In addition to high-fat diet (HFD) and inactivity, inflammation and microbiota composition contribute to obesity. Inhibitory immune receptors, such as NLRP12, dampen inflammation and are important for resolving inflammation, but their role in obesity is unknown. We show that obesity in humans correlates with reduced expression of adipose tissue NLRP12. Similarly, Nlrp12 −/− mice show increased weight gain, adipose deposition, blood glucose, NF-κB/MAPK activation, and M1-macrophage polarization. Additionally, NLRP12 is required to mitigate HFD-induced inflammasome activation.
Wang joined the NCI in 2012. He is currently the Project Officer for GDC and a Biomedical Informatics Specialist at the NCI Center for Cancer Genomics (CCG). Before joining the NCI, Dr. Wang spent two years as a senior bioinformatics scientist with The Cancer Genome Atlas (TCGA) Data Coordinating Center. (13)National Cancer Institute, National Institutes of Health, Bethesda, Maryland 20892, USA. Somatic mutations have been extensively characterized in breast cancer, but the effects of these genetic alterations on the proteomic landscape remain poorly understood.
Co-housing with wild-type animals, antibiotic treatment, or germ-free condition was sufficient to restrain inflammation, obesity, and insulin tolerance in Nlrp12 −/− mice, implicating the microbiota. HFD-fed Nlrp12 −/− mice display dysbiosis marked by increased obesity-associated Erysipelotrichaceae, but reduced Lachnospiraceae family and the associated enzymes required for short-chain fatty acid (SCFA) synthesis. Lachnospiraceae or SCFA administration attenuates obesity, inflammation, and dysbiosis. These findings reveal that Nlrp12 reduces HFD-induced obesity by maintaining beneficial microbiota. Graphical Abstract.
In contrast to pro-inflammatory immune receptors, NLR proteins that dampen inflammation are important for the resolution of inflammation, yet their roles in obesity are completely unknown. NLRP12 is identified as a potent mitigator of inflammation. It is primarily expressed by dendritic cells, granulocytes, and macrophages, where it inhibits both canonical and non-canonical NF-κB and ERK activation. Nlrp12-deficient mice are highly susceptible to experimental models of colitis, colitis-associated colon cancer (CAC) (.) and found that human NLRP12 expression is significantly lowered in the obese population (A). Its expression is negatively correlated with body mass index (BMI), body weight, crown-like structures (CLSs) in the adipose tissue, and serum ferritin, but not with waist to hip ratio, age, or gender (A–S1G).
To determine if Nlrp12 in mice affected obesity, 5- to 6-week-old WT and Nlrp12 −/− mice were fed HFD (45% kcal from fat) or a control low-fat diet (LFD, 10% kcal from fat) for 20 weeks (B). Nlrp12 −/− mice consumed the same amount of food and water as WT mice (A and S2B) and showed no significant differences in the total respiratory exchange ratio (RER) (ratio of carbon dioxide made during metabolism to oxygen used) (C), but they had less energy expenditure (D). They also gained significantly more weight (C and 1D) and a greater percentage of body fat than WT mice during 22 weeks of HFD (E). Conversely, the percentage of lean mass in HFD-fed Nlrp12 −/− mice was significantly reduced (F).
Gonadal white adipose tissues (GWAT, open arrow) and inguinal white adipose tissues (InWAT, black arrows) (G) were isolated and stained with hematoxylin and eosin to reveal significant increases in adipocyte size in both GWAT and InWAT in Nlrp12 −/− versus WT mice on LFD, but this difference was absent in mice on HFD, possibly because adipocytes reached their maximum size in both strains on HFD (H and 1I). Nlrp12 −/− mice exhibited a significant increase in adipose tissue weight (J and 1K), but not in spleen, heart, and kidney weights, compared to HFD-fed WT controls and LFD-fed animals (E–S2G). TNF and IL6 have been positively correlated with obesity and insulin tolerance (.), and NLRP12 is primarily expressed by the myeloid lineage. The RNA sequencing (RNA-seq) profile of LPS-treated CD14 + monocytes revealed an inversed relationship between NLRP12 expression, which was reduced by LPS, and IL1B, TNF, and IL6, which were increased by LPS (A, top). Activation epigenetic markers were also downregulated by LPS in the NLRP12 promoter but increased in IL1B, TNF, and IL6 promoters.
