to healthy Bcon cells. (C) Fold-change in expression of Il10, p35, and Ebi3 from sorted CD19⁺CD24^hiCD38^hi Bregs in healthy donors or PDA patients determined by qPCR. (D) Representative immunofluorescence staining for CD8, CXCR3, and pSTAT3 in samples of human PDA. Arrow indicates pSTAT3⁺CD8⁺ T cells, arrowhead indicates pSTAT3^–CD8⁺ T cells, and asterisks indicate pSTAT3⁺CD8^– cells. Scale bars, 25μm. (E) Proportion of Ebi3-high and -low tumor regions as a function of immune aggregates (IA). Data was derived from counting 3–6 field-of-view (FOV)/tumor sample (n=11 tumor samples). (F) Proportion of pSTAT3⁺CD8⁺ T cells as function of low vs. high numbers of Ebi3⁺ immune cells. Each data point is the percent of pSTAT3⁺CD8⁺ T cells out of all CD8⁺ T cells/20x FOV. Data was derived from counting 3–6 FOV/tumor sample (n=11 tumor samples). (G) Proportion of CXCR3⁺CD8⁺ T cells as function of low vs. high numbers of Ebi3⁺ immune cells. Each data point is the percent of CXCR3⁺CD8⁺ T cells out of all CD8⁺ T cells/20x FOV. Data was derived from counting 3–6 FOV/tumor sample (n=11 tumor samples). (H) Proportion of CXCR3⁺CD8⁺ T cells as function of pSTAT3⁺ vs pSTAT3^– in tumor regions high for Ebi3⁺ immune cells. Each data point is the percent of CXCR3⁺CD8⁺ T cells out of all CD8⁺ T cells/20x FOV. Data was derived from counting 3–6 FOV/tumor sample (n=11 tumor samples). (I) Correlation of the cancer Breg signature with the cytotoxic CD8⁺ T-cell index in PAAD from TCGA. (J) Correlation of the cancer Breg signature with the cytotoxic CD8⁺ T-cell index across TCGA subtypes. Error bars indicate S.E.M., p-values were calculated using Student’s t-test (unpaired, two-tailed); NS: not significant, **p<0.01; ***p<0.001. Data represents 3 independent experiments.