NLRP3 inflammasome [NLR (nucleotide-binding domain, leucine-rich repeat containing protein) Pyrin-domain-containing 3 ] functions as an innate sensor of several PAMPs and DAMPs (pathogen- and damage-associated molecular patterns). It has been also reported as a transcription factor related to Th2 pattern, although its role in the adaptive immunity has been controversial, mainly because the studies were performed using gene deletion approaches. In the present study, we have investigated the NLRP3 gain-of-function in the context of encephalomyelitis autoimmune disease (EAE), considered to be a Th1- and Th17-mediated disease. We took advantage of an animal model with NLRP3 gain-of-function exclusively to T CD4+ lymphocytes (CD4CreNLRP3fl/fl). These mice presented reduced clinical score, accompanied by less infiltrating T CD4+ cells expressing both IFN-γ and IL-17 at the central nervous system (CNS) during the peak of the disease. However, besides NLRP3 gain-of-function in lymphocytes, these mice lack NLRP3 expression in non-T CD4+ cells. Therefore, in order to circumvent this deficiency, we transferred naive CD4+ T cells from WT, NLRP3−/− or CD4CreNLRP3fl/fl into Rag-1−/− mice and immunized them with MOG35–55. Likewise, the animals repopulated with CD4CreNLRP3fl/fl T CD4+ cells presented reduced clinical score and decreased IFN-γ production at the peak of the disease. Additionally, primary effector CD4+ T cells derived from these mice presented reduced glycolytic profile, a metabolic profile compatible with Th2 cells. Finally, naive CD4+ T cells from CD4CreNLRP3fl/fl mice under a Th2-related cytokine milieu cocktail exhibited in vitro an increased IL-4 and IL-13 production. Conversely, naive CD4+ T cells from CD4CreNLRP3fl/fl mice under Th1 differentiation produced less IFN-γ and T-bet. Altogether, our data evidence that the NLRP3 gain-of-function promotes a Th2-related response, a pathway that could be better explored in the treatment of multiple sclerosis.
NLR (Nucleotide-binding domain, leucine-rich repeat containing protein) is a group of intracellular proteins that regulate inflammation. Among NLRs, NLRP3 inflammasome (NLR Pyrin-domain-containing 3) has a broad spectrum of specificity, including crystals of silica and uric acid , extracellular adenosines, extracellular matrix components, toxins, among others [2,3]. Induction of NLRP3 inflammasome assembly through endogenous ligands, also known as DAMPs (damage-associated molecular patterns), indicates that there is a potential link between the innate immune system and maintenance of the body’s homeostasis. NLRP3 activation by DAMPs or PAMPs (pathogen-associated molecular patterns) is a process mediated by the innate immune system, since its specificity is not restricted to a single epitope, and its diversity is not distributed clonally between cells. However, NLRP3 deficiency leads to milder damage in the development of autoimmune diseases, typical of an adaptive immune response profile . The deficiency of NLRP3 in the context of experimental autoimmune encephalomyelitis (EAE), for instance, leads to reduced lymphocytic infiltration and lower levels of IFN-γ and IL-17, resulting in diminished disease score .
The promoter region of the NLRP3 gene presents responsive elements for NF-κB , an important transcription factor in the inflammatory process. The role of NLRP3 as a participant in adaptive immune system has been controversial. It has been demonstrated that NLRP3 inflammasome assembles in human CD4+ T cells and initiates caspase-1 and IL-1β secretion, thereby promoting IFN-γ production and Th1 differentiation after stimulation with anti-CD3. Moreover, constitutively active NLRP3 in T cells from patients with cryopyrin-associated autoinflammatory syndrome (CAPS) induced hyperactive Th1 responses that could be normalized with a NLRP3 inhibitor . On the other hand, it was found that NLRP3 expression in CD4+ T cells specifically supported a Th2 transcriptional program in a cell-intrinsic manner, when T cells are stimulated with Th2-related cytokine cocktail . According to Bruchard and colleagues, NLRP3 binds to the IL-4 promoter in conjunction with the transcription factor IRF4, leading to IL-4 production by T lymphocytes . These results were also supported by another study indicating that Th2 genes such as c-maf or il4 were not induced in NLRP3 deficient cells .
