All experiments were conducted in compliance with the IACUC guidelines of Virginia Tech. Nlrp12-deficient B6/lpr was generated by cross-breeding B6.Nlrp12^-/- (42) with B6.Fas^lpr/lpr mice (The Jackson Laboratory, Bar Harbor, ME). Offspring were genotyped for both the Nlrp12 locus and Fas^lpr mutation ( Figure S1 ). We monitored the disease progression in both female and male mice housed under specific pathogen-free environment in an AAALAC accredited animal facility at Virginia Tech. All factors including housing, handling, light cycle (12-hour light/dark) were consistent for all mice, which received the hormone-free NIH-31 Modified 6% Mouse/Rat diet. Food and water were provided ad libitum.

The development of lupus nephritis was assessed through weekly testing of proteinuria levels. Weekly urine samples were collected, and proteinuria levels were measured using a Pierce Coomassie Protein Assay Kit (Thermo Scientific) as we previously described (55). Additionally, upon euthanasia at 39 weeks of age, kidneys were harvested to determine the deposition of immune complexes in the renal compartments through immunohistochemical staining for IgG as described below. Renal deposition of complement C3 was also determined with immunohistochemical staining.

Endpoint serum samples were sent to the Ligand Assay & Analysis Core of the Center for Research in Reproduction (CRR) at the University of Virginia for measurement of testosterone levels. Mouse serum testosterone levels were determined using Testosterone Mouse & Rat ELISA (IBL America) following the manufacturer’s recommendations.

Total splenocytes and bone marrow (BM) cells were isolated and red blood cell exclusion was achieved following our previously published protocols (55). Both Splenocytes and BM cells were analyzed using flow cytometry as described below. Furthermore, for in vitro generation of BM-derived myeloid cells, BM cells from femurs were cultured at a density of 10⁶ cells/ml for 6 days in complete RPMI medium (RPMI 1640 supplemented with 10% fetal bovine serum, 1 mM sodium pyruvate, 1% 100 MEM non-essential amino acids, 10 mM HEPES, 55 μM 2-mercaptoethanol, 2 mM L-glutamine, and 100 U/ml penicillin–streptomycin; all from Life Technologies, Grand Island, NY) supplemented with 10 ng/ml recombinant murine GM-CSF (PeproTech) and cultured for 6 days as previously described (56). For in vitro stimulation of BM-derived myeloid cells, cultures were treated with 50 ng/ml or 1 μg/ml lipopolysaccharide (LPS; eBioscience) for four hours before analysis. At the end of the stimulation period, cells were harvested for both flow cytometry and RT-qPCR analysis whereas the supernatants were collected for ELISA.

Cells were initially blocked with anti-mouse CD16/32 (eBioscience) then stained with fluorochrome-conjugated antibodies following our previously published procedures (55). Zombie Aqua (BioLegend) staining was performed to exclude dead cells. For quantification of B cells in total splenocytes, the following anti-mouse antibodies were used: CD19-Pacific blue, CD27-PE, CD138-APC-Cy7, CD44-PerCP-Cy5.5, IgD-PE-Cy7, GL7-AF647. For splenic T cells, CD3-APC, CD4-FITC, CD8-PE, CD44-PerCP-Cy5.5, CD62L-APC-Cy7, CD69-Pacific blue, CXCR5-PerCP-Cy5.5, and PD-1-APC-Cy7 (BioLegend) were used. For myeloid cell analysis, the following anti-mouse antibodies were used: CD11b-PE, CD11c-PerCP-Cy5, F4/80-PE-Cy7 (BioLegend), Gr1-V540 (BD Bioscience). Analysis was performed with a BD FACSAria II flow cytometer (BD Biosciences). Flow cytometry data were analyzed with FlowJo.

