A multi-omic landscape of steatosis-to-NASH progression

Figure 1. Transcriptomic profiling associated with steatosis−to−NASH progression. (a−f) DEGs in the livers of NAFL and NASH mice. (a and b) Heatmap showing the expression patterns of DEGs in the livers of NAFL (a) and NASH mice (b) compared with their age-matched normal controls after batch-effect correction, respectively. (c and d) Volcano plot of DEGs in the livers of NAFL (c) and NASH mice (d). The top dysregulated mRNAs are labeled as indicated. (e) mRNA expression levels of top 5 up-regulated and 5 down-regulated genes in NASH mice analyzed by qRT-PCR. (f) Validation of DEGs in two public datasets of NASH patients (GSE89632 and GSE24807). (g) Comparison of up-regulated (left) or down-regulated (right) genes in the livers of NAFL and NASH mice. (h and i) GO (h) and KEGG (i) analysis of NASH-specific genes. Data are represented as mean ± SEM. Two-tailed Student’s t-test was used in (e). Wald test by Deseq 2 was used in (f). **P < 0.01, ***P < 0.001.

Figure 2. Gdf3 is specifically up-regulated in NASH mice. (a and b) Intersection of mRNAs specifically up-regulated (a) or down-regulated (b) in NASH mice with a mouse secretome database (NCBI GEO, accession number: GSE10246). (c) mRNA expression levels of Gdf3 in the livers of NAFL mice (left) and NASH mice (right) with their age-matched normal controls, respectively. (d) Protein expression of GDF3 in the livers of NAFL mice (top) and NASH mice (bottom) with their matched normal controls. (e) Plasma GDF3 levels of NAFL mice (left) and NASH mice (right). (f) Relative mRNA levels of Gdf3 in the livers of normal mice and NASH mice induced by CD-HFD (left) or MCD (right). In the CD-HFD experiment, C57BL/6J mice were fed a normal diet or CD-HFD for 10 weeks. In the MCD experiment, C57BL/6J mice were fed a normal diet or MCD for 6 weeks. (g) Protein expression of GDF3 in the livers of normal mice and NASH mice induced by CD-HFD (top) or MCD (bottom). (h) Proportion of cells expressing Gdf3 in the livers of normal mice or HFHC diet-induced NASH mice analyzed by single-cell sequencing. (i) The Gdf3-expressing cell types in the livers of NASH mice. Data are represented as mean ± SEM. **P < 0.01,…

Figure 3. Plasma GDF3 concentration is elevated in patients with NASH. (a) Relative fold change of hepatic Gdf3 expression from public datasets of human individuals with different liver diseases compared to normal healthy individuals. (b−i) Analysis of plasma GDF3 concentrations in human subjects. Normal: n = 315, NAFL: n = 110, NASH: n = 107. (b) Comparison of plasma GDF3 levels in patients with NAFL or NASH compared to normal individuals. (c and d) The correlation between plasma GDF3 concentrations and ALT (c) and AST (d) in total participants. (e−h) Plasma GDF3 levels in patients with NAFL or NASH with different NASH score (e), steatosis score (f), ballooning score (g), and lobular inflammation (h). (i) NASH diagnostic accuracy in the total NAFLD cohort (n = 217). AUROC, area under the receiver-operating characteristic curve. Wald test by Deseq 2 was used in (a). One-way ANOVA with Tukey’s post hoc test was used in (b, e−h). Spearman correlation was used in (c, d). DeLong’s test was used in (i). **P < 0.01, ***P < 0.001.

Figure 4. Proteomic profiling associated with steatosis-to-NASH progression. (a and b) Volcano plot of DEPs in the livers of NAFL (a) and NASH mice (b). The top DEPs are labeled as indicated. (c) Comparison of up-regulated (top) or down-regulated (bottom) proteins in the livers of NAFL and NASH mice. (d) Volcano plot of specifically altered proteins in NASH. (e and f) Integration of proteomic and transcriptomic profiles specifically altered in NASH. (e) Heatmap of differentially expressed genes and protein pairs. (f) Expression of gene-protein pairs. The blue circle dots represent up-regulated gene-protein pairs. The brown circle dots represent down-regulated gene-protein pairs. Spearman correlation was used in (e).

