NLRP3 activation in neutrophils induces lethal autoinflammation, liver inflammation, and fibrosis (original) (raw)

Abstract

Sterile inflammation is a central element in liver diseases. The immune response following injurious stimuli involves hepatic infiltration of neutrophils and monocytes. Neutrophils are major effectors of liver inflammation, rapidly recruited to sites of inflammation, and can augment the recruitment of other leukocytes. The NLRP3 inflammasome has been increasingly implicated in severe liver inflammation, fibrosis, and cell death. In this study, the role of NLRP3 activation in neutrophils during liver inflammation and fibrosis was investigated. Mouse models with neutrophil‐specific expression of mutant NLRP3 were developed. Mutant mice develop severe liver inflammation and lethal autoinflammation phenocopying mice with a systemic expression of mutant NLRP3. NLRP3 activation in neutrophils leads to a pro‐inflammatory cytokine and chemokine profile in the liver, infiltration by neutrophils and macrophages, and an increase in cell death. Furthermore, mutant mice develop liver fibrosis associated with increased expression of pro‐fibrogenic genes. Taken together, the present work demonstrates how neutrophils, driven by the NLRP3 inflammasome, coordinate other inflammatory myeloid cells in the liver, and propagate the inflammatory response in the context of inflammation‐driven fibrosis.

Synopsis

The alternative text for this image may have been generated using AI.

Sterile inflammation is central to many liver diseases. This work shows how neutrophils coordinate other inflammatory myeloid cells in the liver and propagate the inflammatory response in the context of inflammation‐driven fibrosis.

Similar content being viewed by others

Introduction

Sterile inflammation is a central element in the development of liver diseases including fibrosis (Kubes & Mehal, 2012; Szabo & Petrasek, 2015; Alegre et al, 2017). Interactions between immune and non‐immune cells mediate an inflammatory response that results in the restoration of liver homeostasis or its progression to chronic liver inflammation and fibrosis (Kubes & Jenne, 2018). Inflammasomes are intracellular multiprotein complexes that are key regulators of the innate immune response and are critical in defining outcomes of hepatic inflammation. Recognition of danger and intracellular stress signals culminate in the assembly of the inflammasome and pro‐inflammatory cytokine release, changes in hepatic cell fate, and can set the stage for chronic sterile inflammatory diseases including nonalcoholic fatty liver disease (NAFLD) and alcoholic liver disease (ALD; Wree et al, 2013, 2014a; Luan & Ju, 2018; Van Opdenbosch & Lamkanfi, 2019).

The NLR Family Pyrin Domain Containing 3 (NLRP3) inflammasome has been implicated in severe liver inflammation by induction of hepatocyte pyroptotic cell death, thereby exacerbating fibrosis and promoting hepatic stellate cell (HSC) activation as well as liver damage (Wree et al, 2014a; Mridha et al, 2017; Inzaugarat et al, 2019). Increasing recognition of the communication between cells of the immune system and hepatocytes has helped in understanding how NLRP3 contributes to the development of liver disease (Heymann & Tacke, 2016; Koyama & Brenner, 2017). NLRP3 is expressed in immune cells, hepatocytes, and non‐parenchymal cells of the liver such as hepatic stellate cells. The stimulation of NLRP3 activates caspase‐1 (Casp1), subsequently releasing mature interleukin (IL)‐1β and contributing to a pro‐inflammatory response (Heymann & Tacke, 2016; Koyama & Brenner, 2017; Wree et al, 2018; Barbier et al, 2019).

The immune response in the liver following injurious stimuli is adaptable according to environmental stimuli and intercellular communication mediated in part by inflammatory cytokines (Heymann & Tacke, 2016; Koyama & Brenner, 2017; Krenkel & Tacke, 2017; Kubes & Jenne, 2018). Neutrophils are generated in the BM. Long‐term hematopoietic stem cells (LT‐HSC) commit to a series of developmental checkpoints including short‐term hematopoietic stem cells (ST‐HSC) to multipotent progenitors (MPP) to common myeloid progenitors (CMP), which branch into unipotent monocyte progenitors (MonP) and unipotent neutrophil progenitors (NeP). MonP gives rise to monocyte lineage cells including monocytes and monocyte‐derived macrophages and dendritic cells (DC), while NeP gives rise to immature and mature neutrophils (Zhu et al, 2018). Mature neutrophils are the first responder immune cells that are rapidly recruited to the site of inflammation and can augment the recruitment and activation status of other leukocytes such as monocytes (Heymann & Tacke, 2016; Ye et al, 2016).

NLRP3‐dependent secretion of IL‐1β as well as TNF‐α are established mediators of inflammation and associated with liver injury (Szabo & Petrasek, 2015). Induction of chemokines amplifies the inflammatory response and involves C‐X‐C Motif Chemokine Ligand (CXCL) 1, CXCL2 and C‐C Motif Chemokine Ligand 2 (CCL2) to recruit an array of hematopoietic cells (Heymann & Tacke, 2016; Koyama & Brenner, 2017). During the development of NAFLD, C‐X‐C Motif Chemokine Receptor 2 (CXCR2), the key receptor of the ligands CXCL1 and CXCL2, was shown to be upregulated via neutrophil‐derived Lipocalin 2 (LCN2; Syn: neutrophil gelatinase‐associated lipocalin; Ye et al, 2016; Moschen et al, 2017). P‐selectin glycoprotein ligand‐1 (PSGL‐1) and FcγRIII/II mediate neutrophil aggregation and activation (Guyer et al, 1996; Wang & Jönsson, 2019). Chitinase‐3‐like proteins expressed in activated neutrophils and macrophages contribute to the release of CXCL2 and CCL2 to further promote and guide leukocyte recruitment to sites of inflammation (Kzhyshkowska et al, 2016; Zhao et al, 2020). The hepatocyte‐derived acute phase protein Serum Amyloid A (SAA) is secreted in response to hepatic and systemic inflammation and can amplify signaling via cathepsin B and P2X7‐Receptor‐mediated activation of the NLRP3 inflammasome (Migita et al, 2014; De Buck et al, 2016). Defining master regulators of these recruitment processes has the potential to block fibrosis at an early stage of development.

The extent to which neutrophil‐specific activation of NLRP3 can modulate diverse aspects of the hepatic inflammatory response remains unclear. In this study, we investigate how NLRP3 activity in neutrophils controls intracellular inflammatory pathways in other hepatic cells, promotes hematopoietic cell infiltration, and contributes to liver injury.

