An inflammasome-independent role for epithelial-expressed Nlrp3 in renal ischemia-reperfusion injury

Abstract

Cytoplasmic innate immune receptors are important therapeutic targets for diseases associated with overproduction of proinflammatory cytokines. One cytoplasmic receptor complex, the Nlrp3 inflammasome, responds to an extensive array of molecules associated with cellular stress. Under normal conditions, Nlrp3 is autorepressed, but in the presence of its ligands, it oligomerizes, recruits apoptosis-associated speck-like protein containing a caspase recruitment domain (Asc), and triggers caspase 1 activation and the maturation of proinflammatory cytokines such as IL-1β and IL-18. Because ischemic tissue injury provides a potential source for Nlrp3 ligands, our study compared and contrasted the effects of renal ischemia in wild-type mice and mice deficient in components of the Nlrp3 inflammasome (Nlrp3(-/-) and Asc(-/-) mice). To examine the role of the inflammasome in renal ischemia-reperfusion injury (IRI) we also tested its downstream targets caspase 1, IL-1β, and IL-18. Both Nlrp3 and Asc were highly expressed in renal tubular epithelium of humans and mice, and the absence of Nlrp3, but not Asc or the downstream inflammasome targets, dramatically protected from kidney IRI. We conclude that Nlrp3 contributes to renal IRI by a direct effect on renal tubular epithelium and that this effect is independent of inflammasome-induced proinflammatory cytokine production.

Figure 1. Nlrp3 and Asc in murine and human cells

Figure 1 shows three panels depicting the Nlrp3, Asc or housekeeping GADPH genes. Each panel illustrates gene expression, detected by semiquantitative PCR in renal tubular epithelium of either Nlrp3^−/− mice (Nlrp3^−/− mRTE), wild type (WTmRTE) mice, lymph node cells from WT mice (WTLN), nontransfected HEK 293 cells (HEK), Thp1 cells (Thp1) or human renal tubular epithelial cells extracted from normal human kidneys (hRTE). The results are representative of 4 mice per group and 5 different human kidneys extracted for renal tubular cells with identical results.

Figure 2. Kidney ischemia-reperfusion injury in WT, NLR and caspase-1 deficient mice

WT, Asc^−/−, caspase 1^−/− or Nlrp3^−/− mice were subjected to 25 min ischemia/24hr reperfusion and plasma obtained at 24 hrs to detect creatinine from injured WT, Asc^−/−, caspase 1^−/−, and Nlrp3^−/− mice (n=6 mice per each group). Preinjury creatinines were obtained 2 weeks earlier on each of experimental mice; WT is shown in the figure (labeled, Pre); Asc^−/− (0.42±0.01); caspase 1^−/− (0.47±0.01); and Nlrp3^−/− (0.48±0.04). Creatinines were also obtained from sham operated mice: WT (0.4±0.1); Asc^−/− (0.3±0.1); Caspase 1^−/− (0.4±0.2); and Nlrp3^−/− (0.3±0.2). Error bars represent SD of creatinines and statistical significance was determined with a one-way ANOVA with Bonferroni’s multiple comparison test. The data represent one of five identical experiments with the same results.

Figure 3. Histological renal injury after bilateral renal artery clamping

Figure 3, Panel A shows blinded scoring of tubular necrosis from WT (n=6), caspase 1^−/− (n=6), Asc^−/− (n=6), Nlrp3^−/− (n=6) and sham (n=3) operated mice 24 hrs after bilateral renal artery clamping (25 min ischemia/24hr reperfusion). The top panel shows the blinded necrosis score (see materials and methods for details) and the bottom panel shows a representative micrograph of each group of mice (PAS, 200X). Means of necrosis scores were as follows: WT (5.0±1.1); caspase 1^−/− (4.7±1.0); Asc^−/− (3.7±2.2); Nlrp3^−/− (2.3±0.8). There were no statistical differences between WT vs. caspase 1^−/− (p>0.50), or WT vs. Asc^−/− (p>0.50); but significant differences in histological injury were observed between WT vs. Nlrp3^−/− (p<0.01) mice. Panel B shows blinded scores of neutrophil infiltration on the same sections (see materials and methods). Means of neutrophil scores were: WT (2.1±0.8), caspase 1^−/− (1.8±1.3), Asc^−/− (1.6±1.1), Nlrp3^−/− (1.4 ±0.7). No statistical differences in neutrophil infiltration were seen between the groups, WT vs. caspase 1^−/− (p>0.05), vs. Asc^−/− (p>0.05), or vs. Nlrp3^−/− (p>0.05). Statistical significance was determined by a one-way ANOVA with Bonferroni’s multiple comparison test.

