Pancreatic Ductal Adenocarcinoma Cells Regulate NLRP3 Activation to Generate a Tolerogenic Microenvironment

Abstract

Defining feature of pancreatic ductal adenocarcinoma (PDAC) that participates in the high mortality rate and drug resistance is the immune-tolerant microenvironment which enables tumors to progress unabated by adaptive immunity. In this study, we report that PDAC cells release CSF-1 to induce nucleotide-binding domain, leucine-rich containing family, pyrin domain-containing-3 (NLRP3) activation in myeloid cells. Increased NLRP3 expression was found in the pancreas of patients with PDAC when compared with normal pancreas which correlated with the formation of the NLRP3 inflammasome. Using human primary cells and an orthotopic PDAC mouse model, we show that NLRP3 activation is responsible for the maturation and release of the inflammatory cytokine IL1β which selectively drives Th2-type inflammation via COX2/PGE2 induction. As a result of this inflammation, primary tumors were characterized by reduced cytotoxic CD8+ T-cell activation and increased tumor expansion. Genetic deletion and pharmacologic inhibition of NLRP3 enabled the development of Th1 immunity, increased intratumoral levels of IL2, CD8+ T cell–mediated tumor suppression, and ultimately limited tumor growth. In addition, we observed that NLRP3 inhibition in combination with gemcitabine significantly increased the efficacy of the chemotherapy. In conclusion, this study provides a mechanism by which tumor-mediated NLRP3 activation exploits a distinct adaptive immunity response that facilitates tumor escape and progression. Considering the ability to block NLRP3 activity with safe and small orally active molecules, this protein represents a new promising target to improve the limited therapeutic options in PDAC.

Significant: This study provides novel molecular insights on how PDAC cells exploit NLRP3 activation to suppress CD8 T-cell activation. From a translational perspective, we demonstrate that the combination of gemcitabine with the orally active NLRP3 inhibitor OLT1177 increases the efficacy of monotherapy.

FIGURE 1

NLRP3 expression and activation are increased in PDAC biopsy. A, NLRP3, ASC, IL1β, and IL18 median expression levels in human PDAC biopsy and normal pancreas samples (TCGA dataset). B, Immunofluorescence staining from NLRP3 (red) and ASC (green) in biopsy of normal pancreas (n = 2) and patients with PDAC (n = 4). Each row represents a single patient. C, Quantification of ASC specks–like structure samples in B. *, P < 0.05.

FIGURE 2

PDAC paracrine activity mediates NLRP3 activation. A, IL1β and IL18 secretion from freshly isolated PBMCs from healthy donors (n = 6) following stimulation with PDAC-conditioned media obtained from AsPC-1, MiaPaCa-2, and Panc-1 cells. B, IL1β and IL18 secretion from parenteral (WT) and NLRP3 deficient (NLRP3^−/−) THP-1 cells following stimulation with PDAC-conditioned media obtained from Panc-1 cells (n = 3). C, IL1β and IL18 secretion from THP-1 cells following stimulation with PDAC-conditioned media obtained from Panc-1 cells in presence of the NLRP3 inhibitors OLT1177 (OLT) and MCC950 (MCC; n = 3). Cytokine levels were measured by specific ELISA kits. Data expressed as mean ± SEM; ****, P < 0.0001; **, P < 0.01; *, P < 0.05.

FIGURE 3

NLRP3 inhibition reduces PDAC progression. A, Tumor volume and weight in WT and Nlrp3^−/− mice (n = 8/group). B, Tumor volume and weight in mice fed standard (STD) or OL1177 (OLT) diet (n = 16/group). C, Survival curve in mice fed standard or OLT1177 diet (n = 20/group). D, Flow cytometry analysis of CD8, CD8/PD1, CD8/CD44/PD-1 cells in primary tumors of mice in B (n = 5/group). E, Flow cytometry analysis of memory (CD62L⁺/CD44⁺), naïve (CD62L⁺/CD44⁻), and effector (CD62L⁻/CD44⁺) CD8 cells in primary tumors of mice in D–F (n = 5/group). F, Gene expression of Ifng, Prf1, and Gzmb in primary tumors of mice in B (n = 5/group). Data expressed as mean ± SEM; ****, P < 0.0001; **, P < 0.01; *, P < 0.05.