Results

Neutrophil‐specific mutant NLRP3 mice show impaired growth, weight and survival

To explore the consequences of neutrophil‐specific NLRP3 overactivation on liver pathobiology, Nlrp3 D301NneoR and Nlrp3 A350VneoR conditional Nlrp3 mutant mice were used as previously described (Brydges et al, 2009; Bonar et al, 2012) and bred with S100a8‐Cre mice to generate heterozygous Nlrp3 D301N‐MRP8 ‐mutant mice and Nlrp3 A350V‐MRP8 mutant mice (Fig 1A). Cre‐recombinase‐negative mice (WT) and Cre‐recombinase‐positive mice (Nlrp3 D301N‐MRP8 or Nlrp3 A350V‐MRP8) of the same litters were used for the analysis of each group. Mice from both mutant models showed an impaired growth and weight profile compared with littermate controls (Nlrp3 D301NneoR: 0.6‐fold; P < 0.001, Nlrp3 A350VneoR: 0.5‐fold; P < 0.001; Fig 1B and C). Lifespan of S100a8‐Cre‐positive mice of both mutant models Nlrp3 D301N‐MRP8 and Nlrp3 A350V‐MRP8 was impaired reaching a maximum of 30 days in Nlrp3 D301N‐MRP8 and 12 days in Nlrp3 A350V‐MRP8 mutant mice (Fig 1D). Furthermore, S100a8‐Cre‐positive mutant mice showed similar survival curves compared with lysozyme‐Cre‐positive mutant mice, which drive expression in most myeloid cells including neutrophils (Fig 1D). Liver weight/body weight ratio was increased in Nlrp3 A350V‐MRP8 mutant mice compared with WT (1.15‐fold; P < 0.01; Fig EV1). Gasdermin D (GSDMD) has been shown to be crucial in macrophage and neutrophil‐driven NLRP3 Inflammasome‐related diseases. Hence, we generated a neutrophil‐specific Nlrp3 A350V/+_Gsdmd_−/− mouse model. Nlrp3 A350‐MRP8 _Gsdmd_−/− mice show a normal survival reaching adulthood, normal growth, and no signs of inflammation similar to healthy control mice (Fig 1E). Taken together, these mutant models reveal that neutrophil‐specific NLRP3 activation phenocopies systemic NLRP3 activation, with a similar spectrum of severity between Nlrp3 D301N and Nlrp3 A350V models as previously reported (Brydges et al, 2009; Bonar et al, 2012).

Figure 1

Figure 1

The alternative text for this image may have been generated using AI.

Full size image

Neutrophil‐specific mutant NLRP3 mice show impaired growth, weight, and survival

A. Schematic illustration of generation of both mutant neutrophil‐specific Nlrp3 D301NneoR and Nlrp3 A350VneoR mice.

B. Representative images of both Nlrp3 _D301N‐MRP8_and Nlrp3 A350V‐MRP8 mutant mice.

C. Body weight of both Nlrp3 _D301N‐MRP8_and Nlrp3 A350V‐MRP8 mutant mice at day 10 of age. n = 6–8.

D. Survival curve of both neutrophil‐ and myeloid‐specific Nlrp3 D301NneoR and Nlrp3 A350VneoR mutant mice. n ≥ 18.

E. Weight and survival curves of neutrophil‐specific Nlrp3 A350VneoR _Gsdmd_−/− mice. n ≥ 18.

Data information: Significance determined by unpaired _t_‐test (***P < 0.001), data are expressed as mean ± SD, and dots represent individual biological replicates.

Figure 2

Figure 2

The alternative text for this image may have been generated using AI.

Full size image

Neutrophil‐specific NLRP3 activation induces systemic changes in hematopoiesis and accumulation of inflammatory neutrophils in the liver

A. Flow cytometry gating strategy used to identify mature leukocytes and HSPCs in the bone marrow, peripheral blood, and liver.

B. Frequency of LSKs, MPs, and leukocytes in bone marrow and peripheral blood of live CD45+ cells from WT (n = 6) and Nlrp3 D301N‐MRP8 (n = 6) neonatal mice at P8‐P9.

C. Frequency of LSKs, MPs, and leukocytes in liver of live CD45+ cells from WT (n = 6) and Nlrp3 D301N‐MRP8 (n = 6) neonatal mice at P8‐P9.

D. Markers of neutrophil activation FcγRIII/II and PSGL‐1 illustrated by geometric mean fluorescence intensity (gMFI) in neutrophils from the bone marrow (BM), blood (BLD), and liver (LIV) of WT (n = 3) and Nlrp3 D301N‐MRP8 (n = 4) neonatal mice at P8‐P9.

E. Loss of Ly6G expression in Nlrp3 D301N‐MRP8 neutrophils (n = 4) relative to WT littermate neutrophils (n = 3) by flow cytometry evaluation of gMFI.

Data information: Significance determined by unpaired _t_‐test (*P < 0.05, **P < 0.005, ***P < 0.0005, ****P < 0.0001), data are expressed as mean ± SD, and dots represent individual biological replicates. CMP, common myeloid progenitor; HSPCs, hematopoietic stem and progenitor cells; LSKs, lineage− ckit+ Sca‐1+; LT/ST‐HSC, long‐term/short‐term hematopoietic stem cell; Mac/DC, macrophage/dendritic cell; Mono, monocyte; MonP, monocyte progenitor; MPP, multipotent progenitor; MPs, myeloid progenitors; NeP, neutrophil progenitor; Neutro, neutrophil.

Neutrophil‐specific NLRP3 activation induces systemic changes in hematopoiesis and accumulation of inflammatory neutrophils in the liver

To investigate the effects of neutrophil‐specific NLRP3 activation on hematopoiesis, and the specific immune cell types recruited to the liver by NLRP3 inflammasome activation in neutrophils, we investigated the changes in the neutrophil lineage, in other myeloid cells, and in hematopoietic stem and progenitor cells in the liver tissue, blood, and bone marrow of Nlrp3 D301N‐MRP8 and Nlrp3 A350V‐MRP8 mice by flow cytometry (Fig 2A). Alterations in myelopoiesis were evident in the bone marrow of Nlrp3 D301N‐MRP8 mice at 8–9 days of life (P8‐P9) when they exhibited the full inflammatory phenotype, with elevated numbers of MPP and MonP, and a reduction in CMP. These changes were accompanied by accelerated differentiation into the monocyte and neutrophil lineages (Fig 2B). In the blood, increased numbers of MonP and mature monocytes were evident (Fig 2B). The hematopoietic compartments in the liver of Nlrp3 D301N‐MRP8 mice displayed similar patterns with increased populations of LT‐HSC and MPP in the lineage− ckit+ Sca‐1+ (LSKs) fraction, reduced CMP, and elevated MonP (Fig 2C). In the neutrophil lineage, NeP was decreased in the liver whereas total numbers of terminally differentiated neutrophils varied widely in the liver (Fig 2C). These data suggest increased myeloid cell‐lineage differentiation accompanied by mobilization of myeloid cell precursors into the peripheral blood. Elevated CD162 (PSGL‐1) was evident in Nlrp3 D301N‐MRP8 neutrophils and was accompanied by increased expression of CD16/32 (FcγRIII/II), particularly in the liver, suggesting neutrophil aggregation and activation (Fig 2D). These changes in cells of the neutrophil lineage were surprisingly accompanied by a near absence of mature neutrophils (Ly6G‐high), a marker for mature neutrophils (Sturge et al, 2015) in the bone marrow, blood, and liver of Nlrp3 D301N‐MRP8 mice (Fig 2E). To investigate the myeloid lineage of the knock‐in mice at baseline before prolonged exposure to the environment outside the womb, we studied the more severe model of Nlrp3 activation, Nlrp3 A350V‐MRP8 mice, immediately prior to birth at P0 and identified elevated LSKs and MonP in both bone marrow and blood, and elevated neutrophils in the peripheral blood (Fig EV2A). Consistent with data from the older Nlrp3 D301N‐MRP8 mice, neutrophils from Nlrp3 A350V‐MRP8 mice also displayed a near absence of Ly6G expression in the bone marrow and blood (Fig EV2B). These data indicate major changes in hematopoiesis of neonatal mice with neutrophil‐restricted expression of mutant NLRP3 evident from birth and continuing in neonatal life. The lack of mature neutrophils and expansion of immature neutrophils may be a result of emergency granulopoiesis in these neonates. The changes in the liver, blood, and bone marrow including an expansion of monocyte progenitors may be a consequence of altered activation status and lifespan of cells in the neutrophil lineage that accompany neutrophil‐specific NLRP3 inflammasome activation.