Spleen and kidney were harvested at the endpoint and embedded in Tissue-Tek OCT Compound (Sakura Finetek) and rapidly frozen in a freezing bath of dry ice and 2-methylbutane. Frozen OCT samples were cryosectioned and unstained slides were stored at −80°C. Immunohistochemical staining procedures were performed as we previously described (55). Splenic sections were stained for germinal center (GC) formation using the following anti-mouse antibodies: CD3-APC, IgD-PE, GL7-FITC (BioLegend). Renal immune complex deposition was determined using anti-IgG-PE (eBioscience) and anti-C3-FITC (Cedarlanelabs, Burlington, Canada). Slides were mounted with Prolong Gold containing DAPI (Life Technologies). Pictures were visualized with both an EVOSVR FL microscope (Advanced Microscopy Group, Grand Island, NY) and a Zeiss LSM 880 confocal microscope (Zeiss,USA, Fralin Imaging Center, Virginia Tech). Image processing and quantification of the fluorescent intensity were performed with ImageJ and ZEN 2.1 Lite software. Sections from at least 3 mice per group were quantified and the unit used for calculation was “integrated density score.”

Total RNA extraction was performed from snap-frozen pre-weighed splenic tissue or snap-frozen cultured cells as we previously reported (55, 57). Tissues or cell pellets were homogenized in Qiazol lysis reagent using TissuelyserII homogenizer (Qiagen). Total RNA was isolated using RNeasy Plus Universal Kit (Qiagen) with the elimination of gDNA. Reverse transcription (RT) was carried out using iScript™ Reverse Transcription Supermix (Bio-Rad). Quantitative PCR (qPCR) was performed utilizing the Fast SYBR^® Green Master mix and the ABI 7500 Fast Real-Time PCR System (Applied Biosystems). Relative transcript quantities were calculated using the 2−ΔΔCt method and normalized to the level of the 18S rRNA housekeeping gene level. Primer sequences for mouse Bcl6, Prdm1/Blimp1, Tnfsf13b/BAFF, Il21, Tnf, Il1β, Cxcl13, Ccl19/MIP-3β, Ccr7, and androgen receptor are available in Table S1 .

Serum samples were obtained at euthanasia, and aliquots were stored at −80°C until processing. Anti-doubles stranded (ds)DNA IgG antibodies were determined following our previously reported procedures (55). Serum BAFF, IL-6 and IFNγ were determined using ProcartaPlex™ Multiplex Immunoassay (Invitrogen) following manufacturer’s procedures and the data were acquired and analyzed using the Luminex FlexMAP3D™ system (Chicago, USA). For culture supernatants, TNFα was determined using mouse TNFα ELISA MAX kit (BioLegend) following manufacturer’s procedures.

Fecal microbiota samples from each mouse at the indicated time points were obtained by taking a mouse out of the cage and collecting a fecal pellet. To avoid cross-contamination, each microbiota sample was collected by using a new pair of sterile tweezers. Samples were stored at −80°C. Similarly, at euthanasia, different intestinal sections (duodenum/jejunum, ileum, and colon) were recovered immediately, and the contents of each section were separately collected by manual extrusion and frozen immediately at −80°C until use. All samples were processed at the same time. Sample homogenization, cell lysis, and DNA extraction were performed as previously described (55, 58). For 16S rRNA sequencing, the V4 region (ca. 252 bp) of 16S rRNA gene was PCR amplified with 515F and 12-base GoLay barcoded 806R primers (59). The purified amplicons were sequenced bidirectionally (150 bp PE chemistry) on an Illumina MiSeq at Argonne National Laboratory. Samples were analyzed using the R package phyloseq (60). Reads were processed and amplicon sequence variants (ASVs) were generated using DADA2 in R. Reads were quality trimmed and filtered using the command fastqPairedFilter with parameters truncLen=c(140,140), maxEE=c(2,2), rm.phix=TRUE, maxN=0, compress=TRUE, multithread=FALSE. DADA2 was used to learn error rates, perform sample inference, dereplicate and merge paired-end reads, and construct a sequence table (61). Taxonomy was assigned using the SILVA 138 ribosomal RNA (rRNA) database training set (62) using the DADA2 functions, assignTaxonomy and addSpecies. A total of 3327 ASVs were detected in 212 total samples. ASVs seen fewer than three times in at least 20% of samples and samples with fewer than 1000 reads were removed from the dataset, resulting in 205 samples and 187 ASVs used for downstream analyses. ASVs were aggregated at the genus level using the phyloseq function tax_glom. Counts were used for alpha diversity and differential abundance tests, while proportions were used to calculate Bray-Curtis dissimilarity. Differentially abundant and variable taxa between groups were identified using the function differentialTest in corncob (63) and significance was assessed using a Wald test with an FDR cutoff of 0.05. Shannon diversity was calculated using the DivNet (64) functions divnet and testDiversity. Bray-Curtis distances were calculated using the phyloseq function ordinate, specifying “method=“NMDS”, distance=“bray”, trymax=1000”. Significance was assessed using the adonis test in the vegan package with 999 permutations.

