NLRP3 deficiency protects from type 1 diabetes through the regulation of chemotaxis into the pancreatic islets
Contributed by Richard A. Flavell, July 16, 2015 (sent for review October 30, 2014; reviewed by Myung-Shik Lee and Matthias von Herrath)
Significance
Our study demonstrated that the nucleotidebinding oligomerization domain, leucine-rich repeat and pyrin domaincontaining protein 3 (NLRP3) pathway plays an important role in type 1 diabetes (T1D) using a mouse model. NLRP3 is critical for chemokine receptors CCR5 and CXCR3 expression on T cells, influencing pathogenic T-cell migration to the islets. It also affects the chemokines CCL5 and CXCL10 expression in the islets, preventing pathogenic T cells from infiltration. Targeting this pathway may be useful in prevention and treatment of T1D as it affects both immune cells and pancreatic beta cells.
Abstract
Studies in animal models and human subjects have shown that both innate and adaptive immunity contribute to the pathogenesis of type 1 diabetes (T1D). Whereas the role of TLR signaling pathways in T1D has been extensively studied, the contribution of the nucleotide-binding oligomerization domain, leucine-rich repeat and pyrin domain-containing protein (NLRP) 3 inflammasome pathway remains to be explored. In this study, we report that NLRP3 plays an important role in the development of T1D in the nonobese diabetic (NOD) mouse model. NLRP3 deficiency not only affected T-cell activation and Th1 differentiation, but also modulated pathogenic T-cell migration to the pancreatic islet. The presence of NLRP3 is critical for the expression of the chemokine receptors CCR5 and CXCR3 on T cells. More importantly, NLRP3 ablation reduced the expression of chemokine genes CCL5 and CXCL10 on pancreatic islet cells in an IRF-1–dependent manner. Our results suggest that molecules involved in chemotaxis, accompanied by the activation of the NLRP3 inflammasome, may be effective targets for the treatment of T1D.
Sign up for PNAS alerts.
Get alerts for new articles, or get an alert when an article is cited.
Type 1 diabetes (T1D) is a T-cell–mediated autoimmune disease characterized by the destruction of insulin-producing pancreatic beta cells in genetically predisposed individuals. Studies in animal models and human subjects have shown that both innate and adaptive immunity play a role in disease pathogenesis. Strategies targeting either T or B cells have shown some efficacy in T1D in both animal and human studies (1–4). Recently, the role of innate immunity in T1D has been increasingly appreciated. We, and others, have demonstrated that Toll-like receptor (TLR) signaling pathways are essential for the development of T1D. Nonobese diabetic (NOD) mice deficient in TLR2, TLR9, or MyD88 showed delayed disease development or were protected from diabetes (5–9). However, the development of autoimmune diabetes was accelerated in TLR4−/− NOD mice (5–7, 10). Whereas the role of TLR signaling has been intensively studied, the contribution of the nucleotide binding domain-like receptor (NLR) signaling pathway to the pathogenesis of T1D remains to be explored.
Nucleotide-binding oligomerization domain, leucine-rich repeat and pyrin domain-containing protein (NLRP) 3 is a NLR family member, together with ASC and caspase-1, forms protein complexes that are responsible for the innate immune response to pathogens and/or “danger” signals (11). Increasing evidence indicates that the NLRP3 inflammasome plays an important role in obesity and type 2 diabetes (12–14). However, little is known about the role of NLRP3 in autoimmune diabetes. Whereas the inflammasome has been extensively studied in the control of infection, only recently has the role of the NLRP inflammasome in autoimmune disease been recognized. Polymorphisms in inflammasome genes are involved in the predisposition to systemic lupus erythematosus (15). NLRP3 deficiency dramatically delayed the course and reduced severity of experimental autoimmune encephalomyelitis by suppression of Th1 and Th17 responses (16). Mice deficient in ASC, the adaptor protein of the NLRP3 inflammasome pathway, were also less susceptible to collagen-induced arthritis (17). Nevertheless, the role of the inflammasome pathway in the pathogenesis of T1D is unclear. Although caspase-1 or IL-1β deficiency did not protect NOD mice from T1D (18, 19), IL-1 blockade showed a synergistic protective effect when combined with anti-CD3 therapy for T1D in a mouse model (20). Interestingly, recent genetic association studies suggested that polymorphisms in inflammasome genes might be involved in the predisposition to T1D. A coding polymorphism in NLRP1 was demonstrated to confer susceptibility to T1D (21). Furthermore, two single-nucleotide polymorphisms in NLRP3 were identified in a separate association study as a predisposing factor for T1D (22).
Thus, we generated NLRP3-deficient (NLRP3−/− or KO) NOD mice to understand the role of NLRP3 in the pathogenesis of T1D. Here, we show that NOD mice deficient in NLRP3 were protected from T1D development. Mechanistic studies suggested that the expression of NLRP3, in both hematopoietic and nonhematopoietic cells, was important for diabetes development. Whereas NLRP3 deficiency in the hematopoietic compartment reduced the diabetogenicity of immune cells, its ablation in nonhematopoietic cells, particularly in the pancreatic islets, compromised the migration of immune cells into the target tissue. Destruction of beta cells was reduced via the down-regulation of chemokine gene expression in the pancreatic islets leading to protection from diabetes.
Results
NLRP3 Deficiency Prevented T1D in NOD Mice.