Figure 3

Figure 3

The alternative text for this image may have been generated using AI.

Full size image

Neutrophil‐specific NLRP3 activation leads to myeloid cell accumulation and elevated cell death in the liver

A, B MPO staining of liver issue used as a marker for myeloid cells including monocytes in both Nlrp3 D301N‐MRP8 and Nlrp3 A350V‐MRP8 mutant (n ≥ 3 mice per genotype, representative images shown, bar indicates 250 μM). H&E staining of liver tissue with histologic scoring of liver inflammation (bar indicates 100 μM, representative images are shown). TUNEL staining of liver tissue (Representative images shown, bar indicates 250 μM).

C Expression of mRNA level in whole liver tissue of MPO, MMP8, MMP9 LCN2, Chitinase‐3‐like proteins CHIL1 and CHIL3.

Data information: n ≥ 6 mice per genotype, significance determined by unpaired _t_‐test (*P < 0.05; **P < 0.01; ***P < 0.001), data are expressed as mean ± SD, and dots represent individual biological replicates.

Neutrophil‐specific NLRP3 activation leads to accumulation of neutrophils and elevated cell death in the liver

To understand the organ‐specific effects of neutrophil NLRP3 inflammasome activation, we studied the inflammatory and profibrotic changes developing in the livers of Nlrp3 D301N‐MRP8 and Nlrp3 A350V‐MRP8 mutant mice. Livers were collected 8–10 days after birth, and hepatic infiltration of neutrophils was assessed by immunostaining. Myeloperoxidase (MPO)‐positive immunohistochemical staining, a marker for myeloid cells including monocytes, revealed a significant increase of myeloid cell infiltration in the liver of both Nlrp3 D301N‐MRP8 and Nlrp3 A350V‐MRP8 mutant mice models consistent with the flow cytometry data (Nlrp D301NneoR: 1.4‐fold; P < 0.05, Nlrp3 A350VneoR: twofold; P < 0.01; Fig 3A and B). To further study the effects of NLRP3 overactivation in neutrophils, we focused on Nlrp3 A350V‐MRP8 mutant mice, a model system with severe inflammatory disease. Histopathology of H&E sections revealed severe liver inflammation in Nlrp3 A350V‐MRP8 mice that was not observed in WT mice. Hepatic inflammation including large neutrophil foci was disseminated in the whole liver from Nlrp3 A350V‐MRP8 mice (Fig 3B). Cell death was significantly increased in the livers of Nlrp3 A350V‐MRP8 mice, as assessed by TUNEL staining (twofold; P < 0.05; Fig 3B). We have previously shown using Nlrp3 knock‐in models that NLRP3‐triggered cell death is likely a consequence of pyroptosis, a Casp1‐dependent form of cell death (Wree et al, 2014a). Gene expression of MPO was significantly higher in the livers of Nlrp3 A350V‐MRP8 mice (2.5‐fold; P < 0.01; Fig 3C), as was expression of the neutrophil‐specific metalloproteinase 8 (MMP8) and metalloproteinase 9 (MMP9; eightfold; P < 0.001, 1.8‐fold; P < 0.05; Fig 3C). LCN2 is mainly expressed in neutrophils and promotes the secretion of pro‐inflammatory cytokines. Analysis of LCN2 gene expression demonstrated a significant upregulation in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (22.7‐fold; P < 0.001; Fig 3C). Chitinase‐3‐like proteins are highly expressed in both activated neutrophils and macrophages and promote the secretion of the chemoattractants CXCL2 and CCL2 (Kzhyshkowska et al, 2016; Zhao et al, 2020). Both Chitinase‐3‐like protein 1 (CHIL1) and Chitinase‐3‐like protein 3 (CHIL3) gene expression were upregulated in the liver of Nlrp3 A350V‐MRP8 mutant mice (5.4‐fold; P < 0.001, 3.7‐fold; P < 0.001; Fig 3C).

Neutrophil‐specific NLRP3 activation leads to increased cytokine production and acute phase proteins in the liver

The pathology of NLRP3‐driven inflammation is mediated in part via the actions of IL‐1 family members including IL‐1β. To study the effects of neutrophil‐specific NLRP3 activation, we examined mRNA and protein levels of NLRP3 inflammasome components and downstream mediators. The levels of NLRP3 and IL‐1β mRNA were upregulated in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (1.9‐fold; P < 0.05, twofold; P < 0.05, respectively; Fig 4A); however, similar levels of Casp1 mRNA were present in the livers of both groups. After activation of the NLRP3 inflammasome, or in the presence of a mutant active NLRP3 inflammasome, Casp1 and IL‐1β are proteolytically processed to become biologically active, although exceptions have been reported for Casp1 (Szabo & Petrasek, 2015; Alegre et al, 2017). The posttranslational protein expression of cleaved and mature forms of Casp1 (p20) and IL‐1β was significantly higher in the Nlrp3 A350V‐MRP8 mutant mice (1.4‐fold; P < 0.05, 3.9‐fold; P < 0.001; Fig 4B and C). Pro IL‐1β expression did not show a significant difference in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (Fig 4B and C). TNF‐α, a key pro‐inflammatory cytokine, was previously shown to have an essential role in liver inflammation and fibrosis downstream of Nlrp3 activation (Wree et al, 2013). In Nlrp3 A350V‐MRP8 mice, the mRNA and protein levels of TNF‐α were elevated compared with littermate controls (2.2‐fold; P < 0.05, 2.2‐fold; P < 0.05; Fig 4A–C). SAA is an acute phase protein that is mainly released in the liver by hepatocytes in high concentrations during inflammation. SAA 1 and 3 mRNA were upregulated 300‐fold in Nlrp3 A350V‐MRP8 mice compared with littermate controls (P < 0.001; Fig 4D). The chemoattractants CXCL1, CXCL2, and CCL2 have a key role in orchestrating the interplay and recruitment of monocytes, macrophages, and neutrophils during inflammation. Expression of mRNA of CXCL1, CXCL2,its receptor CXCR2, and CCL2 were upregulated in Nlrp3 A350V‐MRP8 mutant mice (22.9‐fold; P < 0.05, 8.2‐fold; P < 0.01, 3.6‐fold; P < 0.001, 2.5‐fold; P < 0.05; Fig 4D).

