Metabolic dysfunction-associated steatohepatitis reduces hepatic H2S-producing enzymes altering persulfidome composition

Abstract

Metabolic dysfunction-associated steatohepatitis (MASH) is a progressive disease driven by obesity-related hepatic inflammation and oxidative stress. Recently, cysteine persulfidation (PSSH), a protective post-translational modification by hydrogen sulfide (H₂S), was established to play a role in redox regulation. Despite the role of the liver in H₂S metabolism, the function of PSSH in MASH remains underexplored. We demonstrated that H₂S-producing enzymes are downregulated in both human and mouse livers with steatosis and fibrosis, resulting in a decline in global PSSH levels. Dimedone-switch mass spectrometry in dietary mouse models of distinct obesity-associated liver disease stages revealed dysregulated PSSH on specific proteins. Surprisingly, increased hepatic PSSH levels of protein tyrosine phosphatases and redox regulators were found in advanced disease stages, suggesting a targeted adaptive response to oxidative stress. Overall, our findings demonstrated that impaired H₂S production disrupts protective PSSH networks in MASH. However, selective PSSH preservation on redox-sensitive proteins may represent a compensatory mechanism, underscoring the therapeutic potential of persulfidation in restoring redox homeostasis during obesity-associated chronic liver disease.

Fig. 1

H₂S-producing enzymes are downregulated in metabolic dysfunction–associated steatohepatitis (MASH) (A) Heatmap of protein expression levels of H₂S-producing enzymes and redox regulators in liver biopsies from patients with healthy liver (H, n = 3), metabolic-associated steatotic liver (S, n = 4), and metabolic-associated steatohepatitis (M, n = 4). The proteins were identified and analyzed by mass spectrometry. (B) Expression of CTH (CSE), CBS and MPST across different clusters in human livers. The size of each circle represents the percentage of cells expressing the gene, while the colour of circles indicates the average expression level of the gene. Data obtained from human liver atlas. (C) mRNA expression levels of CTH (CSE), CBS, and MPST in human livers from GSE126848 from patients with healthy liver (n = 14), obesity (n = 12), metabolic-associated steatotic liver (MASL, n = 15), and metabolic-associated steatohepatitis (MASH, n = 16). (D) mRNA expression levels of CTH (CSE), CBS, and MPST in human livers from GSE164760 representing different stages of MASLD. Samples from patients with healthy liver (n = 6), MASH (n = 74), cirrhosis (n = 8), peritumor (n = 29), and tumor (n = 53). (E) Heatmap of protein expression levels of H₂S-producing enzymes and redox regulators in the livers from mice fed with different diets (n = 4–5). Samples were analyzed by mass spectrometry. (F) Western blotting to validate protein expression levels of H₂S-producing enzymes in the livers of mice fed different diets. The data are mean ± SD (n = 4–5). ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001 and ∗∗∗∗P < 0.0001.

Fig. 2

Protein persulfidation is downregulated in mouse liver with steatohepatitis (A) H₂S production capacity in liver tissue from control and CDHFD-fed mice (100 μg of proteins). H₂S production was measured using lead acetate plate assay with the substrate after 2 h of incubation. The value is the mean intensity measured by ImageJ. Each dot represents an individual mouse liver sample. The data are mean ± SD (n = 5–7). (B) PSSH levels in liver tissues of mice fed with control and CDHFD. The data are mean ± SD (n = 4) (C) PSSH levels and H₂S-producing enzymes in isolated primary mouse hepatocytes treated with TGF- β1 (5 ng/ml) for 24 h. The data are mean ± SD (n = 5). (D) H₂S-producing enzymes and albumin expression levels in isolated primary mouse hepatocytes cultured for up to 48 h. The data are mean ± SD (n = 4). (E) H₂S production capacity in isolated primary mouse hepatocytes culture for up to 48 h H₂S production from lysate (100 μg) was measured by lead acetate plate assay with the substrate after 3 h incubation. The data are mean ± SD (n = 3) (F) PSSH levels in isolated primary mouse hepatocytes cultured for up to 48 h. The data are mean ± SD (n = 3). ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001.

Fig. 3

Mapping the protein persulfidation landscape of MASH (A) Volcano plot illustrating the changes in total proteome and persulfidome profiles in CDHFD-/control-fed mice. (B) Venn Diagram of the proteins identified in the total proteome and persulfidome. (C) Fold change of persulfidated proteins in CDHFD-fed mice compared with control diet-fed mice after normalization. The fold change less than 0.76 (marked as blue) and more than 1.3 (marked as red) are considered changed. (D) KEGG pathway enrichment analysis of significantly decreasing and increasing proteins at PSSH levels was performed using DAVID and plotted. The size of the bubbles is indicative of the number of proteins annotated with that term; bubbles are color-coded according to the significance of the enrichment. (E) PSSH changes in subcellular compartments. Colors indicate the proportion of proteins with increased (red) or decreased (blue) PSSH levels in CDHFD-fed mice compared with control-fed mice; grey represents the proportion of proteins that did not change. Protein subcellular localization was determined using DAVID and Gene ontology cellular compartment. (F) Box plot of log2 transformed abundancies of PRDX2 (peroxiredoxin 2), THIO (Thioredoxin), mitochondrial superoxide dismutase (SOD2), and TRXR2 (thioredoxin reductase 2) identified in the total proteome (left) and persulfidome (right). The data are presented as box-and-whisker plots, showing the minimum, maximum, and interquartile range (IQR) for n = 4–5. (G) ROS levels in liver tissue from control and CDHFD-fed mice were measured using DCFDA. The data are mean ± SD (n = 5–7). (H) Lipid peroxidation (MDA) levels in liver tissue from control and CDHFD-fed mice were measured using the TBRAS assay. The data are mean ± SD (n = 5–7). (I) Networks of enriched biological processes in CDHFD-Control group. The biological processes were selected based on a p-value (<0.05) and an FDR (<0.25) cutoff from the GSEA enrichment (red for processes enriched in the CDHFD group, while blue for processes enriched in control group). (J) NADPH levels in liver samples from mice fed a control diet or CDHFD were analyzed. The data are mean ± SD (n = 5–7). ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001.