EAE is the animal model most frequently used to study multiple sclerosis (MS), a disease that causes central nervous system (CNS) inflammation and demyelination. Several cells play an important role in the development of EAE, such as myelin sheath producers, oligodendrocytes, CD8+ T lymphocytes , astrocytes, , microglia , macrophages  and dendritic cells . Initially, MS was considered a Th1 type disease, but experiments in which IFN-γ deficient animals were more susceptible to EAE did not explain the proposed immune profile for the disease . It has also been shown that IL-23 deficiency was more protective against the development of EAE than IL-12 deficiency . Further, a new population of T cells, which produced cytokines such as IL-17A and IL-17F , IL-21  and GM-CSF  were described as the main players in the development of EAE. It is now well accepted that both CD4+ T lymphocyte populations are important in the development of EAE, with Th17 profile lymphocytes reaching the CNS initially via the choroid plexus through CCR6  and the Th1 profile lymphocytes subsequently reaching the CNS by integrin VLA-4 capillary transmigration . Myelin-specific Th1 and Th17 cells, for example, add to MS  and EAE  physiopathology. The dysregulated activity of effector Th1 and Th17 cells results in the development of tissue inflammation and autoimmunity . Thus, it is of great relevance to understand the mechanisms by which Th1 and Th17 cells are modulated in vivo. In the present study, we investigated the role of NLRP3 in the generation of Th1 and Th17 cells and the consequent development of EAE. We tested both the deletion and the gain of function of NLRP3 specifically in CD4+ T cells in the onset of EAE. Interestingly, we observed that, in fact, NLRP3 gain-of-function in T cells favors a Th2 cytokine profile both in vitro and in vivo, reducing overall clinical scores of EAE mice.
Materials and methods
Isogenic female C57BL/6, Rag1−/−, NLRP3−/− and CD4CreNLRP3flox mice, aging 8–12 weeks (25–28 g), obtained from our Isogenic Breeding Unit (University of Sao Paulo), were housed in standard cages and had free access to water and food. NLRP3−/− (also NLRP3flox/flox) was bought from The Jackson Laboratory under the line name B6N.129- Nlrp3tm3Hhf/J. These mice contain a floxed neomycin cassette in intron 2 of the NLRP3 gene, in opposite orientation to gene, abolishing gene expression. These mice also contain a point mutation in exon 3, which results in a missense mutation related to gain-of-function in NLRP3, commonly found in humans with CAPS syndrome. When bred to mice that express Cre recombinase, resulting offspring will have the floxed-neoR deleted in the Cre-expressing tissues, allowing expression of the mutated gene, while Cre-negative cells will present a NLRP3−/− phenotype. Once heterozygous littermates present altered phenotype, we have used WT as control mice.
For the EAE development, mice were immunized subcutaneously with 150 μg of MOG35– 55 emulsified in CFA (v/v) containing 500 μg of Mycobacterium tuberculosis Des, H37Ra (Becton & Dickinson-BD). Mice also received two doses, 0 and 48 h after immunization of 200 ng of Bordetella pertussis toxin intraperitoneally. All animals were followed daily and scores were given as followed: 0, no disease; 1, limp tail; 2, weak/partially paralyzed hind legs; 3, completely paralyzed hind legs; 4, complete hind and partial front leg paralysis; 5, complete paralysis/death.
For EAE induction into C57BL/6 mice, 12 animals were used being 4 euthanized at day 7 post MOG35–55 inoculation and 4 euthanized at day 16 post MOG35–55 inoculation in order to performed the protein expression and protein quantification analysis. The remaining four animals were used to evaluate the clinical score until later time points. For EAE induction into Rag1−/− mice, seven animals were used and all euthanized at day 16 post MOG35–55 inoculation. For euthanasia, a supra dose of anesthesia consisting of Ketamine-Xylazine (Agribrands do Brazil, São Paulo, Brazil) was injected i.p. in mice. The present study was carried out in accordance with the principles of the Basel Declaration and recommendations of National Animal Experimentation Control Council (CEUA) at Transplantation Immunology Laboratory, in the Department of Immunology, Institute of Biomedical Sciences, University of São Paulo, São Paulo, Brazil. The protocol was approved by the institutional Ethical Committee of the University of Sao Paulo, detailed at document number 50/2017.
T-cell differentiation conditions
Spleen were harvested and CD4+ naive cells were collected from donor C57Bl/6 (WT), NLRP3−/−, and CD4CreNLRP3flox mice using a negative selection kit (Miltenyi-Biotec). For differentiation of CD4+ cells, they were plated on in a flat-bottom 96-well plate with plated-bound anti-CD3 (3 µg/ml) and soluble anti-CD28 (2 µg/ml). Differentiation condition were employed as follows: Th1 IL-12 (10 g/ml), IFN-γ (10 ng/ml) and anti-IL-4 (200 µg/ml). Th2 IL-4 (10 ng/ml), anti-IFN-γ (200 µg/ml) and anti-IL-12 (200 µg/ml). Th17 TGF-β1 (1 ng/ml), IL-6 (50 ng/ml) and IL-23 (10 g/ml). All recombinant cytokines were purchased from Peprotech.