Figure 4

Figure 4

The alternative text for this image may have been generated using AI.

Full size image

Neutrophil‐specific NLRP3 activation leads to increased cytokine production and acute phase proteins in the liver

A Expression of NLRP3, IL‐1β, and TNF‐α on mRNA level in whole liver lysate.

B, C Protein expression in liver tissue of cleaved and mature forms of Casp1 (p20), IL‐1β and TNF‐α (17 kDa; n = 3 mice per genotype for Western blots).

D Expression of mRNA of SAA 1 and 3, the chemoattractants CXCL1, CXCL2, its receptor CXCR2, and CCL2.

Data information: n ≥ 6 mice per genotype, significance determined by unpaired _t_‐test (*P < 0.05; **P < 0.01; ***P < 0.001), data are expressed as mean ± SD, and dots represent individual biological replicates.

NLRP3 activation in neutrophils alters macrophages in the liver

To understand how neutrophils alter resident liver macrophages, we examined phenotypic and functional markers on macrophages. Immunohistochemical staining of F4/80, a marker of macrophages, revealed reduced expression in liver tissue of Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (Fig 5A). These findings were in line with immunofluorescence staining of C‐Type Lectin Domain Family 4 Member F (CLEC4F), a specific surface marker of Kupffer cells, that showed reduced CLEC4F expression in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (Fig 5B). Analysis of mRNA expression of both F4/80 and CLEC4F also showed decreased expression of both genes in Nlrp3 A350V‐MRP8 mutant mice (0.6‐fold; P < 0.001, 0.7‐fold; P < 0.001; Fig 5A and B). In contrast, CD11b, a surface protein that is present on neutrophils, monocytes, and macrophages, was significantly upregulated in Nlrp3 A350V‐MRP8 mutant mice (2,7‐fold; P < 0.001; Fig 5D). Ly6C, a cell surface marker of bone marrow‐derived monocytes and neutrophils was significantly increased in Nlrp3 A350V‐MRP8 mutant mice (Fig 5C). Analysis of pro‐inflammatory myeloid markers including inducible nitric oxide synthase (iNOS) showed an increase in gene expression in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (31.0‐fold; P < 0.001; Fig 5D). Together, these data reveal reduced numbers of Kupffer cells in Nlrp3 A350V‐MRP8 mutant mice, together with evidence of pro‐inflammatory signaling in infiltrating myeloid cells initiated by changes in NLRP3 inflammasome activation in neutrophils.

Figure 5

Figure 5

The alternative text for this image may have been generated using AI.

Full size image

NLRP3 activation in neutrophils alters macrophage population in the liver

A. Immunohistochemical staining of F4/80, a marker for resident macrophages, and mRNA expression of F4/80 in liver tissue.

B. Immunofluorescence staining of CLEC4F, a specific marker of Kupffer cells, and gene expression of CLEC4F in liver tissue.

C. Immunohistochemical staining of Ly6C, a marker for pro‐inflammatory macrophages, in liver tissue.

D. Gene expression of the pro‐inflammatory marker of macrophages iNOS as well as gene expression of CD11b.

Data information: Representative images shown, bar indicates 250 μM for F4/80 and Ly6C, bar indicates 100 μM for CLEC4F, n ≥ 6 mice per genotype, significance determined by unpaired _t_‐test (**P < 0.01; ***P < 0.001), data are expressed as mean ± SD, and dots represent individual biological replicates.

Neutrophil‐specific NLRP3 activation impairs collagen deposition and drives liver fibrosis

NLRP3 has an essential role in liver inflammation and fibrosis as previously shown (Wree et al, 2013, 2014a, 2014b), but little is known about the relative role of neutrophils in this process. Furthermore, little is known about its role in collagen deposition during liver development. Here, the role of neutrophil‐specific NLRP3 overactivation on liver development as well as fibrosis was investigated. Sirius red staining revealed a lower overall expression of collagen in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (0.5‐fold; P < 0.05; Fig 6A). Gene expression of collagen 1a and collagen 3a analyzed by qPCR in whole liver lysates was decreased in Nlrp3 A350V‐MRP8 mutant mice (0.1‐fold; P < 0.001, 0.1‐fold; P < 0.05, respectively; Fig 6B). In addition, protein expression of collagen 1a was significantly lower in Nlrp3 A350V‐MRP8 mutant mice (0.3‐fold; P < 0.05; Fig 6C). Notably, WT mice displayed normal collagen deposition of the vessels and bile ducts, and no signs of fibrosis. In contrast, Nlrp3 A350V‐MRP8 mutant mice displayed a diminished and irregular pattern of collagen around vessels and bile ducts and an absence of an adequate vascular and bile duct architecture. Importantly, several foci of dense irregular collagen deposition and fibrosis, corresponding to areas with severe liver inflammation, were detected in Nlrp3 A350V‐MRP8 mice (Fig 6D). While neonatal livers of WT mice show normal organ development including regular collagen deposition, Nlrp3 D301N‐MRP8 mice fail to develop a regular liver architecture. Instead, after birth Nlrp3 D301N‐MRP8 mice start to develop fibrotic foci that progress over time (Fig 6E). The expression of the profibrotic connective tissue growth factor (CTGF) and tissue inhibitor of metalloproteinases 1 (TIMP1) was increased in Nlrp3 A350V‐MRP8 mutant mice compared with WT mice (3.8‐fold; P < 0.001, 73‐fold; P < 0.001, respectively; Fig 6F). At the same time, metalloproteinases 10 (MMP10) and metalloproteinases 13 (MMP13), which are inhibited by TIMPs, showed a decrease in gene expression in Nlrp3 A350V‐MRP8 mice compared with WT mice (0.2‐fold; P < 0.001, 0.2‐fold; P < 0.001, respectively; Fig 6F). These data reveal a liver fibrotic program driven exclusively by NLRP3 inflammasome activation in neutrophils.

Figure 6

Figure 6

The alternative text for this image may have been generated using AI.

Full size image

Neutrophil‐specific NLRP3 activation impairs collagen deposition and drives liver fibrosis

A. Collagen deposition assessed by Sirius red staining.

B. Gene expression of collagen 1a and collagen 3a in whole liver lysate analyzed by qPCR.

C. Protein expression of collagen 1a in whole liver lysate. n = 3.

D. Collagen deposition and fibrosis assessed by Sirius red staining in WT and Nlrp3 A350V‐MRP8 mutant mice to illustrate vascular as well as bile duct collagen deposition and foci of fibrosis (black arrows) in areas with severe liver inflammation.