When activated monocytes reverted to a quiescent state 24 hr after stimulation, proinflammatory cytokine genes ( IL1B, TNF, and IL6) were silenced while NLRP12 expression was restored (B). These results suggest that NLRP12 expression is inversely correlated with cytokine gene expression, and consistent with its role as an inhibitor of inflammatory response. In agreement with this, LFD-fed Nlrp12 −/− mice had more GWAT F4/80 hi macrophages compared to WT mice on LFD, but GWAT from Nlrp12 −/− mice on HFD had the most abundant F4/80 hi cells forming a CLS (C), which is formed by macrophage aggregation (. We isolated GWAT ATMs from WT or Nlrp12 −/− mice after 20 weeks on HFD and found samples from Nlrp12 −/− mice had significantly more F4/80 hiCD11b hi ATM cells than WT controls (D and 2E). There is also a relatively higher percentage of CD11c hi CD301 lo M1 macrophages in the Nlrp12 −/− ATMs compared to the WT control (D, right, and F). These results suggest that the loss of Nlrp12 resulted in M1 enrichment.
An analysis of gene expression in ATMs from WT mice revealed that Nlrp12 expression was reduced in M1 cells and inversely correlated with M1 genes (G).), was similarly enhanced in Nlrp12 −/− mice. Additionally, GWAT of HFD-fed Nlrp12 −/− mice had heightened p-p65 and p-ERK, and inflammasome activation indicated by increased pro-IL1β processing to mature IL1β (I). Collectively, these results demonstrate that Nlrp12 deficiency increases activation of ATMs and contributes to a more inflammatory M1-enriched status in the adipose tissue (J). Macrophage-Specific Deletion of Nlrp12 Resulted in Greater Obesity, Reduced Insulin Signaling, and Increased Inflammation. Analysis of isolated ATMs showed reduced Nlrp12 expression in ATMs isolated from obese mice relative to lean mice (A).
Suggests that during HFD, Nlrp12 restrains M1 proinflammatory macrophages. To directly assess if Nlrp12 in myeloid cells played a role during obesity, we generated myeloid-specific Nlrp12 knockout mice ( Nlrp12 flox/flox LysM-Cre +) (B). Nlrp12 flox/flox LysM-Cre + mice showed greater HFD-induced weight gain (C) and impaired glucose tolerance measured by the OGTT (oral glucose tolerance test) (D) than Nlrp12 flox/flox LysM-Cre − control mice, but only a test of weight gain reached significance. Insulin tolerance test (ITT) showed that Nlrp12 flox/flox LysM-Cre + mice showed significantly increased insulin tolerance (E), with a significant reduction of insulin signaling (p-AKT) in the liver and WAT (F). Adipose tissues in Nlrp12 flox/flox LysM-Cre + mice have elevated inflammation manifested by increased NF-κB nuclear p52, p-p65, and mature IL1β levels (G). These data indicate that the loss of myeloid Nlrp12 led to obesity, insulin tolerance, and increased immune signaling and inflammasome activation. To more stringently investigate the contribution of the microbiota, we examined germ-free (GF) WT and Nlrp12 −/− mice birthed and reared under gnotobiotic conditions.
![Nci guohao wango Nci guohao wango](/uploads/1/2/5/5/125512052/733012260.png)
We were unable to administer HFD to GF animals because our gnotobiotic facility does not allow the introduction of special feeds due to the difficulty in maintaining sterility in irradiated HFDs. However, Nlrp12 −/− mice raised under specific pathogen-free (SPF) condition and fed a standard chow still gained significantly more weight than WT animals on the same regimen (D). The weight gain of Nlrp12 −/− mice fed a normal diet occurred at a later time than mice fed HFD (shown in ). This observation allowed us to compare weight gain in Nlrp12 −/− versus WT mice fed a normal diet in a GF environment. On normal diet, there was more significant weight gain and increased GWAT and InWAT between SPF Nlrp12 −/− versus WT mice at 44 weeks of age and 44-week-old GF Nlrp12 −/− versus WT mice (D–4G).
