Ferroptosis driven by radical oxidation of n-6 polyunsaturated fatty acids mediates acetaminophen-induced acute liver failure

Abstract

Acetaminophen (APAP) overdose is a common cause of drug-induced acute liver failure. Although hepatocyte cell death is considered to be the critical event in APAP-induced hepatotoxicity, the underlying mechanism remains unclear. Ferroptosis is a newly discovered type of cell death that is caused by a loss of cellular redox homeostasis. As glutathione (GSH) depletion triggers APAP-induced hepatotoxicity, we investigated the role of ferroptosis in a murine model of APAP-induced acute liver failure. APAP-induced hepatotoxicity (evaluated in terms of ALT, AST, and the histopathological score), lipid peroxidation (4-HNE and MDA), and upregulation of the ferroptosis maker PTGS2 mRNA were markedly prevented by the ferroptosis-specific inhibitor ferrostatin-1 (Fer-1). Fer-1 treatment also completely prevented mortality induced by high-dose APAP. Similarly, APAP-induced hepatotoxicity and lipid peroxidation were prevented by the iron chelator deferoxamine. Using mass spectrometry, we found that lipid peroxides derived from n-6 fatty acids, mainly arachidonic acid, were elevated by APAP, and that auto-oxidation is the predominant mechanism of APAP-derived lipid oxidation. APAP-induced hepatotoxicity was also prevented by genetic inhibition of acyl-CoA synthetase long-chain family member 4 or α-tocopherol supplementation. We found that ferroptosis is responsible for APAP-induced hepatocyte cell death. Our findings provide new insights into the mechanism of APAP-induced hepatotoxicity and suggest that ferroptosis is a potential therapeutic target for APAP-induced acute liver failure.

Fig. 1. Ferroptosis mediates APAP-induced hepatotoxicity and lethality.

Liver and serum samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice treated with Fer-1 (10 mg/kg) or vehicle 1 h prior to injection. a Serum levels of AST and ALT were assessed (n = 7–9 for each). b, c Liver sections were stained with HE, and histopathology scores were assessed (n = 4–6 for each). d Survival time after injection of a lethal dose of APAP (400 mg/kg) was analyzed using the Kaplan–Meier method (n = 4–6 each). e Hepatic CYP2E1 mRNA expression was assessed (n = 4–6 for each). f Liver samples were analyzed by western blotting for CYP2E1. β-actin served as a loading control. Statistical significance was calculated using the Mann–Whitney test c–e with the Bonferroni correction. Data are expressed as dot plots and/or means ± SEM. *p < 0.05, **p < 0.01.

Fig. 2. APAP induces Ptgs2 expression, lipid peroxidation, and GSH depletion.

Liver samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice were treated with Fer-1 (10 mg/kg) or vehicle 1 h prior to injection. a Hepatic PTGS2 mRNA expression was assessed (n = 4–6 for each). b Liver sections were analyzed by immunohistochemical staining for 4-HNE (n = 4–6 for each). c Hepatic MDA levels were assessed by MDA-TBARS assay (n = 5 for each). d Liver samples were analyzed by western blotting for xCT. β-actin served as a loading control. e Liver samples were analyzed by western blotting for GPX4. β-actin served as a loading control. f Hepatic levels of total GSH (GSH + 2GSSG) and reductant GSH were assessed. GSH was calculated as total GSH –2GSSG (n = 4–6 for each). Statistical significance was calculated using the Mann–Whitney test with the Bonferroni correction a–e. Data are expressed as dot plots and/or means ± SEM. *p < 0.05, **p < 0.01.

Fig. 3. Fer-1 has no effects on other types of cell death or inflammatory responses.

Liver samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice were treated with Fer-1 (10 mg/kg) or vehicle 1 h prior to injection. a, b Liver samples were analyzed by western blotting for caspase-3, cleaved caspase-3, and RIP3K. DEX-treated thymus and L929 cells were used as positive controls for cleaved caspase-3 and RIP3K, respectively. β-actin served as a loading control. c Liver sections were analyzed by immunohistochemical staining for CD45 (n = 4–6 for each). d Hepatic mRNA expression of inflammatory cell markers (Ly6G and EMR) and cytokines (IL-1β, IL-6, TNF-α, and CCL2) was assessed (n = 4–6 for each). Statistical significance was calculated using the Mann–Whitney test with the Bonferroni correction a–d. Data are expressed as dot plots and/or means ± SEM.