E. Fibrotic foci in Nlrp3 D301N‐MRP8 mice over time. n ≥ 3.

F. Gene expression in liver tissue of the profibrotic genes CTGF and TIMP1 as well as the metalloproteinases 10 and 13.

Data information: Representative images shown, bar indicates 250 μM, n ≥ 6 per genotype, significance determined by unpaired _t_‐test (*P < 0.05; **P < 0.01; ***P < 0.001), data are expressed as mean ± SD, and dots represent individual biological replicates.

Discussion

This study provides new insights into the role of neutrophil NLRP3 function in liver inflammation, liver development, and collagen turnover. The results of this study reveal that neutrophil‐specific activation of NLRP3 can independently initiate liver inflammation, impair liver development, and promote liver fibrosis. Neutrophils trigger pro‐inflammatory signaling and a chemokine profile that mediate hepatic monocyte recruitment, and polarization of infiltrating myeloid cells to a pro‐inflammatory phenotype.

For numerous inflammatory‐driven liver diseases, including alcoholic and nonalcoholic steatohepatitis, liver fibrosis, and ischemia‐reperfusion injury, a central role of the NLRP3 inflammasome and neutrophils has been described (Wree et al, 2014a, 2014b, 2018; Mridha et al, 2017; de Oliveira et al, 2018; Neumann et al, 2018; Knorr et al, 2020). Emerging evidence points to leukocytes as the key cells for NLRP3‐dependent liver inflammation. Mice expressing mutant NLRP3 protein in myeloid lineage cells phenocopy the liver pathobiology seen in mice expressing mutant NLRP3 globally (Wree et al, 2014a). Liver inflammation evoked by NLRP3 inflammasome activation is characterized by a significant infiltration of neutrophils (Wree et al, 2014a, 2018). Neutrophils are the largest fraction of circulating leukocytes and are known to initiate liver inflammation (de Oliveira et al, 2018). Activation of NLRP3 in neutrophils was recently shown to trigger autoinflammatory disease in mice and to be a substantial source of IL‐1β (Stackowicz et al, 2021). Consistent with this finding, a crucial role for neutrophils in NLRC4 inflammasome‐dependent autoinflammatory disease has also been shown (Kitamura et al, 2014) and neutrophil‐specific activation using the same Cre but driving NLRC4 inflammasome activation resulted in systemic autoinflammation (Nichols et al, 2017). To investigate the role of neutrophils in the pathogenesis of liver inflammation triggered by the NLRP3 inflammasome, this study was conducted using 2 neutrophil‐specific Nlrp3 conditional gain of function mutant mouse models. We demonstrated that activation of NLRP3 in neutrophils alters hematopoiesis in the BM and neonatal liver accompanied by emergency myelopoiesis in the blood and leads to systemic ablation of mature neutrophils and enrichment of immature neutrophils. These enriched immature neutrophils display increased levels of activation markers and can independently promote NLRP3‐dependent liver inflammation. Histological analysis of liver sections from Nlrp3 D301N‐MRP8 mice revealed the presence of neutrophil foci consistent with rapid recruitment of neutrophils to the liver of neonatal mice, and a pathological role for these cells in liver inflammation. Liver inflammation was characterized by an increase in the pro‐inflammatory NLRP3‐dependent cytokine IL‐1β. This elevated IL‐1β was associated with increased secretion of TNF‐α. Both of these cytokines are known to mediate and promote liver inflammation. These data are also consistent with pharmacological blockade and knock‐out mouse models of IL‐1β and TNF‐α that have shown to alleviate or even revert NLRP3‐driven liver inflammation and fibrosis (Petrasek et al, 2012; Wree et al, 2014a; McGeough et al, 2017).

Altered recruitment, activation, and differentiation of recruited leukocytes can promote liver diseases such as NAFLD and ALD (Heymann & Tacke, 2016). In this context, the interplay between neutrophils and macrophages is crucial. Numerous cytokines mediate the recruitment as well as the local interactions between leukocytes. CXCL1, CXCL2, and CCL2 are vital chemoattractants for neutrophils and monocytes (Heymann & Tacke, 2016). We demonstrate that activation of NLRP3 in neutrophils increases the expression of these chemoattractants in the liver. In addition, CXCR2, the principal receptor of the two ligands CXCL1 and CXCL2, was upregulated in the liver. LCN2 modulates neutrophil recruitment, and its expression has been reported in neutrophils and in hepatocytes during acute inflammation (Schroll et al, 2012; Li et al, 2016; Wieser et al, 2016; Moschen et al, 2017). Emerging evidence supports a crucial role of LCN2 in liver diseases such as NAFLD, ALD, and liver fibrosis (Alwahsh et al, 2014; Wieser et al, 2016; Chen et al, 2020). In ALD, neutrophil‐derived LCN2 was shown to be an important determinant for hepatic neutrophil infiltration (Wieser et al, 2016). LCN2 mediates the recruitment and crosstalk between neutrophils and macrophages by inducing CXCR2 in NASH (Ye et al, 2016). Our results reveal an upregulation of LCN2 and CXCR2 in mutant mice, supporting the current data that highlight the interplay of neutrophils and macrophages via LCN2‐induced CXCR2.

SAA, an acute phase protein, is mainly secreted by hepatocytes but may form a feed‐forward loop in the context of lethal autoinflammation and liver disease. SAA is primarily upregulated by IL‐1β, IL‐6, and TNF‐α, but can also induce the production of numerous cytokines including IL‐1β and TNF‐α. Furthermore, SAA mediates the expression of numerous chemokines like CXCLs as well as CCL2 and is a chemoattractant itself for neutrophils and monocytes (Sack, 2018; De Buck et al, 2016). SAA activates the NLRP3 inflammasome, thereby further increasing IL‐1β expression (Ather et al, 2011; Niemi et al, 2011). In particular, SAA was shown to have this effect in neutrophils (Migita et al, 2014). In this study, we demonstrate a drastic increase of SAA in liver tissue that is initially induced by NLRP3‐mediated cytokine release. We therefore hypothesize that the secretion of SAA enhances NLRP3‐mediated inflammation by acting in a feed‐forward loop.