Generation of BMDCs
C57Bl/6 mice were euthanized and BM cells were flushed from femurs. Red blood cells were lysed, and cells were plated at a density of 1.0 × 106 cells/ml in DMEM high-glucose medium (Gibco) containing 10% FBS (Hyclone) and 20 ng/ml of GM-CSF. The culture medium was changed on day 5, and cells were harvested on day 7 to obtain DCs.
Cell sorting and T-cell co-culture with DC cells
Splenocytes from donor C57Bl/6 (WT), NLRP3−/− and CD4CreNLRP3flox mice were stained with the following fluorochrome-conjugated antibodies: PerCP anti-CD4, phycoerythrin (PE) anti-CD44 and Pacific Blue anti-CD62L. Primary effector CD4+CD44+CD62L−CD25+ cells were sorted (BD FACSAria II, BD Biosciences) and cell purity was greater than 98%. For some experiments, naive T cells were plated at a density of 2.0 × 105 cells/well in a flat-bottom 96-well plate and co-cultured in DMEM low-glucose medium containing 10% FBS for 5 days with the 1.0 × 105 BMDCs and MOG35–55 peptide (MEVGWYRSPFSRVVH LYRNGK).
Flow cytometry and intracellular cytokine staining
Cells were resuspended in PBS supplemented with 2% FBS and stained with a subset of the following mAbs: APC anti-CD4 and FITC anti-CD25 (BD Biosciences). Cells were acquired using an LSR II flow cytometer (BD Biosciences), and the results were analyzed with FlowJo 8.7 software (Treestar). To determine the number of Th1, Th2 and Th17 cells, 1 × 106 cells were stimulated in vitro for 4 h at 37°C in 5% CO2 with phorbol-12-myristate-13-acetate (PMA, 100 ng/ml), ionomycin (1 μg/ml) and brefeldin-A (1 μg/ml) (Sigma-Aldrich). We have permeabilized the cells using the BD Cytofix/Cytoperm Fixation/Permeabilization solution kit (BD Biosciences). Intracellular staining was performed with Tbet (PE), IFN-γ (PerCP), IL-17 (APC), IL-4 (PE) and IL-13 (APC) (eBioscience).
An ELISA assay for IFN-γ and IL-17 was used to measure the concentration of cell culture supernatant and the tissue, according to manufacturer’s instructions (R&D Systems, U.S.A.). The results were determined in pg/ml.
IL-4, IL-5, IL-10 and IL-12 (p40) were measured by Cytometric Bead Array (CBA) combined with flow cytometry (BD, Biosciences, CA, U.S.A.). Briefly, the cytokine capture beads were incubated with 50 ml of tissue homogenate or cellular supernatant and mixed with PE-conjugated detection antibodies. Cytokine concentrations were determined on a FACSDiva, (BD Biosciences Pharmingen) with the detection limits of 1.9 pg/ml. Results were analyzed with BD FCAP analysis software MACS OS version 9 (BD Bioscience). The results were determined in pg/ml.
Real-time changes in extracellular acidification ratio (ECAR) were evaluated in XF-96 Extracellular Flux Analyzer (Seahorse Bioscience) under basal conditions and after treatment with glucose at 10 mM, oligomycin at 1 µM and 2-deoxu-D-glucose (2-DG) at 100 mM.
Immune response gene array
Rag-1−/− mice were repopulated with naive CD4+CD44−CD62L+ T cells from WT, NLRP3−/− or CD4CreNLRP3flox mice. The animals were inoculated with MOG35–55 emulsified in CFA and 16 days later, their final portion of spinal medulla was extract and total RNA was isolated using TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. A total of 1 mg was then reverse transcribed using the First Strand Synthesis Kit (Thermo Fisher Scientific, Waltham, MA U.S.A.) and subsequently loaded onto immune response RT profiler array according to the manufacturer’s instructions (Thermo Fisher Scientific, Waltham, MA U.S.A.). The qPCR array analyses (n=4 per group) were performed in sets of 96-well plates, following the manufacturer’s recommendations (TaqMan™ Array, Mouse Immune, cat. number 4418724, Thermo Fisher Scientific, Waltham, MA U.S.A.). Gene expression levels were considered significantly different if P<0.05 and assuming a false discovery rate (FDR) of 20%. Plots were constructed in the GraphPad PRISM® 5 software, using the 2−ΔCT values for each gene corrected by the housekeeping Gapdh. Comparative heat maps, and fold change was calculated by determining the ratio of mRNA levels to control values using the D threshold cycle (Ct) method (2ΔDDCt). All data were normalized to an average of five housekeeping genes, Gusb, Hprt1, Agtr2, Gapdh and Actb. PCR conditions used: hold for 10 min at 958°C, followed by 45 cycles of 15 s at 958°C and 60 s at 608°C.