To understand the role of NLRP3 in the pathogenesis of T1D, we backcrossed NLRP3−/− C57BL/6 mice with NOD mice for more than 10 generations. The purity of the NOD genetic background was further confirmed by verification of known Idd loci (type1diabetes.jax.org/gqc). To study the effect of NLRP3 ablation on the development of T1D, we set up a cohort of NLRP3−/− NOD mice, as well as a group of wild-type littermates, to observe the natural history of disease development. As shown in Fig. 1A, compared with WT littermates, diabetes development in NLRP3−/− NOD mice was significantly reduced. Moreover, to test whether pharmaceutical inhibition of NLRP3 had a similar effect, we treated a group of WT NOD mice with either the NLRP3 inhibitor, parthenolide, an extract of a natural herb, or vehicle, via oral gavage for 4 wk. Intriguingly, chemical inhibition of NLRP3 also significantly delayed and reduced disease development in NOD mice (Fig. 1B). To understand the effect of NLRP3 deficiency on the cellular infiltration in pancreatic islets, we randomly selected some pancreata from nondiabetic female NOD mice (NLRP3−/− NOD and NLRP3+/+ littermates, n = 5–8 per group) and examined their insulitis after H&E staining. Consistent with the reduction in diabetes development, NLRP3−/− NOD mice showed considerably attenuated insulitis (Fig. S1A). There were significantly fewer T and B cells in the islet infiltrates (Fig. S1B).
Fig. 1.
Fig. S1.
The Expression of NLRP3 in Both Hematopoietic and Nonhematopoietic Cells Contributes to Diabetes Development.
NLRP3 is expressed in both immune cells and tissue cells (23, 24). To further investigate whether deficiency of NLRP3 in the hematopoietic or nonhematopoietic cell compartment was responsible for the observed protective effect, we performed a set of bone marrow chimera (BMC) experiments. NLRP3−/− NOD or WT recipients (all females) were lethally irradiated (1,000 cGy) and transplanted with 107 bone marrow (BM) cells from NLRP3−/− NOD or WT mice (age-matched females). The BMC mice were observed for diabetes development. It is interesting that although WT mice that received NLRP3−/− BM cells showed modest protection from diabetes (solid triangles in Fig. S2) in comparison with WT mice that received WT BM cells (solid dots in Fig. S2), the development of diabetes was highly inhibited in NLRP3−/−NOD recipients (squares and diamonds in Fig. S2) regardless of the genotype of the donor BM cells, which suggests that NLRP3 in nonhematopoietic cells plays a more important role in diabetes protection.
Fig. S2.
The Effect of NLRP3 Deficiency on the Hematopoietic Compartment.
The BMC experiment showed that NOD WT recipients that received NLRP3−/− NOD BM cells exhibited delayed and reduced diabetes development, which suggested that NLRP3 expression on immune cells plays a role in promoting disease development. Indeed, when splenocytes from NLRP3−/− NOD mice were adoptively transferred into RAG1−/− NOD mice, diabetes onset was significantly delayed in recipients in comparison with RAG1−/− mice that received WT splenocytes (Fig. 2A). To understand which subsets of immune cells expressing NLRP3 were essential, we first characterized the NLRP3−/− NOD mice. Compared with WT NOD mice, NLRP3−/− NOD mice had reduced splenocyte numbers, including significantly fewer CD4+ T cells, CD19+ B cells, and CD11b+ macrophages (Fig. S3A). There was also a reduced number of T and B cells, dendritic cells, and macrophages in pancreatic draining lymph nodes (Fig. S3B). However, we did not find any reduction of T cells in the thymus (CD4+; Fig. S3C) or CD8+ (Fig. S4A) or in B cells from the BM (Fig. S4B), but we did observe an increase in thymic B cells (Fig. S3C). Interestingly, the reduction of immune cells in the periphery of NLRP3−/− NOD mice appears to be genetic background-dependent because this reduction was not observed in NLRP3−/− C57BL/6 mice (Fig. S3D). It has been reported that T cells express NLRP3 (25), and deficiency of NLRP3 in our model system also led to less activated CD4+ T cells in both pancreatic draining lymph nodes and islet infiltrates as shown by a reduction in CD69 (Fig. 2B). Moreover, there were fewer CD4+ T cells producing IFN-γ in NLRP3−/− NOD mice in comparison with WT mice (Fig. 2C).
Fig. 2.
Fig. S3.
Fig. S4.
Furthermore, the deficiency of NLRP3 in T cells induced a change in T-cell function. Upon anti-CD3 stimulation, NLRP3−/− NOD T cells showed impaired proliferation in vitro (Fig. S5A). Interestingly, the reduced T-cell proliferation was not observed in NLRP3−/− C57BL/6 T cells (Fig. S5B). To test whether NLRP3 deficiency affected diabetogenic T-cell function in vivo, we labeled purified CD4+ T cells from WT and NLRP3−/− BDC2.5 mice with CFSE and then transferred the cells into WT NOD recipients. We harvested pancreatic draining lymph nodes from the recipient mice 3 and 7 d later and analyzed CFSE dilution, i.e., the proliferation of BDC2.5 cells. The proliferation of diabetogenic NLRP3−/− BDC2.5 CD4+ T cells was markedly reduced on day 3; however, the reduced proliferation was not obvious on day 7 (Fig. 2D) and diabetogenicity of BDC2.5 CD4+ T cells was not affected by the deficiency of NLRP3, as the WT NOD recipients became diabetic on day 9 (Fig. S6).
Fig. S5.
Fig. S6.
Macrophages and dendritic cells are the major immune cells expressing NLRP3. To investigate the effect of NLRP3 deficiency on these professional antigen-presenting cells (APCs), we stimulated APCs with LPS, CpG, Pam3csk4, and Poly I:C. Both WT and knockout APCs showed comparable responses to these TLR ligands based on CD80 and MHC II molecule expression (Fig. S7A). Next, we studied their antigen processing and presenting functions. Neither purified splenic APCs nor BM-derived APCs from WT or knockout mice showed any difference in antigen processing and presentation (Fig. S7B). However, APCs from NLRP3−/− NOD mice suppressed the differentiation of naïve CD4 T cells into IFN-γ–producing Th1 cells (Fig. 2E).