MMP8 and MMP9 are released primarily by neutrophils as well as monocytes. Both MMP8 and MMP9 have a chemotactic effect on neutrophils via the proline–glycine–proline tripeptide that binds to the CXCR2 (Lin et al, 2008; Koelink et al, 2014). MMP9 has been shown to promote the infiltration of neutrophils into the liver (de Oliveira et al, 2018). In line with these findings, we report an upregulation of MMP8 and MMP9 in Cre_MRP8_ mice that contributes to the observed recruitment of neutrophils in the liver. Chitinase‐3‐like protein 1 and Chitinase‐3‐like protein 3 (only present in mouse/rat) are expressed in neutrophils as well as macrophages and are induced by various factors and cytokines including IL‐1β, IL‐18, and TNF‐α (Kzhyshkowska et al, 2016; Xie et al, 2020; Zhao et al, 2020). By promoting the production of CCL2, CXCL2, and MMP9, Chitinase‐3‐like protein 1 contributes to the recruitment of leukocytes at the site of inflammation (Kzhyshkowska et al, 2016; Zhao et al, 2020). Progression of liver fibrosis in patients with NASH correlated with increased expression of Chitinase‐3‐like protein 1 (Kumagai et al, 2016). Mice deficient for Chitinase‐3‐like protein 1 showed ameliorated liver fibrosis (Higashiyama et al, 2019). The findings of our study also reveal an upregulation of Chitinase‐3‐like proteins in mutant mice and support a pro‐inflammatory role of Chitinase‐3‐like proteins. Taken together, these results demonstrate the central role of neutrophils in orchestrating liver inflammation via pro‐inflammatory signaling and chemokine release.

Monocytes adapt their phenotypic profile to environmental stimuli and modulate inflammatory responses in the liver (Heymann & Tacke, 2016; Koyama & Brenner, 2017; Krenkel & Tacke, 2017). Pro‐inflammatory macrophages are important in the initial phase of liver inflammation and are characterized by high expression of Ly6C and iNOS, whereas the surface protein CD11b is generally prevalent in monocytes and macrophages. Reduced infiltration of pro‐inflammatory macrophages in NLRP3‐deficient mice was reported in a model of NASH (Wree et al, 2018). Resident liver macrophages are characterized by high expression of F4/80 and CLEC4F (Yang et al, 2013; Tacke & Zimmermann, 2014; Guillot & Tacke, 2019; Krenkel et al, 2020). Our data demonstrate an increase in Ly6C and iNOS expression in mutant mice revealing a pro‐inflammatory infiltration of recruited macrophages. Immunophenotyping of hematopoietic cells in the bone marrow and blood highlighted an expanded population of monocyte progenitors and monocytes consistent with an expanded profile of infiltrating macrophages in the liver. This expanded monocytopoiesis in bone marrow, blood, and liver suggested that monocytes and macrophages may be interacting with activated neutrophils to promote liver fibrosis or potentially to counteract neutrophil‐driven inflammatory disease. Increased CD11b expression points to a higher overall population of myeloid‐derived cells in the liver in these mice. In contrast, Kupffer cells expressing F4/80 and CLEC4F were decreased. These findings suggest that neutrophil‐specific NLRP3 activation promotes an increase in macrophage infiltration while reducing resident macrophages.

Chronic inflammation in the liver often progresses to liver fibrosis (Koyama & Brenner, 2017). NLRP3 has a key role in the development and progression of liver inflammation. Mice overexpressing NLRP3 in myeloid‐derived cells develop liver fibrosis at an early age and show increased expression of profibrotic genes (Wree et al, 2014a, 2018). At the same time, genetic depletion of Nlrp3 or pharmaceutical blockade of NLRP3 ameliorates liver fibrosis in mouse models of NASH (Wree et al, 2014b; Mridha et al, 2017). The effect of NLRP3‐driven sterile inflammation on organ development is unknown. In this study, we demonstrate that the expression of mutant NLRP3 in neutrophils leads to impaired liver development at an early age. Livers of mutant mice had a significant decrease in regular collagen deposition and an altered architecture. In particular, vascular and bile duct structure lacked adequate collagen deposition. In contrast to these findings, modest liver fibrosis was observed in the mutant mice but not in WT. In addition, a profibrotic gene profile with increased expression of CTGF and TIMP1 was observed in these mice compared with WT mice. These results give new cell‐specific insights in the profibrotic role of NLRP3 in liver fibrosis. It also shows the role of NLRP3‐driven inflammation on liver development at an early age.

Taken together, our study provides new insight to the role of neutrophil‐ and NLRP3‐driven liver inflammation and early‐stage fibrosis (Fig EV3). These findings have important implications for the understanding of sterile liver inflammation and provide novel therapeutic avenues in liver fibrosis via targeting the NLRP3 inflammasome or potentially neutrophil depletion strategies.

Materials and Methods

Mice

Nlrp3 knock‐in mouse strains with an aspartate 301 to asparagine (D301N) or an alanine 350 to valine (A350V) substitution were generated as previously described (Brydges et al, 2009; Bonar et al, 2012). The D301N and A350V point mutation leads to a conformational change causing a ligand‐independent constitutive activation of the mutant NLRP3 inflammasome. Briefly, the targeting construct pPNTlox2PNLRP3 D301N or pPNTlox2PNLRP3 A350V were created by cloning 4–7 kb regions directly upstream and downstream of a targeted position in intron 2 of Nlrp3 around the neoR antibiotic resistance cassette in plasmid pPNTlox2P. Due to the presence of an intronic floxed neomycin resistance cassette, the expression of the mutation does not occur unless the Nlrp3 knock‐in mice (Nlrp3 D301NneoR, Nlrp3 A350VneoR) are first bred with mice expressing Cre recombinase. Floxed mice were bred to mice expressing Cre recombinase under control of Calcium‐binding protein A8 (MRP8; Cre_MRP8_, constitutive activation of the mutated NLRP3 inflammasome in neutrophils) to generate heterozygous gain of function mutant Nlrp3 _D301N_‐MRP8 and Nlrp3 A350V_‐_MRP8, respectively (Fig 1A). In addition, floxed mice were bred to mice expressing Cre recombinase under control of lysozyme (CreLyz, constitutive activation of the mutated NLRP3 inflammasome in myeloid lineage) to generate heterozygous gain of function mutant Nlrp3 D301N_‐_LysM and Nlrp3 A350V_‐_LysM, respectively. Furthermore, we generated a neutrophil‐specific Nlrp3 A350V/+_Gsdmd_−/− mouse model. Mouse strain B6.Cg‐Tg (S100A8‐cre,‐EGFP) 1Ilw/J and B6.129P2‐Lyz2tm1(cre)Ifo/J were purchased from Jackson Laboratories. Littermates lacking the Cre recombinase were used as control mice. Mice were housed in a temperature and light cycle‐controlled room. The experimental protocols were approved by the Institutional Animal Care and Use Committee at the University of California, San Diego.

Liver fixation and storage

NLRP3 knock‐in mice were sacrificed at day 8–10 after birth. Mice were anesthetized, and liver tissue was either: (i) fixed in 10% formalin for 24 h and embedded in paraffin; (ii) embedded in O.C.T. for frozen tissue sections; or 50 μg samples were placed in 0.5 ml of RNAlater® Solution (Lifetechnologies, Carlsbad, CA, USA) and frozen at −80°C.