Experiments were performed in duplicate or triplicate and at least two independent experiments were performed for each assay. The data are presented as the mean ± S.E.M. Differences among groups were compared using a nonparametric ANOVA (Bonferroni post-test), and the differences between two groups were compared with a nonparametric Student’s t-test (Mann–Whitney). All statistical analyses were performed using GraphPad PRISM® 5 software, and the differences were considered significant when P<0.05, P<0.01 or P<0.001, according to the figure.
We have initially evaluated the clinical score of mice lacking NLRP3 (NLRP3−/−), mice with gain-of-function on NLRP3 into TCD4+ lymphocytes (CD4CreNLRP3fl/fl) and WT mice. As observed in Figure 1A, both NLRP3−/− and CD4CreNLRP3fl/fl animals presented reduced clinical score when compared with WT animals in the peak of the disease (from day 14 to 18). Similarly, both NLRP3−/− and CD4CreNLRP3fl/fl animals presented reduced mean peak score when compared with EAE control group (Figure 1B). Additionally, we have analyzed the IFN-γ and IL-17 production in the spleen at day 7 post-MOG35-55 inoculation and in the terminal spinal cord (CNS) at day 16. Despite no differences in the splenic IFN-γ mean fluorescence intensity (MFI) among the groups (Figure 1C), the percentage of T CD4+ cells expressing IFN-γ in the spleen and in the CNS of CD4CreNLRP3fl/fl animals were reduced when compared with WT mice (Figure 1D,E and Supplemental Figure S1). This was further corroborated as the protein levels of IFN-γ in the CNS was also reduced in CD4CreNLRP3fl/fl when compared with WT mice (Figure 1F). Moreover, the percentage of CD4+IFN-γ+ as well as IFN-γ secretion in the spinal cord was also reduced in NLRP3−/− mice when compared with WT animals (Figure 1E,F and Supplementary Figure S1). There were no differences in splenic IL-17 MFI among the groups (Figure 1G). On the other hand, both NLRP3−/− and CD4CreNLRP3fl/fl mice exhibited reduced IL-17- expressing CD4+ T cells in the spleen at day 7 and in the CNS at day 16 after MOG35–55 (Figure 1H,I and Supplementary Figure S1). However, we did not observe differences in IL-17 protein production in the CNS (Figure 1J) at day 16. We also searched for Th1 and Th2-related cytokines at 16 days after MOG inoculation in the final third of spinal medulla. We observed that CD4CreNLRP3fl/fl mice exhibited reduced IL-12 (p40) levels when compared with WT animals (Figure 1K). Additionally, CD4CreNLRP3fl/fl mice present increased levels of IL-4 (Figure 1L), IL-5 (Figure 1M) and IL-10 (Figure 1N) when compared with NLRP3−/− ones. There were no differences in IL-1β production among the groups (data not shown). Such data clearly evidences the role of NLRP3 in the development of EAE clinical scores. Interestingly, either the absence of NLRP3 in all cells or the gain-of-function only in CD4+ T cells predicts a better scenario in the EAE model.