Figure 4. Cytokine secretion after ischemia reperfusion or injection of LPS

Panel A shows WT or Nlrp3^−/− mice (n=6/gp) subjected to 25 min ischemia and indicated times of reperfusion. At the indicated time points plasma (for IL-18) or kidney homogenates (for IL-1 β) was assayed for the cytokines (top panels, IRI). Bottom panels show plasma cytokine secretion over time after LPS injection in WT (λ) or Nlrp3^−/− (σ) mice (n=6/gp). The data are representative of 3 experiments with the same results. Error bars represent standard deviations. Panel B shows IL-1β and caspase 1 in WT vs. Nlrp3^−/− kidneys from sham operated (sham) or IRI mice. The top panel shows mature IL-1β (mature IL-1β identified by the p17 fragment). The middle panel shows caspase 1 (active subunit p20) and the lower panel shows hsp-90 for a loading control. This represents one of 3 experiments with identical results.

Figure 5. IL-1 and IL-18 blockade and kidney IRI

Figure 5, Panel A shows WT (WT), WT mice treated with Anakinra (per materials and methods) (Anakinra), or IL-1R^−/− (IL-1R^−/−) mice that were treated with 25 min ischemia/24hr reperfusion followed by plasma creatinine determination at 24 hrs. Sham mice underwent laparotomy without bilateral renal artery clamping. These data represent 6 mice per group. There were no significant differences between WT vs. Anakinra (p>0.05), WT vs. IL-1R^−/− (p>0.05) groups. Error bars represent SDs. Panel B shows blinded necrosis scores of the IR injured mice. The means of necrosis score were: WT (4.2±0.8), Anakinra (4.5±0.8), IL-1R^−/− (4.0±0.6). There were no statistical differences in necrosis scores: Wt vs. Anakinra (p>0.05), vs. IL-1R^−/− (p>0.05). The means of neutrophil infiltration were: WT (2.3±0.8), Anakinra (1.7±0.5), IL-1R^−/− (2.0±0.6); and statistical differences – WT vs. Anakinra (p>0.05), vs. IL-1R^−/− (p>0.05). Statistical significance was determined by a one-way ANOVA with Bonferroni’s multiple comparison test.

Figure 6. Bone marrow chimeric mice subjected to IRI

Panel A shows WT mice that were irradiated and injected with either WT (n=6), Asc−/− (n=6) or Nlrp3^−/− (n=6) bone marrow 6 weeks earlier. After engraftment the three groups of mice were treated with 25 min bilateral renal artery clamping followed by 24hrs reperfusion. Pre creatinine values were obtained 2 weeks before injury (pre). Sham creatinines were obtained from mice that underwent laparotomy without undergoing bilateral renal artery clamping (n=3). Error bars represent SDs. Panel B shows the necrosis score of the injured animals. The means of scores were: WT (4.3±0.8), Asc^−/− (4.0±1.1), Nlrp3^−/− (4.3±0.8) and statistical differences: Wt vs. Asc^−/− (p>0.05), vs. Nlrp3^−/− (p>0.05). Panel C shows the scores of neutrophil infiltration: WT (2.8±0.4), Asc^−/− (2.5±0.5), Nlrp3^−/− (2.3±0.8) and the statistical differences were: WT vs. Asc^−/− (p>0.05), WT vs. Nlrp3^−/− (p>0.05). Statistical significance was obtained by a one-way ANOVA with Bonferroni’s multiple comparison test. Panel D shows that chimeric bone marrow was engrafted in the indicated experimental groups. The control panels show the Nlrp3 knockout allele (indicating the Nlrp3 gene) was present in the Nlrp3^−/− mice, but not in the WT or the Asc^−/− mice (controls). The WT mice transplanted with the Nlrp3^−/− bone marrow show that the Nlrp3 knockout allele was highly expressed (bone marrow chimeras, Nlrp3^−/− ⇒ WT). Likewise the Asc knockout allele was expressed in Asc^−/− controls, but not in the WT controls and also highly expressed in the mice transplanted with Asc^−/− bone marrow (bone marrow chimeras, Asc^−/− ⇒ WT).

Figure 7. Apoptosis of renal tubular epithelial cells after renal IRI

WT vs. Nlrp3^−/−, Asc^−/−, or sham mice were treated with 25 min of bilateral renal artery clamping followed by 4hr of reperfusion. The mice were sacrificed, their kidneys removed and placed into formalin for paraffin embedding and TUNEL assay. Panel A shows representative images of TUNEL and DAPI staining 4hrs after reperfusion. Panel B shows results of blinded counting of TUNEL-positive cells, 4hr after reperfusion, in nine random, non-overlapping, sections (n=3 mice per group). Data are expressed as means ± SD. Statistical significance was determined by a one-way ANOVA with Bonferroni’s multiple comparison test.