Liver histology, immunostaining, and Sirius Red

Paraffin‐embedded liver tissues were stained for hematoxylin and eosin (H&E). Inflammation was quantified according to the NAFLD activity score (NAS) by an experienced pathologist (Kleiner et al, 2005). For the assessment of liver fibrosis, Sirius Red was performed. Liver sections were incubated for 1 h at room temperature with an aqueous solution of saturated picric acid containing 0.1% Direct Red. Immunohistochemistry staining for myeloperoxidase (Myeloperoxidase, RB‐372‐1, Thermo Scientific, Waltham, MA, USA), lymphocyte antigen 6 complex, locus C1 (Ly6C, Abcam, ab 15627, Cambridge, MA, USA) and F4/80 (F4/80, BioLegend, 123102, San Diego, CA, USA) was performed on formalin‐fixed, paraffin‐embedded liver sections. After specimens were deparaffinized and hydrated in ethanol, the antigens were retrieved in citrate buffer pH 6.0 for 30 min at 95°C (MPO), applying Proteinase K at room temperature (Ready‐to‐use Proteinase K, Dako, Agilent Technologies, Santa Clara, CA, USA) followed by blocking with 3% BSA in TBS‐T for 1 h at room temperature (Ly6C) or treated with 2% BSA 1x Triton for 30 min at room temperature followed by blocking with 1% BSA in TBS‐T for 10 min at room temperature (F4/80). After overnight incubation with primary antibodies, respective secondary antibodies for 1 h were applied and developed with a streptavidin‐peroxidase complex, using 3,3‐diaminobenzidine tetrahydrochloride (DAB). Slides were counterstained with hematoxylin. Immunohistochemistry staining for Clec4F (R&D Systems, AF2784, Minneapolis, MN, USA) was performed on frozen liver sections. Following fixation in acetone for 20 min at room temperature, antigen retrieval with 1% BSA 0.3% Triton for 30 min at room temperature was performed. After overnight incubation with the primary antibody, Alexa‐Fluor 488‐conjugated secondary antibody (Invitrogen, MA, USA) for frozen section was applied for 1 h. Apoptosis was assessed by terminal deoxynucleotidyl transferase dUTP nick‐end labeling (TUNEL) assay according to the manufacturer's instructions (ApopTag® Peroxidase In Situ Apoptosis Detection Kit, Millipore, Billerica, MA, USA). Photographs of all sections were taken by NanoZoomer 2.0HT Slide Scanning System (Hamamatsu, Japan). The number of positive cells per field for TUNEL, MPO, and Ly6C staining was determined using ImageJ software (Version 1.52a, National Institute of Health, USA).

Flow cytometry

Bone marrow, blood, and liver were harvested from Nlrp3 D301N_‐_MRP8 mice at day 8–9 after birth. Bone marrow and blood were harvested from Nlrp3 A350V‐MRP8 mice at day 0 after birth. Bone marrow and blood were collected in ice‐cold D‐PBS (GIBCO) with 2 mM EDTA to prevent cation‐dependent cell‐cell adhesion. Neonatal liver was mechanically disrupted, filtered through a 70 μm strainer, and collected into ice‐cold D‐PBS with 2 mM EDTA. All samples were subjected to red blood cell lysis (1 l H2O, 7.79 g NH4Cl, 0.037 g disodium EDTA, 1 g NaHCO3, pH7.3). Cells were washed twice with FACS buffer (D‐PBS + 1% BSA + 0.1% sodium azide + 2 mM EDTA) and stained according to published methods (Zhu et al, 2018). Briefly, anti‐CD16/32 (APC‐R700) antibody was added to cells (1:400) for 15 min on ice to block the Fc receptors. Then, a surface marker staining antibody cocktail (CD45‐APC‐Cy7, TCRβ‐FITC, B220‐FITC, NK1.1‐FITC, CD127‐FITC, CD11c‐FITC, F4/80‐FITC, Siglec F‐PE‐Cy7, CD11b‐BUV737, CD117‐PE, Sca‐1‐PerCP‐Cy5.5, Ly6G‐BUV395, CD11a‐BV605, CD162‐BV510, CD150‐BV711, and CD48‐BV786) was added directly to the cells in 100 μL PBS. The cells were incubated with antibody at 4°C for 30 min and protected from light. The viability of cells was assessed using the LIVE/DEAD Fixable Blue Dead Cell Stain Kit (ThermoFisher). Cells were washed twice with 200 μl FACS buffer before acquisition on a LSR Fortessa flow cytometer (BD Biosciences). Data were analyzed using FlowJo (version 10.7).

Immunoblot analysis

For protein extraction, whole liver tissue was homogenized in radioimmunoprecipitation assay buffer (Cell Signaling, Danvers, MA, USA) together with cocktails of protease and phosphatase inhibitors (Sigma‐Aldrich, St. Louis, MO, USA). For immunoblot analysis, 30 μg of protein lysate was resolved on Any kD™ Criterion™ TGX™ Precast Gels (Biorad, Hercules, CA, USA), transferred to nitrocellulose membrane, and blocked with Intercept™ Blocking Buffer (LI‐COR, Lincoln, NE, USA) for 1 h before incubation with primary antibodies overnight at 4°C. Membranes were incubated with IRDye secondary antibody (LI‐COR, Lincoln, NE, USA) and protein bands visualized with the LI‐COR Imaging System (LI‐COR, Lincoln, NE, USA). Expression intensity was quantified by Image Studio Licor™ (LI‐COR, Lincoln, NE, USA). Protein load was verified with a β‐Actin antibody. Antibodies specific to IL‐1β (Abcam, Cambridge, UK, 1:500), caspase‐1 (Adipogen, AG‐20B‐0042‐C100, CA, USA, 1:500), TNF‐α (Cell Signaling, Danvers, MA, USA, 1;500), Typ1 Collagen (SouthernBiotech, 1310‐01, Birmingham, AL, USA,1:500), and β‐Actin (Millipore/Sigma, A5441, Burlington, MA, USA, 1:6,000) were used.

Real‐time PCR

RNA was isolated from liver tissue using the RNeasy Tissue Mini kit according to the manufacturer's instructions (Qiagen, Valencia, CA, USA). Complementary DNA was synthesized from 1 μg of total RNA using the qScript cDNA SuperMix according to the manufacturer's instructions (Beverly, MA, USA). Real‐time PCR quantification was performed using TaqMan™ Fast Advanced Master Mix (ThermoFisher Scientific, Vilnius Lithuania). Briefly, 20 μl of reaction mix contained cDNA, TaqMan™ Fast Advanced Master Mix, and respective primers. Primers were purchased from ThermoFisher Scientific (Waltham, MA, USA). Assay IDs of the primers are given in Table 1. QuantStudio Design Software (ThermoFisher Scientific, Waltham, MA, USA) was used for analyses.

Table 1 Assay ID of primers used for qPCR.