FIGURE 4

NLRP3 activation mediates a Th2-like inflammatory response. A, Flow cytometry analysis of CD4, CD4/PD1, CD4/CD44/PD-1 cells in primary tumors of mice fed standard (STD) and OLT1177 (OLT)-enriched diets (n = 5/group). B, Flow cytometry analysis of memory (CD62L⁺/CD44⁺), naïve (CD62L⁺/CD44⁻), and effector (CD62L⁻/CD44⁺) CD4 cells in primary tumors of mice fed STD and OLT-enriched diets (n = 5/group). C, Gene expression of Il2, Il4, and Il10 in primary tumors of mice of mice fed STD and OLT-enriched diets (n = 5/group). D, COX-2 and PGE2 levels measured by ELISA in primary tumors of mice of mice fed STD and OLT-enriched diets (n = 5/group). E, PGE2 levels measured by ELISA in primary tumors of mice of mice fed STD and OLT-enriched diets (n = 5/group). Data expressed as mean ± SEM; ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05.

FIGURE 5

NLRP3 activation mediates lympthocyte suppression. A, Tumor volume and weight in NCG mice fed standard (STD) or OL1177 (OLT) diet (n = 5/group). B, COX-2 levels measured by ELISAs in primary tumors of mice of mice fed STD and OLT-enriched diets. C, PGE2 levels measured by ELISAs in primary tumors of mice of mice fed STD and OLT-enriched diets. Data expressed as mean ± SEM. *, P < 0.05.

FIGURE 6

CSF-1R engagement induces IL1β and IL18 release. A, Correlation between CSF-1R and NLRP3 expression in PDAC (n = 96) from TGCA dataset. B, Spontaneous CSF-1 and IL34 secretion in Panc-1 cells at 72 hours (n = 3). C, IL1β and IL18 secretion in THP-1 cells following stimulation with PDAC-conditioned media obtained from Panc-1 cells in presence of IgG (control), α-CSF-1, α-IL34, α-CSF-1/IL34 (n = 3). D, Phosphorylation levels of PyK2 (Tyrosine 402) in THP-1 stimulated in D and E (n = 3). E, IL1β and IL18 secretion from THP-1 cells following stimulation with PDAC-conditioned media obtained from Panc-1 cells in presence of α-CSF-1R and an CSF-1R inhibitor (AZD; n = 3). F, IL1β secretion from freshly isolated PBMCs from healthy donors (n = 4) following stimulation with PDAC-conditioned media obtained from Panc-1 cells in presence of α-CSF-1R. Cytokine levels were measured by specific ELISAs. Data expressed as mean ± SEM; ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05.

FIGURE 7

CSF-1R engagement induces NLRP3 activation in vivo. A, Tumor volume and weight in mice treated with α-CSF-1R or matching IgG (n = 6/group). B, Immunofluorescence staining from NLRP3 (red) and ASC (green) in primary tumors treated with α-CSF-1R or matching IgG (n = 3/group). C, Quantification of speck-like structure in mice treated with α-CSF-1R or matching IgG (n = 3/group). D, Schematic graphic representation of the proposed NLRP3 activation and signaling in PDAC (created with BioRender.com).

FIGURE 8

NLRP3 inhibition increases gemcitabine efficacy. A, Tumor volume and weight in mice treated with vehicle (PBS), gemcitabine (GEM), OLT1177 (OLT), and GEM+OLT (n = 8/group). B, Flow cytometry analysis of Ki67 expression in CD45⁻ cells from the primary tumors obtained in A (n = 8/group). C, Viability of human PDAC organoids after 72 hours of culture in presence of gemcitabine (GEM) or OLT1177 (OLT; n = 3). D, Representative images of organoids from C. E, NLRP3 expression in primary tumors obtained in A (n = 3/group). Data expressed as mean ± SEM; ****, P < 0.0001; ***, P < 0.001; **, P < 0.01; *, P < 0.05.