Both NLRP3 deficiency and NLRP3 gain-of-function in T CD4+ cells predict a better outcome in the context of EAE
As an increased Th2-related immune response and reduced Th1 and Th17 pattern predicts a protection in the context of EAE, we investigated the specific role of T cells derived from WT, NLRP3−/−, and CD4CreNLRP3fl/fl animals. For that, we have differentiated splenocytes-derived Th0 naive precursors into Th1, Th2 and Th17 cells in vitro. Despite the lack of difference in the percentage of CD4+ T cells expressing IL-17 among the groups regardless of the differentiation pattern, we observed that IFN-γ-expressing CD4+ T cells were decreased in CD4CreNLRP3fl/fl animals when compared with WT under Th1 differentiation conditions (Figure 2). This was correlated with a reduced Tbet protein expression when compared with cells derived from WT mice (Figure 3). It is noteworthy that there was no difference in the IFN-γ-expressing CD4+ T cells when we compared CD4CreNLRP3fl/fl and NLRP3−/− animals under Th1 stimuli (Figure 2); however, a balance between Th1 and Th2-related cytokines could lead to different immune responses. In this sense, we searched for IL-14 and IL-13. As expected, CD4+ T cells from CD4CreNLRP3fl/fl animals exhibited increased expression of IL-4 (Figure 4) and increased expression of IL-13 (Figure 5) when compared with cells derived from WT mice under Th2 differentiation. Nevertheless, there were no differences in the percentage of IL-4 and IL-13 between NLRP3−/− and WT mice. To analyze the impact of these findings in an antigen-specific context, we loaded WT DCs with MOG35–55 peptide and cultured them for 5 days with primary effector CD4+CD44+CD62L−CD25+ T cells from WT, NLRP3−/− or CD4CreNLRP3fl/fl mice immunized with MOG35–55 for 7 days. As observed, the production of IFN-γ (Figure 6A) and IL-17 (Figure 6B) was decreased in the context of CD4CreNLRP3fl/fl cells when compared with NLRP3−/−, despite no differences when compared with WT CD4+ T cells. Additionally, the production of IL-10 (Figure 6C) was increased in the context of CD4CreNLRP3fl/fl cells when compared with WT or NLRP3−/− CD4+ T cells. On the other hand, there was no detectable IL-4 and IL-5 production (data not shown). Altogether, these results suggest that gain-of-function of NLRP3 in CD4+ T cells leads to an increased Th2 immune response, in opposition to a Th1 and a Th17 immune pattern.
NLRP3 gain-of-function in T CD4+ cells leads to reduced IFN-γ production under Th1-related polarization
NLRP3 gain-of-function in T CD4+ cells leads to reduced Tbet production under Th1-related polarization
NLRP3 gain-of-function in T CD4+ cells leads to increased IL-4 production under Th2-related polarization
NLRP3 gain-of-function in T CD4+ cells leads to increased IL-13 production under Th2-related polarization
NLRP3 gain-of-function in T CD4+ cells leads to reduced IFN-γ and IL-17 production in an antigen-specific context
In an attempt to investigate whether CD4CreNLRP3fl/fl mice exhibit reduced clinical scores due to gain-of-function of NLRP3 in CD4+ T cells and not due to deletion of NLRP3 in all other cells but T lymphocytes (inherent to the genetic model), we repopulate Rag-1−/− mice with naive CD4+CD44−CD62L+ T cells from WT or CD4CreNLRP3fl/fl and five days later we subjected the animals to the EAE protocol. Again, Rag-1−/− mice repopulated with CD4CreNLRP3fl/fl CD4+ T exhibited a decreased clinical score (Figure 7A) and a reduced mean peak score (Figure 7B) when compared with animals repopulated with WT cells. Moreover, IFN-γ production in the terminal spinal cord at day 16 post-MOG inoculation was reduced in Rag-1−/− mice repopulated with CD4+ T from CD4CreNLRP3fl/fl when compared with those repopulated with CD4+ T from WT (Figure 7C). Additionally, IL-10 levels were increased in Rag-1−/− mice repopulated with CD4+ T from CD4CreNLRP3fl/fl when compared with those repopulated with CD4+ T from WT (Figure 7D). During the peak of the disease, splenocytes of both animals were removed and CD4+ T cells were sorted for further metabolic studies. The ECAR analysis between CD4+ T cells from WT and CD4CreNLRP3fl/fl mice was performed following glucose, oligomycin and 2-DG stimuli (Figure 7E). In comparison to CD4+ T cells from Rag-1−/− mice repopulated with WT cells, CD4CreNLRP3fl/fl CD4+ T lymphocytes exhibited reduced glycolysis (Figure 7F), glycolytic capacity (Figure 7G) and reduced glycolytic reserve (Figure 7H), a metabolic profile compatible with Th2 cells, when compared with Th1 cells .