Fig. S7.
NLRP3 Deficiency Impaired the Migration of Diabetogenic CD4 T Cells to Islets.
NLRP3 deficiency impairs the migration of pathogenic cells into the brain in the EAE model (26, 27). Thus, we speculated that the improved insulitis in NLRP3−/− NOD mice could be a result of impaired immune cell migration to pancreatic islets. To understand whether NLRP3 deficiency impairs CD4+ T-cell homing, we performed cell trafficking experiments. We labeled WT BDC2.5 CD4+ T cells (Thy1.1+) and NLRP3−/− BDC2.5 CD4+ T cells (Thy1.2+) with CFSE and mixed the cells at a 1:1 ratio before transferring into WT NOD.SCID mice. Interestingly, 16 h after adoptive transfer, more labeled NLRP3−/− BDC2.5 CD4+ T cells than WT BDC2.5 CD4+ T cells resided in the spleen and pancreatic draining lymph nodes (Fig. 3A, Left). In contrast to WT BDC2.5 CD4+ T cells, significantly fewer NLRP3−/− BDC2.5 CD4+ T cells migrated into islets (Fig. 3A, Left). To clarify whether NLRP3 deficiency in the tissue organ prevents the recruitment of diabetogenic CD4+ T cells, we transferred the same CFSE-labeled mixture of BDC2.5 CD4+ T cells into NLRP3−/− NOD.SCID mice. Compared with WT NOD.SCID recipients, WT BDC2.5 CD4+ T cells also showed reduced recruitment (8 vs. 4 cells) to NLRP3−/− NOD.SCID islets 16 h after the transfer (Fig. 3A, Right). Furthermore, we found that the proliferation of CFSE-labeled WT BDC2.5 CD4+ T cells was impaired in NLRP3−/− NOD.SCID recipients compared with proliferation that occurred in the WT NOD.SCID recipients (Fig. 3B). More importantly, when we transferred WT BDC2.5 CD4+ T cells into WT or NLRP3−/− female NOD.SCID mice, diabetes development was significantly delayed and reduced in NLRP3−/− NOD.SCID recipients in comparison with WT NOD.SCID recipients (Fig. 3C).
Fig. 3.
NLRP3 Deficiency Affects Chemotaxis of Immune Cells Both Intrinsically and Extrinsically.
There were two possible reasons for the impaired T-cell proliferation and function in NLRP3−/− NOD islets. One was that there were more regulatory T cells in NLRP3−/− NOD islets, which suppressed T-cell function. However, we did not find a significant difference in the numbers of regulatory T cells in spleens and pancreatic draining lymph nodes between WT and NLRP3−/− NOD mice (Fig. S8). The other explanation was that NLRP3 deficiency impaired the chemotaxis of T cells to pancreatic islets, which was more likely, as evidenced by fewer BDC2.5 CD4+ T cells migrating to and proliferating in NLRP3−/− NOD islets in comparison with the WT islets. To further dissect the molecular mechanisms, we tested the gene expression of chemokines and chemokine receptors in islet cells and purified T cells, respectively.
Fig. S8.
As shown earlier in Fig. 3A, more NLRP3−/− NOD T cells were homing to, or staying in, the spleen and pancreatic draining lymph nodes. To clarify whether NLRP3 deficiency affected chemokine receptor gene expression on CD4+ T cells, we sorted CD4+ T cells from spleen, pancreatic draining lymph nodes, and islets of NLRP3−/− NOD and WT NOD mice and investigated CCR5, CCR7 and CXCR3 [chemokine (C-X-C motif) receptor 3] expression. Consistent with the alteration in cell migration, chemokine receptor CXCR3 gene expression was up-regulated in CD4+ T cells from NLRP3−/− NOD spleen and pancreatic draining lymph node in comparison with WT CD4 T cells (Fig. 4A). We also found enhanced expression of CCR5 in splenic CD4+ and CCR7 in pancreatic draining lymph node CD4+ T cells of NLRP3−/− mice (Fig. S9). However, the expression of CCR5, CCR7, and CXCR3 genes in islet-derived CD4+ T cells was all down-regulated in NLRP3−/− NOD mice compared with WT mice (Fig. 4B).
Fig. 4.
Fig. S9.
Adoptive transfer experiments indicated that fewer diabetogenic BDC2.5 CD4+ T cells migrated to NLRP3−/− NOD.SCID islets, in comparison with WT NOD.SCID islets (Fig. 3A), which suggested that NLRP3 deficiency also affected homing of diabetogenic cells to islets. An array of chemokine genes is expressed in islet cells (28). Thus, we speculated that the impaired migration of CD4+ T cells to NLRP3−/− NOD islets could be due to suppressed chemokine gene expression by NLRP3−/− NOD pancreatic islets. We isolated islets from 3-wk-old WT and knockout NOD mice and performed quantitative PCR (qPCR) (see Table S1 for primer sequences) to analyze the expression of chemokine genes including CCL3 [chemokine (C-C motif) ligand 3], CCL5, CCL21, CXCL9 [chemokine (C-X-C motif) ligand 9], and CXCL10. As expected, there was a significant down-regulation of all of the chemokine genes tested in NLRP3−/− NOD pancreatic islets compared with WT islets (Fig. 4C). Moreover, the NLRP3 inhibitor, parthenolide, also inhibited LPS-induced chemokine gene expression in WT NOD islets (Fig. 4D). To further confirm the impaired chemo-attractant expression by NLRP3−/− NOD islets at the protein level, we performed in vitro transwell migration experiments. We cultured isolated WT or NLRP3−/− NOD islets overnight at 37 °C. The islet culture supernatant was used as the source of chemokine(s). Consistent with qPCR results, the supernatants from NLRP3−/− NOD islets had reduced capacity to recruit WT CD4+ T cells and the addition of CCL5 and CXCL10 to the supernatants completely restored the migration of CD4+ T cells (Fig. 4E). Moreover, the protein levels of CCL5 and CXCL10 in the supernatants from the islets were also reduced (Fig. S10). However, in comparison with WT islet supernatants, NLRP3−/− NOD islet supernatant showed similar ability to recruit WT macrophages (Fig. 4F), and the addition of CCL5 and CXCL10 enhanced macrophage migration (Fig. 4F). Our results suggest that the migration of immune cells to islets is regulated by chemokine gradients from islet cells. To understand whether NLRP3 deficiency also affected the expression of chemokine receptor gene(s) on immune cells, we sorted splenic CD4+ T cells and macrophages from WT and NLRP3−/− NOD mice to determine their capability to migrate in a transwell migration experiment. As shown in Fig. 4 G and H, in the presence of WT islet culture supernatants, both CD4 T+ cells and macrophages from NLRP3-deficient NOD mice exhibited impaired migration in comparison with WT CD4+ T cells and macrophages. Thus, our data indicate that NLRP3 acts on both immune cells and pancreatic islets through chemokines and chemokine receptors in regulating immune cell migration to islets.