Student’s t test was employed for the comparison between two groups. For in vitro culture data involving more than 2 groups, two-way ANOVA with Sidak’s multiple comparison test was employed. Data are shown as mean ± standard error of the mean (SEM). Significant differences were shown as *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. All analyses were performed with Prism GraphPad.

Since sex differences exist (65), where females are more generally affected with autoimmune disease (66–68), to investigate the roles of NLRP12 in modulating inflammation in the B6/lpr model of autoimmunity, we monitored the disease progression in both male and female mice. Interestingly, the deficiency of Nlrp12 did not alter disease progression in female mice ( Figures S2A, B ). In contrast, while splenomegaly was not affected (data not shown), the gene deletion significantly mitigated several hallmarks of lupus disease in male B6/lpr mice, including reduced proteinuria levels ( Figure 1A ), decreased circulatory levels of anti-dsDNA antibodies ( Figure 1B ), and reduced deposition of IgG and complement C3 in renal compartments ( Figures 1C, D ), indicating sex-specific effects of NLRP12 in modulating lupus pathogenesis. Interestingly, we found a trending increase in both serum testosterone level ( Figure S2C ) and the splenic transcript level of androgen receptor ( Figure S2D ) in male Nlrp12-/- B6/lpr compared to Nlrp12+/+ (WT) B6/lpr mice, suggesting a potential role for sex hormones.

Notably, we monitored male mice from 24 to 39 weeks of age. WT B6/lpr mice generally develop systemic autoimmunity without significant clinical pathology of renal inflammation or nephritis, which was confirmed in our studies. However, Nlrp12-/- B6/lpr mice exhibited even lower proteinuria levels that were significantly different from WT B6/lpr mice during the earlier time window from 24 to 31 weeks of age ( Figure 1A ), while we could not detect differences in proteinuria level during the later period from 32 to 39 weeks of age ( Figure S1E ).

Together, these findings indicate that NLRP12 might have pathological roles in modulating systemic autoimmunity in male B6/lpr mice. From now on, we will focus on describing male mice unless noted otherwise.