Fig. 4

Persulfidation protects protein tyrosine phosphatases (A) Schematic diagram of PTP oxidation and persulfidation. Protein persulfidation can serve as a protective mechanism against irreversible oxidation. Liver samples were tag-switched with DCP-Bio1, and persulfidated proteins were enriched using magnetic streptavidin beads. (B) Total PTP oxidation levels in liver samples from control and CDHFD-fed mice. The data are mean ± SD (n = 4) (C) PTP persulfidation levels in livers. Samples after biotin enrichment were analyzed by western blot. 20 μg of tag-switched samples were loaded as input. Samples without tag-switch using DCP-Bio1 were used as negative controls. The data are mean ± SD (n = 5–7). (D) Total PTP oxidation in isolated primary mouse hepatocytes collected at different time points up to 48 h. Total PTPN2 in the same blot was detected using a fluorescent secondary antibody. The data are mean ± SD (n = 3). (E) PTPN2 persulfidation levels in isolated primary mouse hepatocytes collected at different time points. Samples were tag-switched with DCP-Bio1, and persulfidated proteins were enriched using magnetic streptavidin beads. 20 μg of tag-switched samples were loaded as input. The data are mean ± SD (n = 4). (F) Activity of untreated or treated recombinant human PTPN2 with Na₂S₄ (20 μM) or H₂O₂ (200 μM) for 30 min. After 10 min of measurement, the samples were treated with 1 mM DTT. The data are mean ± SD (n = 5). (G) Activity of recombinant human PTPN2 untreated or treated with Na₂S (200 μM) + H₂O₂ (200 μM) or H₂O₂ alone (200 μM) for 30 min. After 10 min of measurement, the samples were treated with 1 mM DTT. The data are mean ± SD (n = 5). ∗P < 0.05 and ∗∗P < 0.01.

Fig. 5

H₂S regulates the inflammatory pathway in hepatocytes (A) Schematic diagram of PTP and pro-inflammatory signaling pathways. PTPN2 negatively regulates JAK/STAT signaling. Enhanced STAT1/3 signaling drives the progression of MASH/fibrosis and HCC development. While H₂S can both protect PTP activity, its effects on downstream signaling cascades remain unclear. The net impact of H₂S on inflammatory signaling pathways is yet to be determined. (B) HepG2 cells were treated with Na₂S for 30 min. The phosphorylation of STAT3 (Tyr705) was measured by western blotting. The data are mean ± SD (n = 3). (C) Freshly Isolated primary mouse hepatocytes were treated with Na₂S for 30 min. Phosphorylation of STAT3 (Tyr705) were measured by western blotting. The data are mean ± SD (n = 3). (D) HepG2 cells were starved in medium without serum for 2 h and were non-treated or pre-treated with 500 μM Na₂S or Na₂S₄ for 10 min. After removing the medium, cells were treated with (C) IL-6 (1000 U/ml) and collected at the indicated time points. Phosphorylation of STAT3 (Tyr705) was measured by western blotting. The data are mean ± SD (n = 3). Statistical analyses were performed using two-way ANOVA. (E) HepG2 cells were starved in medium without serum for 2 h and were non-treated or pre-treated with 500 μM Na2S or Na₂S₄ for 10 min. After removing the medium, cells were treated with IFN-γ (50 U/ml) and collected at the indicated time points. Phosphorylation of STAT1 (Tyr701) was measured by western blotting. The data are mean ± SD (n = 3). Statistical analyses were performed using two-way ANOVA. (F) Freshly Isolated primary mouse hepatocytes were starved in medium without serum for 2 h and were non-treated or pre-treated with 500 μM Na₂S for 10 min. After removing the medium, cells were treated with IL-6 (1000 U/ml) and collected at the indicated time points. Phosphorylation of STAT3 (Tyr705) was measured by western blotting. The data are mean ± SD (n = 3). Statistical analyses were performed using multiple t-test. (G) Freshly Isolated primary mouse hepatocytes were starved in medium without serum for 2 h and were non-treated or pre-treated with 500 μM Na₂S for 10 min. After removing the medium, cells were treated with IFN-γ (50 U/ml) and collected at the indicated time points. Phosphorylation of STAT3 (Tyr705) was measured by western blotting. The data are mean ± SD (n = 3). Statistical analyses were performed using the multiple t-test. (H) Box plot of log2 transformed abundancies of STAT3 identified in the total proteome (upper) and persulfidome (lower). The data are presented as box-and-whisker plots showing the minimum, maximum, and interquartile range (IQR) with n = 4–5. ∗P < 0.05, ∗∗P < 0.01.