Table S1.
Primer | Sequence |
---|---|
CCL2-forward | TAAAAACCTGGATCGGAACCAAA |
CCL2-reverse | GCATTAGCTTCAGATTTACGGGT |
CCL3-forward | TGTACCATGACACTCTGCAAC |
CCL3-reverse | CAACGATGAATTGGCGTGGAA |
CCL5-forward | TTTGCCTACCTCTCCCTCG |
CCL5-reverse | CGACTGCAAGATTGGAGCACT |
CCL21-forward | GTGATGGAGGGGGTCAGGA |
CCL21-reverse | GGGATGGGACAGCCTAAACT |
CXCL9-forward | GGAGTTCGAGGAACCCTAGTG |
CXCL9-reverse | GGGATTTGTAGTGGATCGTGC |
CXCL10-forward | CCAAGTGCTGCCGTCATTTTC |
CXCL10-reverse | GGCTCGCAGGGATGATTTCAA |
CCR5-forward | TTTTCAAGGGTCAGTTCCGAC |
CCR5-reverse | GGAAGACCATCATGTTACCCAC |
CCR7-forward | TGTACGAGTCGGTGTGCTTC |
CCR7-reverse | GCTAGGTATCCGTCATGGTCTTG |
CXCR3-forward | TACCTTGAGGTTAGTGAACGTCA |
CXCR3-reverse | CGCTCTCGTTTTCCCCATAATC |
NLRP3-forward | CGAGACCTCTGGGAAAAAGCT |
NLRP3-reverse | GCATACCATAGAGGAATGTGATGTACA |
IL-1β-forward | GAAATGCCACCTTTTGACAGTG |
IL-1β-reverse | TGGATGCTCTCATCAGGACAG |
IRF1-forward | ATGCCAATCACTCGAATGCG |
IRF1-reverse | CCTGCTTTGTATCGGCCTGT |
SphK1-forward | TATGCTGGGTACGAGCAGGT |
SphK1-reverse | CCCACTGTGAAACGAATCTCC |
TRAF6-forward | AAAGCGAGAGATTCTTTCCCTG |
TRAF6-reverse | ACTGGGGACAATTCACTAGAGC |
cIAP2-forward | ACGCAGCAATCGTGCATTTTG |
cIAP2-reverse | CCTATAACGAGGTCACTGACGG |
HPRT-forward | GTTGGATACAGGCCAGACTTTGTTG |
HRPT-reverse | GAGGGTAGGCTGGCCTATAGGCT |
Fig. S10.
IFN-Regulatory Factor 1 Pathway Is Involved in the Regulation of Chemokine Gene Expression.
IFN-regulatory factor 1 (IRF-1) deficiency leads to reduction in antigen-specific autoimmune diseases including collagen-induced arthritis and experimental autoimmune encephalomyelitis (EAE) (29). More recently, Harikumar et al. reported that the transcription factor IRF1 is essential for IL-1β–induced production of the chemokines CCL5 and CXCL10 through its formation of a signaling complex with TRAF6, SphK1, and cIAP2 (30). To understand whether this signaling pathway is involved in controlling the production of CCL5 and CXCL10 in pancreatic islets, we isolated islets from NOD mice and analyzed the expression of IRF1, TRAF6, Sph1, and cIAP2 by real-time PCR (Table S1). Interestingly, all of the four genes were detected in WT NOD islet cells (Fig. 5A). Moreover, the expression of these genes was up-regulated in NOD islets upon stimulation by LPS (Fig. 5A). However, NLRP3 ablation repressed their up-regulation induced by LPS (Fig. 5A). Consistent with published data (30), the expression of the chemokine genes CCL5 and CXCL10 also correlated with the expression of IRF1 signaling complex genes (Fig. 5B), which suggested that the IRF1 signaling pathway is most likely involved in the impaired homing of diabetogenic T cells in to pancreatic islets by regulating chemokine gene expression in islets of NLRP3−/− NOD mice.
Fig. 5.