We detected a significantly lower level of autoantibodies and their renal deposition in the absence of NLRP12. Therefore, to delineate the mechanisms through which Nlrp12 deficiency protects against inflammation in our autoimmune model, we investigated its effects on B cell responses. Deficiency of Nlrp12 suppressed B cell responses in male B6/lpr mice ( Figure 2 ). Nlrp12^-/- B6/lpr mice had significantly reduced plasma cells (gated as CD19^-CD27^-CD138⁺IgD^-) to plasmablasts (gated as CD19 ^+/low CD27 ^+/low CD138⁺IgD^-) ratio in total splenocytes ( Figure 2A ), suggesting a blockade right before terminal differentiation of B cells. This is consistent with a significant reduction of the splenic transcript level of Prdm1 ( Figure 2B ). Interestingly, we also found a nearly significant reduction of the transcript levels of the master regulator of the germinal center (GC) reaction, Bcl6, in splenic tissues of Nlrp12-deficient mice ( Figure S3A ). Although GC formation shown as GL7 staining in immunohistochemically stained splenic sections was not different ( Figure 2C , Figure S3B ), there was a significant reduction of the percentage of GL7⁺ cells in total CD19⁺ splenic B lymphocytes ( Figure 2D ; gating strategy is shown in Figure S3C ). Moreover, we found that Nlrp12-deficient mice had a significantly reduced percentage of splenic T follicular helper (Tfh) cells ( Figure 2E ; gated as CXCR5⁺PD-1⁺CD4⁺CD3⁺ in Figure S3D ), as well as significantly reduced staining of CD3⁺ cells in the GCs ( Figures 2C , S3E ). Notably, the percentages of Tfh cells were low and highly variable in the WT mice, and the deficiency of NLRP12 further decreased the frequencies of these cells. Furthermore, while we only found a trending reduction of serum IL-21 ( Figure S3F ), a major cytokine produced by Tfh cells (69), we detected a significant reduction in its splenic transcript level in Nlrp12-deficient mice ( Figure 2F ). These results indicate that Nlrp12 deficiency might dampen GC reaction by suppressing the functions of Tfh cells. Finally, we found downregulated levels of factors assisting B cells (70) including the splenic transcript levels of the B cell chemoattractant Cxcl13 ( Figure 2G ) and the circulatory level of the B cell survival factor BAFF ( Figure 2H ) as well as its splenic transcript level ( Figure 2I ). These results indicate that Nlrp12 deficiency dampens terminal differentiation, GC reaction, and survival of potentially autoreactive B cells, which might be the reason for decreased production of autoantibodies and ameliorated autoimmune pathologies. Further studies will elucidate whether NLRP12 targets Bcl-6 and/or Blimp-1 to control autoreactive B cell responses.

Activation of NLR proteins can shape T cell differentiation and responses. For instance, activation of inflammasome-forming NLRs such as NLRP3 often results in the production of proinflammatory cytokines that could drive the differentiation of inflammatory T cells including Th1 and Th17 (71). However, the exact immunoregulatory functions of NLRP12 in modulating T cell differentiation and responses are still elusive (44, 54). Since T cells play pivotal roles in amplifying and maintaining inflammation particularly through activating autoreactive B cells (72), producing disease-promoting cytokines, and accumulating autoreactive memory (73), we sought to determine how Nlrp12 deficiency modulates the frequencies and responses of different T cell populations. Deficiency of Nlrp12 significantly reduced percentage of CD3⁺ T cells in total splenocytes ( Figure 3A ), consistent with the reduced fluorescence intensity of CD3⁺ T cells in immunohistochemically stained splenic GCs ( Figure S3E ). Nlrp12^-/- B6/lpr mice also had significantly fewer CD8⁺ ( Figure 3B ) and double negative (DN)-T cell ( Figure 3C ) percentages in total splenocytes, which possibly contributed to the reduction in CD3⁺ T cells. CD4⁺ T cell response did not change. Importantly, the generation of DN-T cells is one of the prominent alterations of T cell responses reported in SLE (8) and ALPS (74–76). These DN-T cells could have been generated from activated CD8⁺ T cells (74, 77–79). Moreover, we found a reduced proportion of CD44⁺CD62L⁻ effector memory T (T_EM) cells in the spleens of Nlrp12-deficient mice ( Figure 3D ). Together, these results suggest that Nlrp12 deficiency might target T cells to dampen autoimmunity in male B6/lpr mice.