Rag1−/− mice repopulated with CD4CreNLRP3fl/fl T cells presents a better outcome and a reduced Th1-related response in the context of EAE, when compared with WT CD4+ T cells repopulation
Because EAE development is directly related to T helper immune response, we evaluated the expression pattern of 84 genes involved in adaptive immune response in the terminal spinal cord tissue of each repopulated Rag-1−/− mice (with WT, NLRP3−/− or CD4CreNLRP3fl/fl mice) at day 16 post-MOG inoculation (Figure 8A). We found 12 differently expressed genes among the three groups (Figure 8B). Among the differently expressed genes, we observed that Rag-1−/− mice repopulated with naive CD4+ T cells from both NLRP3−/− and CD4CreNLRP3fl/fl animals expressed increased mRNA levels of chemokines, such as Cxcl10 and Ccl12 when compared with mice repopulated with WT lymphocytes. Moreover, when compared with mice repopulated with NLRP3−/− cells, those repopulated with CD4CreNLRP3fl/fl ones exhibited increased mRNA levels of Spp1, and reduced mRNA levels of Ccr1, Ccl4, Ccl19, Cxcl9, Ccr2, Il2rg and Il1r2 (Figure 8B).
Spinal cord gene expression profiling of Rag-1−/− mice repopulated with naive T CD4 lymphocytes from WT, NLRP3−/− or CD4CreNLRP3fl/fl animals
We performed a functional enrichment analysis to reveal enriched functions among the 12 differently expressed genes (Figure 8C). We observed that ‘cytokine–cytokine receptor interaction’, ‘chemokine signaling pathway’ and ‘Toll-like receptor signaling pathway’ were the most altered pathways in this analysis (Figure 8C). Additionally, the gene ontology (GO) knowledgebase of biological processes revealed that the inflammatory response was the most enriched pathway among the three different repopulated mice (Figure 8D). Moreover, leukocyte chemotaxis pathways were also enriched among these genes (Figure 8D). The analysis of a protein interaction network related to these genes demonstrated they participate in an inflammatory network centered on Cccl4, Jun and Cxcl9 (Figure 8E). It is predicted that the gain-of-function in NLRP3 in CD4+ T cells is sufficient to protect against the development of EAE, a feature that is accompanied by altered chemotaxis pattern and reduced Th1 response. Thus, our data evidence the role of NLRP3 on CD4 cells regulating its transcriptional profile toward a Th2 phenotype, that further protects mice from EAE.
Adaptive immune responses are dictated by the overall signals provided during T-cell differentiation, tuning the immune response to a more adequate response. In this context, the Th1/Th2 dichotomy was expanded after the discovery of IL-17-producing T cells in the context of autoimmune disorders in the brain . Since then, the role of Th17 cells in the pathogenesis of CNS inflammatory-induced demyelination has been well described , although both Th1 and Th17 lymphocytes are responsible for tissue lesions, whereas Th2 and Treg cells predict a more protective scenario. It has already been demonstrated that the absence of NLRP3 inflammasome (Nlrp3−/− mice) leads to protection in the context of EAE mainly due to the abrogation of IL-1β and IL-18 secretion during pyroptosis . These cytokines induce the expression of chemokine receptors on T cells, facilitating their migration through CNS , as well as promoting Th1 and Th17 differentiation . Nevertheless, despite the close link between IL-1β- and Th17-related response , there were no differences in IL-1β among the studied groups in the present study. This result could be attributed by the fact that different cell types are able to produce IL-1β, including microglia , mast cells , CD4+, CD8+ and γδ T cells . Additionally, such production can also be Nlrp3-independent [31,32]. Besides reduced IFN-γ and IL-17 production, Nlrp3−/− mice exhibited improved histology in the spinal cords with reduced destruction of myelin and astrogliosis . This protection was accompanied by a significant reduction of infiltrating macrophages, dendritic cells, CD4+ and CD8+ T cells in the CNS . On the other hand, it has been also reported that aggressive immunization of mice using high doses of heat-killed mycobacteria in association with MOG35–55, as an alternative EAE protocol, was sufficient to induce severe EAE in Nlrp3−/− mice . Additionally, doubling the amount of MOG35–55 emulsified in complete Freund’s adjuvant injected on day 0, and boosting such injection on day 7 leads to absence in the protection of Nlrp3−/− mice against the EAE development . Although further studies are required, it is possible that strong activation of the immune system substantially impacts the development of EAE in a scenario independent of NLRP3.