To verify the expression of NLRP3 and IL-1β transcripts in pancreatic islet cells, we stained dispersed islet cells with CD45 and sorted CD45+ (hematopoietic) and CD45− (nonhematopoietic) cells and examined the expression of NLRP3 and IL-1β transcripts in these two cell populations. We found that nonhematopoietic cells in WT pancreatic islets clearly expressed NLRP3 and IL-1β transcripts (Fig. 5C). To further confirm the finding, we sorted (i) beta cells (CD45−FluoZin-3-AM+); (ii) nonbeta cells and nonimmune cells (CD45−FluoZin-3-AM−) and (iii) immune cells (CD45+, which includes macrophage) from pancreatic islets of wild-type NOD mice and detected IL-1β and NLRP3 expression by qPCR in the three subsets of cells. As shown in Fig. 5D, compared with CD45+ immune cells, beta cells produced significant amounts of IL-1β and increased expression of NLRP3, and even nonbeta, nonimmune cells (CD45−FluoZin-3-AM−) expressed IL-1β and NLRP3. Thus, our results suggest that the NLRP3 and IL-1β, expressed in nonhematopoietic cells in islets, are involved in the reduced incidence of diabetes in NLRP3−/− NOD mice. Together with the data from BMC experiments (Fig. S2), our results support the notion that expression NLRP3 gene expression in nonhematopoietic cells plays a more important role in diabetes development in the NOD mouse than NLRP3 gene expression in hematopoietic cells. To verify the expression of chemokine genes in nonhematopoietic cells in pancreatic islets, we further tested the gene expression of CCL2 [chemokine (C-C motif) ligand 2], CCL5, CXCL9, and CXCL10 in CD45− islet cells from WT or NLRP3−/− NOD islets. All of the chemokine genes tested could be readily detected in CD45− islet cells; however, except for CCL5, the expression levels were significantly lower in CD45− islet cells from NLRP3−/− NOD mice (Fig. 5E). In line with the chemokine gene expression, not only was the expression of IRF1, TRAF6, Sph1, and cIAP2 genes detected in sorted CD45− islets cells, but the expression of the genes was also significantly down-regulated in NLRP3−/− NOD CD45− islets cells compared with their counterparts in WT nonhematopoietic islet cells (Fig. 5F).
Discussion
It is known that the NLRP3 inflammasome plays a role in the development of insulin resistance and T2D (12–14, 31). However, the role of the NLRP3 inflammasome in autoimmune T1D remained to be explored. Here, we show that NLRP3 is critical for the development of autoimmune diabetes in the NOD mouse model. The genetic ablation or pharmaceutical inhibition of NLRP3 delayed and reduced the development of T1D in the NOD mouse. Mechanistic studies indicated that the expression of NLRP3 is required in both hematopoietic and nonhematopoietic cells for disease development, although its expression in the nonhematopoietic compartment of pancreatic islet cells exerts a stronger influence than expression in hematopoietic cells. Our results demonstrated that NLRP3 deficiency not only led to suppressed T-cell activation and Th1 cell differentiation, but also impaired the migration of diabetogenic cells into pancreatic islets through down-regulation of chemotaxis-related gene expression on both T cells and nonhematopoietic cells from islets. The down-regulation of chemokine gene expression in nonhematopoietic cells from NLRP3−/− NOD islets is likely to be mediated through the IRF1 signaling pathway.
In addition to the delayed and reduced spontaneous diabetes development in NLRP3−/− NOD mice, in two different adoptive transfer model systems, we found that NLRP3-deficient immature BM cells or mature splenocytes could not transfer diabetes and their wild-type counterparts. This result clearly implies that the expression of NLRP3 in hematopoietic cells plays an important role in the pathogenesis of T1D. Our study shows that NLRP3 deficiency suppresses Th1 responses in NLRP3−/− NOD mice, which is consistent with recent work demonstrating that NLRP3−/− mice exhibited significantly milder EAE through the reduction in IFN-γ–expressing T helper cells (16, 26). Chemokine receptors such as CCR5 and CXCR3 are preferentially expressed by Th1 cells (32, 33). Interestingly, we found that CD4+ T cells in NLRP3−/− NOD islets expressed significantly less CCR5 and/or CXCR3 in comparison with CD4+ T cells in WT islets, suggesting that fewer Th1 cells were recruited to the pancreatic islets. In contrast to the study by Gris et al. in the EAE mouse model (16), we did not detect suppressed Th17 responses in NLRP3−/− NOD mice.
Unlike tissue resident cells such as macrophages and dendritic cells, T cells are rarely found in healthy islets. Thus, the recruitment of diabetogenic effector T cells to islets is a critical step for the islet inflammation and beta cell destruction in T1D. The expression of certain chemokines in the hematopoietic and nonhematopoietic cells in islets is essential for the migration of T cells into pancreatic islets. It has been shown that, in both humans and rodents, pancreatic islet cells express many chemokine genes including CCL2, CCL3, CCL5, and CXCL10, and among them, CXCL10 is the most abundant chemokine gene expressed by the pancreatic islet (28). Consistent with a recent study in human islets (28), we also found that CXCL10 was the predominant chemokine expressed by WT NOD islets. However, its expression was significantly down-regulated in NLRP3−/− NOD islets (Fig. 4C and Fig. S10). Both CCL5 and CXCL10 are important for the recruitment of Th1 cells, and the expression of CCL5 was also suppressed in NLRP3-deficient NOD pancreatic islets (Fig. 4C). CCR5 and CXCR3 are the chemokine receptors for CCL5 and CXCL10, respectively. Interestingly, the expression pattern of Th1-targeting chemokine genes in NLRP3−/− NOD islets correlated with the chemokine receptor gene expression profile of the islet-infiltrating CD4+ T cells. Down-regulated CCR5 and CXCR3 expression was observed in CD4+ T cells from NLRP3-deficient NOD islets (Fig. 4B). In line with the chemokine gene expression results, there was impaired recruitment of polyclonal CD4+ T cells (Fig. 4E) and monoclonal BDC2.5 CD4+ T cells (Fig. 3) into NLRP3−/− NOD islets. Thus, it is likely that the genetic ablation of NLRP3 leads to the down-regulation of chemokine gene expression in pancreatic islets (Fig. 4A) and the reduced insulitis (Fig. S1), thus protecting NOD mice from diabetes development (Fig. 1).