NLRP12 can modulate the responsiveness of different myeloid cells including neutrophils, dendritic cells (DCs) and macrophages (40–42, 46, 47). We examined the immunophenotypic changes of these populations (see Figure S4A for gating strategies) in different lymphoid compartments including BM and spleen. We found no significant changes in neutrophils ( Figure S4B ; gated as CD11c^-CD11b⁺Gr1⁺) or DCs ( Figures S4C, D ; gated as CD11c^high CD11b⁺Gr1^- or CD11c^high CD11b⁺Gr1⁺). However, Nlrp12^-/- B6/lpr mice showed a significant reduction of Gr1^-F4/80⁺CD11b⁺CD11c^-/low macrophages as the percentage of total splenocytes ( Figure 4A ). In addition, as the percentage of BM macrophages slightly increased in Nlrp12^-/- B6/lpr mice ( Figure S5A ), the ratio of splenic-to-BM macrophages was significantly reduced with Nlrp12 deficiency ( Figure 4B ), suggesting decreased migration of these cells from BM to the spleen. Importantly, we also detected significantly reduced splenic transcript levels of Tnf ( Figure 4C ) and macrophage inflammatory protein 3-β (MIP-3β, gene name Ccl19; Figure 4D ). These data suggest that Nlrp12 deficiency might dampen pro-inflammatory responsiveness of splenic macrophages in autoimmune environment. Interestingly, following ex-vivo stimulation of BM-derived myeloid cells with LPS – a potent activator of macrophages (80) that could prime DCs (81) – although there were slightly more BM-derived macrophages with Nlrp12 deficiency regardless of stimulation status ( Figure S5B ), the percentage of BM-derived DCs in these cultures was significantly reduced with the deficiency ( Figure S5C ). This suggests decreased priming of DCs and thus reduced functional potential of BM-derived macrophages. In parallel, we detected a significantly reduced level of TNFα in the culture supernatants of BM-derived cells with Nlrp12 deficiency following LPS stimulation ( Figure 4E ). Similarly, LPS-stimulated BM-derived cells from Nlrp12^-/- B6/lpr mice had reduced transcript levels of Tnf ( Figure 4F , Figure S5D following 50 ng/ml and 1 μg/ml LPS stimulation, respectively) and Il1β ( Figure 4G , Figure S5E ), as well as Ccr7 ( Figure 4H , Figure S5F ), a receptor known to be expressed on DCs following their activation (82). These ex-vivo findings suggest a potential pathogenic role of NLRP12 in potentiating macrophages and DCs in response to pro-inflammatory triggers. Importantly, the change of IL-1β suggests that NLRP12 inflammasome may facilitate the production of IL-1β that in turn drives the production of other inflammatory mediators including TNF-α (83) and MIP-3β (84). Together, these results suggest reduced pro-inflammatory innate immunity with Nlrp12 deficiency in male B6/lpr mice.

Changes of microbiota dynamics have been shown to drive autoimmunity (32, 33) or modulate autoimmunity (28). This has been established for SLE (37, 38) but not yet for ALPS. Interestingly, NLRP12 could shape inflammatory outcomes through regulating the gut microbiota (49, 85). Thus, we investigated whether the alteration of Nlrp12 could implicate the gut microbiota in B6/lpr mice. We analyzed the fecal and intestinal microbiotas of both male and female mice seeking to answer the sex-dependent response to Nlrp12 deficiency. We found a clear distinction in the fecal microbiota diversity with or without NLRP12 ( Figure 5 ). Fecal microbiotas had significantly different alpha diversity on the genus level as shown by Shannon diversity estimate in both male ( Figure 5A ) and female ( Figure 5B ) B6/lpr mice, where Nlrp12 deficiency led to significantly increased microbiota diversity. However, the difference in alpha diversity was much more pronounced in male than female mice. Similarly, analysis of beta diversity based on Bray Curtis dissimilarity calculation showed significantly different overall taxonomic composition based on the genotype but not the timepoint between WT and Nlrp12^-/- B6/lpr male ( Figure 5C ) and female ( Figure 5D ) mice. Moreover, the composition of fecal microbiota changed upon alteration of Nlrp12. We detected significant enrichment of various genera in Nlrp12^-/- B6/lpr male ( Figure 5E ) and female mice ( Figure 5F ). Strikingly, the intestinal microbiota diversity showed clear differences only in male mice that might explain the sex-dependent changes in disease phenotype. Analysis of alpha diversity from different intestinal segments (duodenum/jejunum, ileum, and colon) at 39 weeks of age showed that male ( Figure 6A , P=0.028), but not female ( Figure 6B , P=0.963), mice have distinct microbial composition upon alteration of Nlrp12. Similarly, the overall taxonomic composition was different for genotype and intestinal segment only among males ( Figure 6C , P=0.001) but not females ( Figure 6D , P=0.065). While not many changes were observed as in fecal microbiota, several genera were significantly altered in the intestinal microbiota of WT vs. Nlrp12^-/- B6/lpr male ( Figure 6E ) and female mice ( Figure 6F ).