On the other hand, little is known about the role of NLRP3 as a transcription factor specifically in T cells, especially during autoimmune diseases, such as MS. Thus, here we investigated the role of NLRP3 by crossing CD4Cre mice with NLRP3flox/flox generating the CD4CreNLRP3flox/flox mice whose main characteristic is the gain-of-function of NLRP3 in TCD4+ cells, despite presenting an NLRP3 deletion in all other cells. It is relevant to note that during T-cell thymic development, TCD8+ cells once expressed CD4 (during the double positive stage) and, therefore, TCD8+ lymphocytes also show a gain-of-function of NLRP3. We observed that CD4CreNLRP3flox/flox mice had milder clinical EAE scores, probably due to the reduced CD4+ T cells expressing IFN-γ and IL-17 during both the induction phase of EAE (at 7-day post-MOG inoculation) and during the effector phase of the disease (at day 16 post-MOG inoculation—the peak of the disease) . To corroborate this, and to investigate whether CD4CreNLRP3fl/fl mice are protected due to NLRP3 deletion in non-lymphoid cells, as previously demonstrated [4,33], or due to gain of function of NLRP3 in lymphocytes, we repopulated Rag-1−/− mice with T CD4 lymphocytes from either WT, NLRP3−/− or CD4CreNLRP3fl/fl animals following EAE induction.
Interestingly, our results demonstrate that NLRP3 gain-of-function in CD4+ T cells significantly reduced EAE scores by reducing the Th1 and Th17 responses and promoting Th2 and Treg differentiation of CD4 cells. This is consistent with previous data pointing to NLRP3 and IRF4 as transcription factors that lead to IL-4 production [7,37], a well-described cytokine related to Th2 immune pattern. IL-10 levels, a cytokine described as being involved in the suppression of EAE [38,39], were also increased in the context of NLRP3 gain-of-function in CD4+ T cells. It has been also demonstrated that IL-10-deficient mice are more susceptible and develop a more severe EAE . NLRP3 is not only a key inflammasome component, but also a transcription factor that controls Th2 cell polarization. Moreover, NLRP3 also exerts a Th2-related response in leishmaniosis context , as well in allergic airway inflammation model . It has been demonstrated that NLRP3 inflammasome signaling induces a Th2-biased cytokine profile in the footpads of L. major-infected mice through the induction of IL-4, IL-5 and the reduction of IFN-γ production . Also, the NLRP3 inhibitor MCC950 leads to reduced IL-4 and IL-13 production, as well as reduced eosinophil infiltration in OVA-induced inflammatory model of allergy . These data were corroborated by experiments in which the absence of NLRP3 released all symptoms of allergic airway inflammation . Controversially, it is important to notice that NLRP3 has also been related to a Th1 response in human autoinflammatory disease and in mouse models of inflammation, however, in a scenario dependent on the complement protein C5 . In this scenario, mice injected with NLRP3−/− CD4+ T cells developed a more severe colitis despite a less pronounced IFN-γ production, when compared with mice that had received WT CD4+ T cells . Moreover, NLRP3 inflammasome has a crucial role in promoting the expansion of Th1/Th17 immunity in mice infected with Paracoccidioidomycosis brasiliensis , and Trichuris trichiura , both worms present in gastrointestinal tract. It is possible that gain-of-function mutations in NLRP3 led to increased levels of IL-1β and IL-18 due to the activity of NLRP3 into myelocytic cells, and these cytokines amplified the Th-1-driven immune response.
We observed an altered mRNA expression pattern among Rag-1−/− mice repopulated with naive T CD4 lymphocytes from WT, NLRP3−/− or CD4CreNLRP3fl/fl animals, mainly in the genes related to chemokines and chemokine receptors. It is possible that mice repopulated with NLRP3−/− cells present a late migratory pattern when compared with both WT or CD4CreNLRP3fl/fl groups, confirmed by the reduced chemokine expression. Nevertheless, we observed no differences in il4, il5 and il13 mRNA expression, a result that could be explained since these cytokines can be produced by cells other than lymphocytes, masking the T helper-related response of T CD4+ cells with different NLRP3 activities. Moreover, Rag-1−/− animals repopulated with CD4CreNLRP3fl/fl cells, in comparison with those repopulated with NLRP3−/− ones, present a reduced expression of il1r2, a protein that binds to IL-1α, IL-1β and IL-1Ra, preventing them from binding to their regular receptors . This mechanism could be added to the intrinsic protection related to the gain-of-function of NLRP3 into the T CD4+ cells and the consequent Th2-biased response against EAE development. Interestingly, animals that received CD4CreNLRP3fl/fl lymphocytes also presented increased mRNA expression of Spp1 when compared with those receiving NLRP3−/− cells, a molecule also known as osteopontin and previously shown to have a Th1 cytokine function in cell-mediated immunity . On the other hand, osteopontin seems to act as a double-edged sword triggering neuronal toxicity and death in some contexts and functioning as a neuroprotective in others . It is important to notice that Rag-1−/− animals repopulated with CD4CreNLRP3fl/fl cells, in comparison with those repopulated with NLRP3−/− ones, had a reduced expression of Ccr2 and Ccr3. Despite sometimes controversial and dependent of the utilized model of disease [49,50], both chemokine receptors are related to a Th1-related immune response [49,51]. CCR2 is, for example, required for Langerhans cell migration and localization of Th1-inducing dendritic cells . Moreover, CCL12, a chemokine increased in the context of CD4CreNLRP3fl/fl T cells repopulation is associated with a Th2-related response in the context of lung fibrosis  and cancer . However, more studies are required to elucidate the dependency of NLRP3 for the dynamic of chemokines and their receptors expression and the consequent dynamic of leukocyte infiltration into the CNS.