There are two possible explanations for the chemokine gene down-regulation in NLRP3−/− NOD islets: (i) suppressed Th1 responses and (ii) impaired IL-1β production. It is known that IFN-γ can induce the production of CCL5 and CXCL10 (34, 35). Together with IL-1β and TNF-α, IFN-γ also induces significant amount of CXCL10, CXCL9, and CCL5 in human or mouse islets and the NIT-1 insulinoma cell line (28, 36). However, it is likely that the down-regulation of CCL5 and CXCL10 gene expression observed in NLRP3−/− NOD pancreatic islets (Fig. 4 C and D) may only be partially due to the suppressed Th1 responses, because the level of circulating IFN-γ between WT and NLRP3−/− NOD mice was similar at the time points measured (Fig. S11). Regardless, IL-1β alone can also induce significant amounts of chemokine gene expression in islets (28). IRF-1 is essential for IL-1β–induced secretion of the chemokines CCL5 and CXCL10 through the formation of signaling complexes with TRAF6, cIAP2, and SphK1 (30). Thus, it is likely that islet cell-derived IL-1β (Fig. 5D) can up-regulate chemokine gene expression in an IRF-1 signaling pathway-dependent manner. Indeed, we have found IRF-1, TRAF6, cIAP2, and SphK1 expression in islet cells (Fig. 5A), and the expression could be up-regulated by LPS in WT islets but suppressed in NLRP3−/− NOD islets. The low expression level of IRF-1 including the secreted IRF-1 (Fig. S10) is correlated with the low expression of the chemokines CCL5 and CXCL10 in NLRP3−/− NOD islets (Fig. 5 A and B). Importantly, the expression of IRF1, TRAF6, cIAP2, and SphK1 is not only found in CD45+ cells in islets, but also in CD45− islet cells (Fig. 5F). Interestingly, the expression of IRF-1, TRAF6, cIAP2, and SphK1 was also down-regulated in NLRP3−/− CD45− islet cells (Fig. 5F), which correlated with the expression of CXCL10 gene in CD45− islet cells (Fig. 5E).
Fig. S11.
NLRP3 is an important inflammasome family member that plays a key role in various types of sterile autoinflammation. In addition to IL-1β, NLRP3 also activates IL-18. The role of IL-18 in T1D is controversial; some studies suggest that IL-18 promotes T1D development (37), possibly through inducing diabetogenic T-cell expansion in NOD mice (38), whereas other studies, including our own, indicate that IL-18 is not required for T1D development (18, 19). However, the role of IL-18 in diabetes protection in the NLRP3-deficient NOD mice is not clear.
In summary, our study showed that inhibition of NLRP3 protected from T1D development in NOD mice. The ablation of NLRP3 led to suppressed Th1 responses and impaired T-cell migration to pancreatic islets through the down-regulation of chemokine expression in islets. Investigating the role of the inflammasome in the development of autoimmunity adds a further dimension to our understanding of the multifactorial nature of the immunopathology that leads to the development of T1D but also opens a new area of research for potential therapy.
Materials and Methods
For additional information regarding mice, antibodies and reagents, real-time PCR, BMC, transwell migration assay, adoptive transfer, islet beta cell isolation, insulitis score, and statistical analysis, see SI Materials and Methods.
SI Materials and Methods
Mice.
All of the animals used in the current study were housed in specific pathogen-free conditions with a 12-h dark/light cycle at the Yale University animal facility. The NLRP3-deficient NOD mice were generated by backcrossing of NLRP3−/− C57BL/6 mice (25) with NOD mice, for more than 10 generations. The purity of the NOD genetic background was further confirmed by verification of known Idd loci by PCR (type1diabetes.jax.org/gqc). The NLRP3−/− BDC2.5, NLRP3−/− NOD.SCID were generated by intercrossing of NLRP3−/− NOD mice with BDC2.5, and NOD.SCID mice, respectively. The use of the animals in the current study was approved by the Yale University Institutional Animal Care and Use Committee.
Antibodies and Reagents.
All fluorochrome-conjugated monoclonal antibodies used for flow cytometry were purchased from Biolegend and eBioscience. Parthenolide (sc-3523A) was from Santa Cruz Biotechnology. The qPCR kits were from KAPA Biosystems. RNeasy mini kits were from QIAGEN. Reverse-transcription PCR kits were purchased from Bio-Rad. Mouse IRF1 ELISA kit was from Antibodies Online. All of the chemicals used in this study were from Sigma.
Real-Time PCR.
Total RNA was isolated from hand-picked islets, and flow cytometry-sorted pancreatic beta cells were sorted from the hand-picked islets with RNeasy Mini kit. CDNA was produced by using iScript cDNA synthesis kits. The transcriptional level of genes was detected by qPCR on an iCycler (Bio-Rad). Primers sequences are listed in Table S1.
BMC.
BM chimeric mice were generated with 107 freshly isolated total BM cells from the femurs and tibias of WT or NLRP3−/− mice (donors, 8 wk old), respectively. Isolated cells were i.v. injected into lethally irradiated (1,000 cGY) female WT or NLRP3−/− mice (recipients, 6 wk old). The chimerism was confirmed by FACS analysis of blood samples from the recipients, 4 wk after injection. Chimeric mice were observed for diabetes development up to 8 mo of age.
Transwell Migration Assay.
Overnight culture supernatant (0.5 mL) from WT or NLRP3−/− islets (100 islet per mL) in 2% BSA RPMI 1640 medium was added into wells of 12-well plates as a source of chemoattractant. Transwell inserts (Fisher catalog no. 07-200-561) containing 0.5 mL of 5 × 105/mL bead-purified CD4+ T cells or macrophages (kits from Stem Cell Technology) in 2% BSA RPMI 1640 medium were placed in a 12-well plate with islet culture medium. The transwell system was incubated at 37 °C, 5% CO2 incubator for 16 h, and the migrated cells in the lower chamber were then counted under a microscope.
