Inflammasome-mediated dysbiosis regulates progression of NAFLD and obesity

Abstract

Non-alcoholic fatty liver disease (NAFLD) is the hepatic manifestation of metabolic syndrome and the leading cause of chronic liver disease in the Western world. Twenty per cent of NAFLD individuals develop chronic hepatic inflammation (non-alcoholic steatohepatitis, NASH) associated with cirrhosis, portal hypertension and hepatocellular carcinoma, yet the causes of progression from NAFLD to NASH remain obscure. Here, we show that the NLRP6 and NLRP3 inflammasomes and the effector protein IL-18 negatively regulate NAFLD/NASH progression, as well as multiple aspects of metabolic syndrome via modulation of the gut microbiota. Different mouse models reveal that inflammasome-deficiency-associated changes in the configuration of the gut microbiota are associated with exacerbated hepatic steatosis and inflammation through influx of TLR4 and TLR9 agonists into the portal circulation, leading to enhanced hepatic tumour-necrosis factor (TNF)-α expression that drives NASH progression. Furthermore, co-housing of inflammasome-deficient mice with wild-type mice results in exacerbation of hepatic steatosis and obesity. Thus, altered interactions between the gut microbiota and the host, produced by defective NLRP3 and NLRP6 inflammasome sensing, may govern the rate of progression of multiple metabolic syndrome-associated abnormalities, highlighting the central role of the microbiota in the pathogenesis of heretofore seemingly unrelated systemic auto-inflammatory and metabolic disorders.

Figure 1. Increased severity of NASH in inflammasome-deficient mice

To induce NASH, mice were fed with MCDD for 24 d. Their serum ALT and AST activities measured and NAFLD histological activity scores were determined. (a–h) Comparison of ALT, AST, and NAFLD activity, plus histological scores for steatosis and inflammation between singly-housed wild-type (wt) mice and caspase-1^−/− (a,b), Asc^−/− (c,d), Nlrp3^−/−(e,f), or Il18^−/−(g,h). Data represent two independent experiments (n=7–19 mice/treatment group). Error bars represent the SEM of samples within a group. *p≤ 0.05, **p≤ 0.01, ***p≤ 0.001 (Student’s t test).

Figure 2. Increased severity of NASH in Asc and Il18-deficient mice is transmissible to co-housed wild-type animals

Asc^−/− or Il18^−/− mice and wt mice were co-housed for 4 weeks and then fed MCDD. (a–d) ALT (a), AST (b), NAFLD activity scores (c), and H&E-stained sections of livers (d) of singly-housed wt mice (wt), wt mice co-housed with Asc^−/− (wt (Asc^−/−)), and Asc^−/− mice co-housed with wt (Asc^−/−(wt)). (e–h) ALT (e), AST (f), NAFLD activity histological scores (g), and H&E-stained sections of livers (h) of wt, wt(Il18^−/)⁻, and Il18^−/−(wt). Data are representative of two independent experiments. Error bars represent SEM. Scale bars = 200 μm (d,h). *p≤ 0.05, **p≤ 0.01, ***p≤ 0.001.

Figure 3. 16S rRNA sequencing demonstrates diet and co-housing associated changes in gut microbial ecology

(a) Principal Coordinates Analysis (PCoA) of Unweighted UniFrac distances of 16S rRNA sequences demonstrating clustering according to co-housing status on principal coordinate 1 (PC1). (b) PCoA of same plot as in (a) colored for experimental day. Mice were co-housed and fed a regular diet for the first 32 days of the experiment (two timepoints taken at day 20 and 32) before being switched to MCDD (sampled at days 39, 46 and 51 of the experiment). (c–f) PCoA and bargraphs of family level taxa Prevotellaceae, Porphyromonadaceae, Bacteroidaceae and Erysipelotrichaceae demonstrating diet-and microbiota-dependent differences in taxonomic representation. PCoA plots contain spheres represented a single fecal community colored according relative representation of the taxon (Blue represents relatively higher levels; red indicates lower levels). Bar graphs represent averaged taxonomic representation for singly or co-housed mouse while on either regular or MCD diet (n=8 for singly-housed wt, n=12 co-housed Asc−/−(wt) and wt(Asc−/−) animals;) * p < 0.05, ** p < 0.01, *** p < 0.001 by t-test after Bonferonni correction for multiple hypotheses. n.d. = not detected.

Figure 4. Increased severity of NASH in Asc-deficient and co-housed wild-type animals is mediated by TLR4, TLR9, and TNF-α

Asc^−/− mice were co-housed with wt, Tnfα ^−/−, Tlr4^−/−, Tlr9^−/−, or Tlr5^−/− mice for 4 weeks and then fed MCDD. (a–c) ALT levels of Tlr4^−/−(Asc^−/−) (a), Tlr9^−/−(Asc^−/−) (b), and Tlr5^−/−(Asc^−/−) (c) and their singly-housed counterparts. (d) TLR4 agonists in portal vein sera from MCDD-fed wt, wt(Asc^−/−), and Asc^−/− animals. (e) Transmission electron microscopy images of colon from wt and Asc^−/−. (f–h) ALT (f), and NAFLD activity histological scores (g–h) of Tnfα ^−/−, wt(Asc^−/−), and Tnfα ^−/− (Asc^−/−). Data are representative of two independent experiments. Error bars represent SEM. *p≤ 0.05, **p≤ 0.01, ***p≤ 0.001.

Figure 5. Increased severity of NASH in Asc-deficient mice is transmissible to db/db by co-housing and is mediated by CCL5-induced intestinal inflammation

(a–c) ALT (a), AST (b), and NAFLD activity histological scores (c) of wt(Asc^−/−) and Ccl5^−/−(Asc^−/−) mice. Data represents two independent experiments. (d–j) db/db mice were co-housed with wt or Asc^−/− mice for 12 weeks. (d–f) Representative H&E-stained sections of colon (d), terminal ileum (e), and liver (f) from db/db(wt) and db/db(Asc^−/−) mice fed a standard chow diet. Mucosal and crypt hyperplasia (arrow). Hepatocyte degeneration (arrowhead). Scale bars = 500 μm (d–e), Scale bars = 200 μm (f). (g–i) ALT (g), AST (h), and NAFLD activity scores (i) of db/db(wt), and db/db(Asc^−/−) mice. (j) Hepatic Tnfα, il6, and il1β mRNA levels. Error bars represent SEM. *p≤ 0.05, **p≤ 0.01, ***p≤ 0.001.

Figure 6. Asc-deficient mice develop increased obesity and loss of glycemic control on HFD

(a) Weight of db/db(wt) or (db/db(Asc^−/−) at 3 weeks of age and at 12 weeks of co-housing. (b–f) Asc^−/− and wt mice were co-housed for 4 weeks and then fed HFD. (b) Body weights. (c) NAFLD histological activity score. (d–e) Fasting plasma glucose and insulin after 11 weeks of HFD. (f) IPGTT after 12 weeks of HFD. (g–j) Mice were untreated, or treated orally with antibiotics, for 3 weeks prior to HFD feeding for 12 weeks. (g) Body weights. (h–i) Fasting plasma glucose and insulin levels after 8 weeks on a HFD. (j) IPGTT after 10 weeks of HFD. Error bars represent SEM. *p≤ 0.05, **p≤ 0.01, ***p≤ 0.001.