Figure 5. Plasma metabolomics analysis associated with steatosis-to-NASH progression. (a−d) Altered plasma metabolites of NAFL and NASH mice. (a) Intersection of dysregulated metabolites between Univariate analysis and OPLS-DA analysis of NAFL mice (top) and NASH mice (bottom). (b and c) Heatmap showing the patterns between dysregulated metabolites in NAFL mice (b) and NASH mice (c) after batch-effect correction. (d) Comparison of up-regulated (top) or down-regulated (bottom) metabolites in NAFL and NASH mice. (e−g) Analysis of specifically altered plasma metabolites in NASH mice. (e) The category of dysregulated metabolites. (f) Vioplot of top 5 increased or decreased plasma metabolites. (g) Functional enrichment pathway analysis of dysregulated metabolites.

Figure 6. Protein and metabolite interaction network associated with steatosis-to-NASH progression. GO analysis results after integrating protein–protein and protein–metabolite interactomes into a network using the PCSF network model.

Figure 7. Pharmacological suppression of ferroptosis improves long-term HFHC diet-induced NASH in mice. (a) Relative mRNA levels of ferroptosis-related genes involved in iron metabolism, antioxidant metabolism, and energy metabolism in NAFL (top) and NASH mice (bottom). (b) Protein expression of ferroptosis markers in NAFL (top) and NASH (bottom) mice. (c) Hepatic MDA content was measured to determine lipid peroxidation in NAFL (top) and NASH (bottom) mice with their age-matched normal controls. (d) Protein expression of ferroptosis markers in normal mice or NASH mice induced by CD-HFD (up) and MCD (bottom). (e−o) 8-week-old male C57BL/6J mice were fed a normal diet or HFHC diet for 34 weeks. Then, normal mice were treated with vehicle control while NASH mice were administered ZnPP (10 mg/kg, i.p.) or vehicle control for another 4 weeks. (e) Relative mRNA levels of Ptgs2, Slc7a11, Hmox1, and Gpx4 in the liver of three groups of mice. (f−h) Body weights (f), liver/body weight ratio (g), and hepatic triglyceride contents (h) among three groups of mice. (i) Hematoxylin and eosin (H&E) staining of liver sections. (j and k) Plasma ALT and AST levels. (l) Relative mRNA levels of genes involved…

Figure 8. Non-coding RNA profiling associated with steatosis-to-NASH progression. (a−j) Differentially expressed non-coding RNAs in the livers of NAFL and NASH mice. (a and b) Heatmap showing the expression patterns of lncRNAs in NAFL mice (a) and NASH mice (b) with their matched normal controls after batch-effect correction. (c and d) Heatmap showing the expression patterns of miRNAs in NAFL mice (c) and NASH mice (d) with their normal controls after batch-effect correction. (e and f) Volcano plot of dysregulated lncRNAs in NAFL mice (e) and NASH mice (f). (g and h) Volcano plot of dysregulated miRNAs in NAFL mice (g) and NASH mice (h). (i) Comparison of up-regulated (top) or down-regulated (bottom) lncRNAs in NAFL and NASH mice. (j) Comparison of up-regulated (top) or down-regulated (bottom) miRNAs in NAFL and NASH mice. (k) CNC network analysis in NASH. The interaction between dysregulated lncRNAs and mRNAs in NASH was integrated into a network. (l−n) ceRNA network of differentially expressed lncRNAs, differentially expressed miRNAs and DEGs in NASH. (l) The interactions between differentially expressed lncRNAs and miRNAs, as well as related DEGs, are integrated into a regulation…