Islet and Beta Cell Isolation.
Islets were isolated as described. Briefly, mice were killed by cervical dislocation. The pancreas was inflated with 3 mL of cold collagenase P solution (0.3 mg/mL) through the bile duct with a 20 G needle, starting at the gall bladder. The pancreas was then removed from the body and placed in a siliconized vial containing 2 mL of 1 mg/mL collagenase solution and digested at 37 °C in a water bath for 12∼15 min. After three washes of the digested pancreas, islets were hand-picked under a dissecting microscope for further experiments. Purified beta cells from dispersed islets were flow cytometrically sorted after staining with anti-CD45 and FluoZin-3-AM (CD45−FluoZin-3-AM+).
Adoptive Transfer.
Ten million splenocytes from WT NOD (diabetic or nondiabetic) or NLRP3−/− NOD (nondiabetic) mice were i.v. injected into 6-wk-old female RAG1−/− NOD recipients. Alternatively, 2 million magnetic bead-purified BDC2.5 CD4+ T cells were adoptively transferred into 6-wk-old female NOD.SCID or NLRP3−/− NOD.SCID mice by i.v. injection. Disease development was monitored by testing urine glucose and confirmed by a blood glucose level higher than 250 mg/dL.
For the monitoring of diabetogenic CD4+ T-cell trafficking, 2 million purified BDC2.5 CD4+ T cells were labeled with CFSE and then i.v. injected into WT or NLRP3−/− NOD.SCID mice. Sixteen hours later, spleens, pancreatic draining lymph nodes (PLN), and islets were collected for flow cytometric analysis of CFSE-labeled cell trafficking.
Insulitis Score.
Pancreata were fixed in 10% neutral formalin buffer. Paraffin-embedded pancreata were then sectioned and stained with hematoxylin-eosin. Insulitis scores were determined by using the following grading scale: 0, no insulitis; 1, insulitis affecting less than 25% of the islet; 2, insulitis affecting between 25% and 75% of the islet; 3, more than 75% of islet infiltration. Eighty-six to 150 islets were scored for insulitis in each group (n = 5∼8 mice) by an individual blinded to the experimental design.
Statistical Analysis.
All of the statistical analysis was performed with GraphPad Prism software. P values less than 0.05 were considered statistically significant.
Acknowledgments
We thank Jian Peng (Yale University) for genotyping all the mice used in this study, all the laboratory members for insightful discussion, and Caroline Lieber for secretarial assistance. This work was supported by NIH Grants DK088181, DK092882, and JDRF 47-2013-516 (to L.W.); Yale Diabetes Research Center Feasibility and Pilot Grant NIH P30 DK045735 (to C.H.); and Juvenile Diabetes Research Foundation Advanced Postdoctoral Fellowship 10-2010-478 (to C.H.). H.D. was supported by National Natural Science Foundation of China Grant 81200586, and J.A.P. is a Fulbright-Diabetes UK Postdoctoral Research Scholar. R.A.F. is an Investigator of the Howard Hughes Medical Institute.
Supporting Information
Supporting Information (PDF)
Supporting Information
- Download
- 898.64 KB
References
1
KC Herold, et al., Prevention of autoimmune diabetes with nonactivating anti-CD3 monoclonal antibody. Diabetes 41, 385–391 (1992).
2
KC Herold, et al., Anti-CD3 monoclonal antibody in new-onset type 1 diabetes mellitus. N Engl J Med 346, 1692–1698 (2002).
3
CY Hu, et al., Treatment with CD20-specific antibody prevents and reverses autoimmune diabetes in mice. J Clin Invest 117, 3857–3867 (2007).
4
MD Pescovitz, et al., Rituximab, B-lymphocyte depletion, and preservation of beta-cell function. N Engl J Med; Type 1 Diabetes TrialNet Anti-CD20 Study Group 361, 2143–2152 (2009).
5
HS Kim, et al., Toll-like receptor 2 senses beta-cell death and contributes to the initiation of autoimmune diabetes. Immunity 27, 321–333 (2007).
6
L Wen, et al., Innate immunity and intestinal microbiota in the development of Type 1 diabetes. Nature 455, 1109–1113 (2008).
7
FS Wong, et al., The role of Toll-like receptors 3 and 9 in the development of autoimmune diabetes in NOD mice. Ann N Y Acad Sci 1150, 146–148 (2008).
8
CM Filippi, et al., TLR2 signaling improves immunoregulation to prevent type 1 diabetes. Eur J Immunol 41, 1399–1409 (2011).
9
N Tai, FS Wong, L Wen, TLR9 deficiency promotes CD73 expression in T cells and diabetes protection in nonobese diabetic mice. J Immunol 191, 2926–2937 (2013).
10
E Gülden, et al., Toll-like receptor 4 deficiency accelerates the development of insulin-deficient diabetes in non-obese diabetic mice. PLoS One 8, e75385 (2013).
11
K Schroder, J Tschopp, The inflammasomes. Cell 140, 821–832 (2010).
12
N Esser, et al., Obesity phenotype is related to NLRP3 inflammasome activity and immunological profile of visceral adipose tissue. Diabetologia 56, 2487–2497 (2013).
13
B Vandanmagsar, et al., The NLRP3 inflammasome instigates obesity-induced inflammation and insulin resistance. Nat Med 17, 179–188 (2011).
14
T Jourdan, et al., Activation of the Nlrp3 inflammasome in infiltrating macrophages by endocannabinoids mediates beta cell loss in type 2 diabetes. Nat Med 19, 1132–1140 (2013).