Additionally, cellular metabolism has emerged as an exciting field in immunology, since different metabolic pathways are involved in different stages of the immune cells’ activation and differentiation, including T lymphocytes. To date, it has been demonstrated that inflammatory M1 macrophages display enhanced glycolytic metabolism and reduced mitochondrial activity . Conversely, anti-inflammatory M2 macrophages show high mitochondrial oxidative phosphorylation and are characterized by an enhanced spare of respiratory capacity . Increased glycolysis or diminished mitochondrial metabolism can also boost the differentiation of pro-inflammatory T-effector cells . Glycolysis promotes expression of effector molecules including the Th1-related cytokine IFN-γ . When activated T cells are provided with co-stimulation and growth factors but are blocked from engaging glycolysis, their ability to produce IFN-γ is markedly compromised . In a separate report, CD4+ T cells deficient in LDHA, a critical enzyme converting pyruvate to lactate and therefore dictating aerobic glycolysis , had defects in IFN-γ production, which originated from lack of acetylation of the IFN-γ locus . Consistently, here we demonstrate that NLRP3 gain-of-function in T cells led to decreased levels of glycolysis compared with WT lymphocytes, thus promoting the subsequent polarization into Th2 profile. However, further studies are needed in order to investigate the real relationship between NLRP3 and metabolic pathways in T lymphocytes, and whether this has an effective role in protection against EAE. Altogether, our data corroborate previous findings showing that NLRP3 may act as a transcription factor that promotes Th2 differentiation. To our knowledge, this is the first time that this approach was used in a model of autoimmune disease, as EAE. Although more research is needed, our data highlight for the importance NLRP3 signaling pathway as a potential therapeutic target for autoimmune diseases.
MS is an unpredictable, often disabling disease of the CNS. Its symptoms vary widely and depend on the kind and the length of the accumulated lesions. MS can be divided into four different types according to its clinical progression.
The present paper shows that NLRP3 acts as a transcription factor involved in the Th2-related response, an immune pattern involved in the protection against MS. Moreover, our data indicate that gain-of-function of NLRP3 specifically into CD4+ lymphocytes leads to increased IL-4 and IL-13 production and a metabolic cellular status compatible with Th2-related cells.
Specific activation of NLRP3 into CD4+ T cells could lead to a protective scenario.
Altogether, our data highlight the importance NLRP3 signaling pathway as a potential therapeutic target for MS.
We thank F.P. from Centro de Facilidades de Apoio a Pesquisa (CEFAP- USP) for help with the cell-sorting experiments and Professor V.C. for providing us NRLP3−/− mice.
Conceptualization, T.T.B. and N.O.S.C.; Methodology, T.T.B., W.N.B., H.A., F.F.T., A.C.L.M., F.V.P., M.I.H. and V.A.-O.; Review & Editing T.T.B., J.P.S.P. and N.O.S.C.; Resources, N.O.S.C.; Supervision, N.O.S.C.
The authors declare that there are no competing interests associated with the manuscript.
The present study was supported by the São Paulo State Funding Agency (FAPESP) [grants numbers: 2014/06992-8, 2017/05264-7 and 2017/22504-1], Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) and in part by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - Brasil (CAPES) - Finance Code 001.
cryopyrin-associated autoinflammatory syndrome
Cytometric Bead Array
central nervous system
encephalomyelitis autoimmune disease
extracellular acidification ratio
mean fluorescence intensity
Nucleotide-binding domain, leucine-rich repeat containing protein
- NLR3 inflammasome
NLR Pyrin-domain-containing 3