This suggests that the microbiota played a role in excessive weight gain by Nlrp12 −/− mice. To analyze the impact of the microbiota on colon inflammation, colons of Abx-treated Nlrp12 −/− mice displayed significantly reduced inflammasome activation measured by the conversion of pro-IL1β to mature IL1β compared to untreated Nlrp12 −/− mice. Nuclear NF-κB p52, p-p65, and p-ERK were also lowered but did not reach significance (A). Abx also reduced IL6 and TNF in colon explants from HFD-fed Nlrp12 −/− mice (B). These results suggest that Abx-mediated changes in the microbiota resulted in attenuated colonic inflammation. Furthermore, serum IL6 and TNF were higher in SPF Nlrp12 −/− mice than WT controls, while these differences disappeared in mice treated with Abx or raised under GF condition (H and 4I), affirming that the microbiota affected basal inflammation and weight gain in Nlrp12 −/− mice.
To determine if Nlrp12 deficiency caused microbiome changes during HFD, we profiled bacterial 16S rRNA genes in feces of WT and Nlrp12 −/− mice fed HFD or LFD for 20 weeks (J). HFD-fed WT mice showed a loss in bacterial diversity compared to LFD-fed WT mice. Nlrp12 deficiency compounded by an HFD showed even more loss of colonic bacterial diversity (K).
Thus, Nlrp12 deficiency and HFD each led to a significantly altered microbiome composition as shown by a principal component analysis (PCA) and quantified by UniFrac dissimilarity distance (L and 4M). The reduction of microbial diversity in Nlrp12 −/− mice on HFD suggests increased basal intestinal inflammation. Indeed, colon explants from Nlrp12 −/− mice on HFD showed increased nuclear NF-κB p52, p-ERK, cleaved caspase-1 p-20, and mature IL1β when compared to WT mice on HFD (N). The former also contained more anti-microbial peptides, Reg3γ (regenerating islet-derived protein III-gamma) and CRAMP (cathelin-related antimicrobial peptides) (N).
Increased intestinal inflammation and NF-κB promote Reg3γ and CRAMP expression (. To decipher if changes in gut microbiota preceded the onset of obesity, we collected fecal DNA from mice on HFD at earlier times (1, 5, and 15 weeks). Among WT mice, there was no significant diversity loss between 1 and 5 weeks of HFD, but after 15 weeks of HFD treatment, there was a significant loss of diversity (A and S6B). Nlrp12 −/− mice showed a similar pattern, except overall there was more loss of diversity compared to WT mice at all three time points. Furthermore, Clostridiales and Lachnospiraceae were significantly reduced and Erysipelotrichaceae was significantly increased in Nlrp12 −/− mice compared to WT controls (C), which correlated with an increased inflammation status in HFD Nlrp12 −/− mice from weeks 1 to 15 (D). These results indicated altered bacterial groups in Nlrp12 −/− mice were evident before the onset of overt obesity. We next transplanted fecal contents containing a high proportion of Lachnospiraceae or fecal material with a low proportion of Lachnospiraceae but high Erysipelotrichaceae into GF recipient mice followed by HFD (E).
After 20 weeks of HFD, fecal contents with higher proportion of Erysipelotrichaceae promoted weight gain, higher fasting glucose level (F and S6G), and elevated intestinal inflammation from the colon (H). Cohousing with WT Mice Attenuates HFD-Induced Obesity in Nlrp12 −/− Mice.
Weaned age- and gender-matched WT and Nlrp12 −/− mice were either single housed (SiHo) or cohoused (CoHo) at 8 weeks of age for another 20 weeks of HFD (A). Control SiHo Nlrp12 −/− mice showed significantly increased body weight (B) and percentage weight gain (C) compared to SiHo WT mice. In contrast, Nlrp12 −/− mice cohoused with WT mice (CoHo Nlrp12 −/−) showed similar body weight as SiHo WT and CoHo WT given HFD, and all were significantly leaner than SiHo Nlrp12 −/− littermates (B and 5C).
Moreover, CoHo Nlrp12 −/− mice and CoHo WT cagemates had similar GWAT and InWAT weights (D) and percentage of body fat (E).