15
A Pontillo, et al., Polimorphisms in inflammasome genes are involved in the predisposition to systemic lupus erythematosus. Autoimmunity 45, 271–278 (2012).
16
D Gris, et al., NLRP3 plays a critical role in the development of experimental autoimmune encephalomyelitis by mediating Th1 and Th17 responses. J Immunol 185, 974–981 (2010).
17
L Kolly, et al., Inflammatory role of ASC in antigen-induced arthritis is independent of caspase-1, NALP-3, and IPAF. J Immunol 183, 4003–4012 (2009).
18
WH Schott, et al., Caspase-1 is not required for type 1 diabetes in the NOD mouse. Diabetes 53, 99–104 (2004).
19
L Wen, et al., In vivo diabetogenic action of CD4+ T lymphocytes requires Fas expression and is independent of IL-1 and IL-18. Eur J Immunol 41, 1344–1351 (2011).
20
V Ablamunits, et al., Synergistic reversal of type 1 diabetes in NOD mice with anti-CD3 and interleukin-1 blockade: Evidence of improved immune regulation. Diabetes 61, 145–154 (2012).
21
NF Magitta, et al., A coding polymorphism in NALP1 confers risk for autoimmune Addison’s disease and type 1 diabetes. Genes Immun 10, 120–124 (2009).
22
A Pontillo, et al., Two SNPs in NLRP3 gene are involved in the predisposition to type-1 diabetes and celiac disease in a pediatric population from northeast Brazil. Autoimmunity 43, 583–589 (2010).
23
G Guarda, et al., Differential expression of NLRP3 among hematopoietic cells. J Immunol 186, 2529–2534 (2011).
24
Z Huang, et al., Tissue-specific expression of the NOD-like receptor protein 3 in BALB/c mice. J Vet Sci 15, 173–177 (2014).
25
FS Sutterwala, et al., Critical role for NALP3/CIAS1/Cryopyrin in innate and adaptive immunity through its regulation of caspase-1. Immunity 24, 317–327 (2006).
26
M Inoue, ML Shinohara, NLRP3 Inflammasome and MS/EAE. Autoimmune Dis 2013, 859145 (2013).
27
M Inoue, KL Williams, MD Gunn, ML Shinohara, NLRP3 inflammasome induces chemotactic immune cell migration to the CNS in experimental autoimmune encephalomyelitis. Proc Natl Acad Sci USA 109, 10480–10485 (2012).
28
SA Sarkar, et al., Expression and regulation of chemokines in murine and human type 1 diabetes. Diabetes 61, 436–446 (2012).
29
Y Tada, A Ho, T Matsuyama, TW Mak, Reduced incidence and severity of antigen-induced autoimmune diseases in mice lacking interferon regulatory factor-1. J Exp Med 185, 231–238 (1997).
30
KB Harikumar, et al., K63-linked polyubiquitination of transcription factor IRF1 is essential for IL-1-induced production of chemokines CXCL10 and CCL5. Nat Immunol 15, 231–238 (2014).
31
IN Mohamed, et al., Thioredoxin-interacting protein is required for endothelial NLRP3 inflammasome activation and cell death in a rat model of high-fat diet. Diabetologia 57, 413–423 (2014).
32
R Bonecchi, et al., Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J Exp Med 187, 129–134 (1998).
33
F Sallusto, D Lenig, CR Mackay, A Lanzavecchia, Flexible programs of chemokine receptor expression on human polarized T helper 1 and 2 lymphocytes. J Exp Med 187, 875–883 (1998).
34
AD Luster, JV Ravetch, Genomic characterization of a gamma-interferon-inducible gene (IP-10) and identification of an interferon-inducible hypersensitive site. Mol Cell Biol 7, 3723–3731 (1987).
35
F Sallusto, CR Mackay, A Lanzavecchia, The role of chemokine receptors in primary, effector, and memory immune responses. Annu Rev Immunol 18, 593–620 (2000).
36
S Frigerio, et al., Beta cells are responsible for CXCR3-mediated T-cell infiltration in insulitis. Nat Med 8, 1414–1420 (2002).
37
Y Oikawa, et al., Systemic administration of IL-18 promotes diabetes development in young nonobese diabetic mice. J Immunol 171, 5865–5875 (2003).
38
AM Marleau, NE Sarvetnick, IL-18 is required for self-reactive T cell expansion in NOD mice. J Autoimmun 36, 263–277 (2011).
Information & Authors
Information
Published in
Classifications
Submission history
Published online: August 24, 2015
Published in issue: September 8, 2015
Keywords
Acknowledgments
We thank Jian Peng (Yale University) for genotyping all the mice used in this study, all the laboratory members for insightful discussion, and Caroline Lieber for secretarial assistance. This work was supported by NIH Grants DK088181, DK092882, and JDRF 47-2013-516 (to L.W.); Yale Diabetes Research Center Feasibility and Pilot Grant NIH P30 DK045735 (to C.H.); and Juvenile Diabetes Research Foundation Advanced Postdoctoral Fellowship 10-2010-478 (to C.H.). H.D. was supported by National Natural Science Foundation of China Grant 81200586, and J.A.P. is a Fulbright-Diabetes UK Postdoctoral Research Scholar. R.A.F. is an Investigator of the Howard Hughes Medical Institute.
Authors
Competing Interests
The authors declare no conflict of interest.
Metrics & Citations
Metrics
Citation statements
Altmetrics
Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
Loading...
View Options
View options
PDF format
Download this article as a PDF file
DOWNLOAD PDFLogin options
Check if you have access through your login credentials or your institution to get full access on this article.
Personal login Institutional LoginRecommend to a librarian
Recommend PNAS to a LibrarianPurchase options
Purchase this article to access the full text.