Fig. 4. Iron chelator inhibits APAP-induced hepatotoxicity and lipid peroxidation.

Liver and serum samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice were treated with DFO (100 mg/kg/day) or vehicle for 7 consecutive days prior to APAP injection. a Serum AST and ALT levels. b, c HE staining and histopathology score. d Hepatic CYP2E1 mRNA expression. e Immunostaining for 4-HNE. f Hepatic PTGS2 mRNA expression. g Hepatic MDA levels. h Hepatic GSH levels. Statistical significance was calculated using the Mann–Whitney test c–e with the Bonferroni correction. Data are expressed as dot plots and/or means ± SEM (n = 6–8 for each). *p < 0.05, **p < 0.01.

Fig. 5. Arachidonic acid-derived lipid peroxidation is involved in APAP-induced hepatotoxicity.

Liver samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice were treated with Fer-1 (10 mg/kg) or vehicle 1 h prior to injection. a–d Liver samples were analyzed by LC-MS to identify lipid mediators and volcano plots are shown. a APAP vs. vehicle in n-6 fatty acid, b APAP + Fer-1 vs. vehicle in n-6 fatty acid, c APAP vs. vehicle in n-3 fatty acid, d APAP + Fer-1 vs. vehicle in n-3 fatty acid. e–h Relative expression levels of PGD2, LTC4, 12-HETE, 20-HETE, and 8-iso-PGF2α are shown. The data are expressed as relative expression compared with a reference standard (n = 3 for each). Statistical significance was calculated using Student’s t test e–h. Data are expressed as dot plots and/or means ± SEM. *p < 0.05, **p < 0.01.

Fig. 6. ACSL4 deletion attenuates APAP-induced hepatotoxicity and lipid peroxidation.

a Liver and serum samples were obtained from ACSL4^+/Y and ACSL4^–/Y mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Serum levels of AST and ALT were assessed (n = 3–5 for each). b–e The hydrodynamics-based transfection method of CRISPR/Cas9 genome editing was used to create hepatic ACSL4-deleted mice. Liver and serum samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. px330-ACSL4-sgRNA or GFP-sgRNA was injected by the hydrodynamics-based method 7 days prior to APAP injection. b Hepatic ACSL4 expression. c Serum AST and ALT levels (n = 3–5 for each). d HE staining. e Histopathological score. f Hepatic PTGS2 mRNA expression. g 4-HNE immunostaining. h Hepatic MDA levels. i Hepatic total GSH levels. j Hepatic CYP2E1 mRNA expression. Statistical significance was calculated using the Mann–Whitney test with the Bonferroni correction a–e. Data are expressed as dot plots and/or means ± SEM. *p < 0.05, **p < 0.01 (n = 6–8 for each).

Fig. 7. Evaluation of lipid peroxidation mechanisms via analysis of lipid hydroperoxide isomers.

MRM chromatograms of PCOOH isomers in serum. Serum samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice were treated with Fer-1 (10 mg/kg) or vehicle 1 h prior to injection. Extract from a serum sample (10 µL) was analyzed by HPLC-MS-MS. Detailed analytical conditions are described in the Supplementary Information.

Fig. 8. α-Tocopherol inhibits APAP-induced hepatotoxicity and lipid peroxidation.

Liver and serum samples were obtained from mice injected with vehicle or APAP (200 mg/kg) 3 h after injection. Mice were treated with α-tocopherol (α-Toc, 100 mg/kg/day) or vehicle 1 h prior to injection. a Serum AST and ALT levels. b, c HE staining and histopathology score. d Hepatic CYP2E1 mRNA expression. e 4-HNE immunostaining. f Hepatic MDA levels. g Hepatic total GSH levels. Statistical significance was calculated using Mann–Whitney test a–g with the Bonferroni correction. Data are expressed as dot plots and/or means ± SEM (n = 3–6 for each). *p < 0.05, **p < 0.01.