Activation of the NLRP3 inflammasome complex is not required for stress-induced death of pancreatic islets

Abstract

Loss of pancreatic beta cells is a feature of type-2 diabetes. High glucose concentrations induce endoplasmic reticulum (ER) and oxidative stress-mediated apoptosis of islet cells in vitro. ER stress, oxidative stress and high glucose concentrations may also activate the NLRP3 inflammasome leading to interleukin (IL)-1β production and caspase-1 dependent pyroptosis. However, whether IL-1β or intrinsic NLRP3 inflammasome activation contributes to beta cell death is controversial. This possibility was examined in mouse islets. Exposure of islets lacking functional NLRP3 or caspase-1 to H2O2, rotenone or thapsigargin induced similar cell death as in wild-type islets. This suggests that oxidative or ER stress do not cause inflammasome-mediated cell death. Similarly, deficiency of NLRP3 inflammasome components did not provide any protection from glucose, ribose or gluco-lipotoxicity. Finally, genetic activation of NLRP3 specifically in beta cells did not increase IL-1β production or cell death, even in response to glucolipotoxicity. Overall, our results show that glucose-, ER stress- or oxidative stress-induced cell death in islet cells is not dependent on intrinsic activation of the NLRP3 inflammasome.

Figure 1. Oxidative and ER stress induced islet cell death is not mediated by inflammasome activation.

(A) NIT cells were cultured in control medium, or medium containing 100 ng/mL LPS, 5 µg/mL IL-1Ra or 100 nM rotenone for 2 days. The frequency of cells undergoing DNA fragmentation was measured by flow cytometry. Results are mean+SEM of n≥3 independent experiments. ***p<0.001 Rotenone treated NIT cell groups vs controls without rotenone. (B) Representative FACS profiles of DNA fragmentation in islets after treatment with 100 nM rotenone for 2 days, 20 µM H₂O₂ for 2 days or 5 µM thapsigargin for 5 days. The percentage of islet cells with fragmented nuclei is indicated. (C) C57BL/6, Nlrp3^−/− or caspase-1^−/− islets were cultured in control medium or medium containing 100 nM rotenone for 2 days. Control islets were incubated in medium containing an equal volume of solvents. Results are mean+SEM of n≥3 independent experiments. ***p<0.001 rotenone treated islets vs controls of same genotype. (D) DNA fragmentation was measured by flow cytometry after incubation of wild-type C57BL/6, Nlrp3^−/− or caspase-1^−/− islets with 0–40 µM H₂O₂ for 2 days. Results are mean+SEM of n≥3 independent experiments. *p<0.05, **p<0.01, ***p<0.001 H₂O₂ treated islets vs untreated islets of same genotype. (E, F) Wild-type islets (E), or wild-type, Nlrp3^−/− or caspase-1^−/− islets (F) were cultured in control medium, or medium containing 5 µg/mL IL-1Ra or 5 µM thapsigargin for 5 days. Control islets were incubated in medium containing an equal volume of solvents. The frequency of cells undergoing DNA fragmentation was measured by flow cytometry. Results are mean+SEM of n≥3 independent experiments. ***p<0.001 Thapsigargin treated islets vs controls without thapsigargin of same genotype. No significant differences were observed between the different genotypes.

Figure 2. Glucose toxicity of islet cells is not mediated by inflammasome activation.

(A) Representative FACS profiles of DNA fragmentation in islets after treatment with 33.3 mM glucose for 6 days, 50 mM ribose for 4 days or 1 mM palmitate conjugated to 1% BSA for 6 days. The percentage of islet cells with fragmented nuclei is indicated. (B, C) DNA fragmentation was measured by flow cytometry after incubation of wild-type C57BL/6, Nlrp3^−/− or caspase-1^−/− islets with 33.3 mM glucose for 6 days (B) or 50 mM ribose for 4 days (C). Control islets were incubated in medium containing 5.5 mM glucose. Results are mean+SEM of n≥4 independent experiments. *p<0.05 glucose treated islets vs control islets of same genotype. ***p<0.001 ribose treated islets vs control islets of same genotype. (D) DNA fragmentation was measured by flow cytometry after incubation of wild-type or caspase-1^−/− islets with 50 mM ribose or 100 ng/mL LPS for 3 days. Results are mean+SEM of n = 3 independent experiments. **p<0.01, ***p<0.001 ribose treated islets vs controls of the same genotype without ribose. (E) DNA fragmentation was measured by flow cytometry after incubation of wild-type or caspase-1^−/− islets with 33.3 mM glucose or 1 mM palmitate conjugated to 1% BSA for 6 days. Control islets were incubated in a medium containing 5.5 mM glucose and 1% BSA. Results are mean+SEM of n = 3 independent experiments. *p<0.05 glucose vs glucose+palmitate-treated caspase-1^−/− islets, **p<0.01 glucose vs glucose+palmitate-treated wild-type islets.

Figure 3. NLRP3 activating mutation does not increase islet glucose toxicity.

(A) DNA was isolated from 300–500 islets of the indicated genotypes and PCR was performed using appropriate primers. Islets isolated from wild-type (+/+ NLRP3^+/+) mice were used for lanes 1, 2, 4, 6 and 8, while lanes 3, 5 and 7 show genotyping results for islets expressing mutant NLRP3 (Cre/+ NLRP3^A350V/A350V). (B) Islets from wild-type mice, floxed mice with activating NLRP3 mutation knocked into NLRP3 locus (NLRP3^A350V/+ or NLRP3^A350V/A350V), and mice expressing Cre recombinase in beta cells (Cre/+) were treated with 50 mM ribose for 4 days. DNA fragmentation was measured by flow cytometry. Results are mean+SEM of n = 3 independent experiments. ***p<0.001 ribose-treated islets vs control islets of the same genotype.

Figure 4. NLRP3 activating mutation does not induce caspase-1 cleavage in islets.

(A) Four hundred islets isolated from wild-type or Cre/+ NLRP3^A350V/A350V mice were cultured in 1 mL of medium containing 33.3 mM glucose or 50 mM ribose for 2.5 days and IL-1β concentration in the supernatant was quantified by ELISA. Control islets were incubated in a medium containing 5.5 mM glucose. Macrophages treated overnight with 1 mM palmitate were used as positive control. Results are mean+SEM of n = 3–4 independent experiments. (B) Four hundred islets/sample were cultured in control medium containing 1% BSA and 100 nM LPS or a medium containing 1% BSA, 100 nM LPS, 33.3 mM glucose and 1 mM palmitate for 2.5 days. Lysates were prepared in RIPA buffer and western blotting for caspase-1 (full length, FL or cleaved, p20) was performed. Genotypes 1: +/+ NLRP3^+/+, 2: Cre/+ NLRP3^+/+, 3: +/+ NLRP3^A350V/A350V and 4: Cre/+ NLRP3^A350V/A350V. As a positive control, macrophages from wild-type (A) and caspase-1^−/− (B) mice were treated with LPS+BSA+Glucose+Palmitate for 24 h or LPS+Nigericin for 1 h.