Fig 1.
Hepatocyte injury, fibrosis, and cancer in NASH mice.
Six to eight week old mice were fed regular chow or WD for 24 weeks. NASH mice were injected once weekly with low dose CCl4. Representative liver histology demonstrated by A) H&E stain and B) picosirius red stain for fibrillar collagen to label areas of fibrosis. C) NAFLD activity score (NAS) was evaluated by a blinded pathologist. NASH mice have significantly elevated NAS, indicative of robust disease. D) Liver enzymes were measured in serum collected at the time of sacrifice. E) Expression of selected markers of hepatic stellate cell activation measured by quantitative real-time PCR (qRT-PCR). F) Liver tumors counted on the liver surface at 24 weeks. G) Endotoxin measured in serum collected at 12 or 24 weeks. Image magnification is 200x for H&E (A) and 50x for Sirius Red (B). For (A-F) n = 7 control and n = 3 NASH mice were included. (G) n = 3 mice per group. P values are encoded as * = p < 0.05, ** = p < 0.01, and *** = p < 0.001.
Fig 2.
Loss of abundance and altered composition of microbiome in NASH mice.
Feces were collected from chow-fed controls or NASH mice at weeks 0, 12, and 24. Composition of the microbiome was assessed 16S rRNA amplicon sequencing. A) Microbiota density in feces. NASH mice had significantly lower bacterial density than controls. B) Faith’s phylogenetic diversity, calculated to assess species diversity. NASH mice displayed reduced alpha diversity. C) Principal Coordinate Analysis plotting weighted UniFrac distances. NASH mice at 12 and 24 weeks cluster separately from week 0 baseline or chow-fed controls. (A-C) fecal samples from 3–7 mice per condition, per timepoint. P values are encoded as * = p < 0.05, ** = p < 0.01, and *** = p < 0.001.
Fig 3.
NASH shifts abundance of Bacteroidales and Erysipelotrichales.
A) Relative abundance of specific taxa at baseline or after 12, and 24 weeks of NASH protocol. B) Bacteroidales abundance is reduced in NASH while C) Erysipelotrichales increases dramatically. Fecal samples were sequenced from 3–7 mice per condition, per timepoint. P values are encoded as * = p < 0.05, ** = p < 0.01, and *** = p < 0.001.
Fig 4.
Separate effects of NASH and WD on the microbiome.
Mice were randomized to control, CCl4-only, WD-only, or complete NASH protocol and followed for 12 weeks. A) Microbiota density decreased in both WD-only and NASH mice, as did B) Faith’s Phylogenetic Diversity. C) Principal Coordinate Analysis of Bray-Curtis dissimilarity distances shows changes to beta diversity. WD-only and NASH have separate impacts on population structure. Fecal samples were sequenced from 3–10 mice per condition (mean = 5), except NASH week 12 where n = 2. P values are encoded as * = p < 0.05, ** = p < 0.01, and *** = p < 0.001.
Fig 5.
Erysipelotrichales expansion is specific to NASH.
A) Relative abundance of specific taxa over 12 weeks. B) Erysipelotrichales bacteria expand in NASH but not other conditions whereas C & D) decreased abundance of Verrucomicrobiales and Bacteroidales is observed in both WD-only and NASH. Fecal samples sequenced from 3–10 mice per condition (mean = 5), except NASH week 12 where n = 2. P values are encoded as * = p < 0.05, ** = p < 0.01, and *** = p < 0.001.