Full size table

Statistical analyses

Analyses were performed with GraphPad (version 8.4.2; GraphPad Software Inc., La Jolla, CA, USA). Unpaired _t_‐tests (two‐tailed) were used to determine the difference between WT and the mutant mice. The significance level was set at P < 0.05 for all comparisons (*P < 0.05; **P < 0.01; ***P < 0.001; unless otherwise stated). Unless otherwise stated, data are expressed as mean ± SD or as percentage for categorical variables.

Data availability

This study does not contain data deposition in a public database.

References

Download references

Acknowledgment

This work was funded by NIH grants R01 DK113592 to AEF, HMH, LB, R01 AA024206 to AEF. NIH grant RO1HL124209 to BAC, AHA CDA 20CDA35320302 to YPZ, German Research Foundation (DFG‐Grant KA 5089/1‐1) to BK.

Author information

Author notes

  1. These authors contributed equally to this work

Authors and Affiliations

  1. Department of Pediatrics, University of California San Diego, La Jolla, California, USA
    Benedikt Kaufmann, Aleksandra Leszczynska, Agustina Reca, Laela M Booshehri, Janset Onyuru, ZheHao Tan, Lori Broderick, Hal M Hoffman, Ben A Croker, Yanfang Peipei Zhu & Ariel E Feldstein
  2. Department of Surgery, TUM School of Medicine, Klinikum rechts der Isar, Technical, University of Munich, Munich, Germany
    Benedikt Kaufmann, Helmut Friess & Daniel Hartmann
  3. Department of Hepatology and Gastroenterology, Charité, Universitätsmedizin Berlin, Berlin, Germany
    Alexander Wree
  4. Department of Pathology, University of California San Diego, La Jolla, California, USA
    Bettina Papouchado

Authors

  1. Benedikt Kaufmann
  2. Aleksandra Leszczynska
  3. Agustina Reca
  4. Laela M Booshehri
  5. Janset Onyuru
  6. ZheHao Tan
  7. Alexander Wree
  8. Helmut Friess
  9. Daniel Hartmann
  10. Bettina Papouchado
  11. Lori Broderick
  12. Hal M Hoffman
  13. Ben A Croker
  14. Yanfang Peipei Zhu
  15. Ariel E Feldstein

Contributions

Benedikt Kaufmann: Conceptualization; resources; data curation; formal analysis; funding acquisition; investigation; methodology; writing – original draft; writing – review and editing. Aleksandra Leszczynska: Data curation; investigation. Agustina Reca: Data curation; investigation. Laela M Booshehri: Data curation; investigation. Janset Onyuru: Data curation; investigation. ZheHao Tan: Data curation; investigation. Alexander Wree: Supervision. Helmut Friess: Supervision. Daniel Hartmann: Supervision. Bettina Papouchado: Formal analysis; investigation. Lori Broderick: Resources; supervision; writing – review and editing. Hal M Hoffman: Resources; supervision; funding acquisition; writing – original draft; project administration; writing – review and editing. Ben A Croker: Resources; supervision; funding acquisition; writing – original draft; project administration; writing – review and editing. Yanfang Peipei Zhu: Conceptualization; resources; data curation; formal analysis; supervision; funding acquisition; investigation; writing – original draft; project administration; writing – review and editing. Ariel E Feldstein: Conceptualization; resources; supervision; funding acquisition; investigation; writing – original draft; project administration; writing – review and editing.

Corresponding author

Correspondence toAriel E Feldstein.

Ethics declarations

L.B. is a site PI for Novartis, Inc. H.H. is a consultant for Novartis. H.H. has research collaborations with Jecure, Inc., Zomagen, Inc., and Takeda, Inc. A.F. is consultant for Ventyx Bio, Inc, Novo Nordisk, and Inipharm. A.F. has research collaborations with Takeda, Inc.

Additional information

See also: KW Chen (November 2022), UC Frising et al (July 2022)

EMBO reports (2022) 23: e54446

Supplementary Information

Expanded View Figures PDF (download PDF )

Review Process File (download PDF )

PDF+ (download PDF )

EV1 (download PNG )

Figure

Liver weight/body weight ratio of Nlrp3 A350V‐MRP8 and Nlrp3 D301N‐MRP8 mutant mice. Data information: n ≥ 6 mice per genotype, significance determined by unpaired _t_‐test (**P < 0.01), data are expressed as mean ± SD, and dots represent individual biological replicates.

EV2 (download PNG )

Figure

A. Frequency of leukocytes and HSPC in bone marrow and blood of live CD45+ cells from WT (n = 4) and Nlrp3 A350V‐MRP8 (n = 2) neonatal mice at P0.

B. Loss of Ly6G expression in Nlrp3 A350V‐MRP8 neutrophils (n = 2) relative to WT littermate neutrophils (n = 4) at P0 by flow cytometry evaluation of gMFI.

Data information: Data are expressed as mean, and dots represent individual biological replicates. CMP, common myeloid progenitor; HSPCs, hematopoietic stem and progenitor cells; LSKs, lineage− ckit+ Sca‐1+; LT/ST‐HSC, long‐term/short‐term hematopoietic stem cell; Mac/DC, macrophage/dendritic cell; Mono, monocyte; MonP, monocyte progenitor; MPP, multipotent progenitor; MPs, myeloid progenitors; NeP, neutrophil progenitor; Neutro, neutrophil.

EV3 (download PNG )

Figure

The activation of the NLRP3 inflammasome in neutrophils leads to the secretion of active IL‐1β, IL‐18, and TNF‐α. These cytokines trigger a pro‐inflammatory cell response by promoting the secretion of additional cytokines such as CXCL1, CXCL2, and CCL2 that attract more neutrophils and BMDMs to the liver. Once recruited, BMDMs mature to pro‐inflammatory macrophages. Upregulation of LCN2 in neutrophils leads to an increase in CXCR2, which enhances leukocyte recruitment to the liver through CXCL1/CXCL2‐CXCR2 binding. In addition, increased CXCR2 expression enhances the secretion of cytokines. MMP8 and MMP9 have a chemotactic effect on neutrophils via the proline–glycine–proline tripeptide (PGP) that binds to CXCR2. Chitinase‐3‐like protein contributes to the recruitment of leukocytes to the site of inflammation by augmenting the secretion of cytokines such as CXCL2 and CCL2. Inflammation in the liver induces the secretion of the acute phase protein SAA in hepatocytes. SAA forms part of a feed‐forward loop and increases the release of IL‐1β and TNF‐α. While inflammation in the liver increases total leukocyte population, a decrease in the number of Kupffer cells is seen. The inflammatory milieu in the liver leads to the upregulation of profibrotic genes and the activation of stellate cells, which together promote liver fibrosis.

Rights and permissions

Copyright: The Author(s)

Reprints and permissions

About this article

Cite this article

Kaufmann, B., Leszczynska, A., Reca, A. et al. NLRP3 activation in neutrophils induces lethal autoinflammation, liver inflammation, and fibrosis.EMBO Rep 23, EMBR202154446 (2022). https://doi.org/10.15252/embr.202154446

Download citation

Keywords