Autophagy alleviates amiodarone-induced hepatotoxicity

Abstract

Amiodarone is a widely used antiarrhythmic drug that can cause the development of steatohepatitis as well as liver fibrosis and cirrhosis. The molecular mechanisms of amiodarone-mediated liver injury remain largely unknown. We therefore analyzed amiodarone-mediated hepatocellular injury in patients with chronic heart failure, in primary hepatocytes and HepG2 cells. We found that amiodarone-treated patients with chronic heart failure revealed significantly higher serum levels of caspase-cleaved keratin-18, an apoptosis biomarker, compared to healthy individuals or patients not receiving amiodarone. Furthermore, amiodarone treatment of hepatocytes resulted in apoptosis associated with lipid accumulation and ER-stress induction. Liver cell steatosis was accompanied by enhanced de novo lipogenesis which, after reaching peak levels, declined together with decreased activation of ER stress. The decline of amiodarone-mediated lipotoxicity was associated with protective autophagy induction. In contrast, in hepatocytes treated with the autophagy inhibitor chloroquine as well as in autophagy gene (ATG5 or ATG7)-deficient hepatocytes, amiodarone-triggered toxicity was increased. In conclusion, we demonstrate that amiodarone induces lipid accumulation associated with ER stress and apoptosis in hepatocytes, which is mirrored by increased keratin-18 fragment serum levels in amiodarone-treated patients. Autophagy reduces amiodarone-mediated lipotoxicity and could provide a therapeutic strategy for protection from drug-induced liver injury.

Keywords: Amiodarone; Apoptosis; Autophagy; Drug-induced liver injury; ER stress; Keratin-18.

Fig. 1

Amiodarone treatment results in liver cell apoptosis. a Detection of serum AST and ALT levels in chronic heart failure patients with or without amiodarone treatment. Patients with chronic heart failure treated with amiodarone revealed slightly elevated aminotransferase levels compared to patients without amiodarone treatment, which was, however, not significant. b Detection of caspase-cleaved keratin-18 (K18) in sera from chronic heart failure patients. Treatment of chronic heart failure patients with amiodarone resulted in significantly higher serum levels of caspase-cleaved K18 compared to patients not receiving amiodarone or healthy persons. c Amiodarone time-dependently induces caspase-3/-7 activity in HepG2 cells. Cells were treated for the indicated time with amiodarone (200 µM), before caspase activity was measured by a luminometric substrate assay. Data were obtained from 5 independent experiments. d Immunocytochemistry of caspase-3 activation and TUNEL staining after amiodarone treatment of HepG cells. Cells were either left untreated or incubated for 4 h with 200 µM amiodarone. Merged images were produced by overlaying caspase-3 (red), TUNEL (green) and nuclear DAPI (blue) staining. Bars = 100 µm. Significances indicated above the bars refer to control. *p < 0.05, **p < 0.01, n.s. = non-significant (colour figure online)

Fig. 2

Amiodarone induces hepatocyte steatosis and increased expression of lipogenic regulators. a, b Triglyceride accumulation in HepG2 cells, as assessed by Oil Red O staining, was induced after 2–6 h of amiodarone treatment (200 µM) followed by a decline after 8 h. c Transcription of the lipogenic regulators SREBP1c and DGAT1 was induced by amiodarone treatment of HepG2 cells with peak levels at 6 h and a decrease at 8 h. Data were obtained from 3 independent experiments. Significances indicated above the bars refer to control. *p < 0.05

Fig. 3

Amiodarone induces ER stress. a Western blot analysis demonstrated that amiodarone treatment (200 µM) of HepG2 cells results in the phosphorylation of IRE1α and eIF2α as well as in increased expression of the transcription factor CHOP. Phosphorylation of eIF2α and CHOP expression increases up to 4–6 h followed by a decline at 8 h of treatment. b Immunocytochemical detection of phosphorylated and hence activated IRE1α in HepG2 cells treated for 4 h with amiodarone further confirms ER stress induction. c Quantification of CHOP mRNA after amiodarone treatment of HepG2 cells compared to untreated control cells. Data were obtained from 4 independent experiments. Bars = 100 µm. *p < 0.05, **p < 0.01

Fig. 4

Amiodarone treatment triggers autophagy. a Western blot analysis revealed decreased p62/SQSTM1 expression and increased LC3B-I/II conversion in HepG2 cells treated with amiodarone for 2–8 h. b Compared to amiodarone treatment alone, pretreatment of HepG2 cells with the autophagy inhibitor chloroquine increased caspase-3 activation at 4 and 6 h of amiodarone treatment (left panel). The vertical lines indicate juxtaposition of non-adjacent lanes from the same blot. In contrast to amiodarone, chloroquine itself did not induce caspase-3 activation (right panel). c Compared to the corresponding control cells, ATG5- or ATG7-deficient HepG2 cells revealed decreased autophagy as shown by p62/SQSTM1 stabilization and lack of LC3B-I/-II conversion in Western blot analyses. d ATG5- or ATG7-deficiency resulted in enhanced cell death of HepG2 cells after 6 h of amiodarone treatment as compared to control cells. **p < 0.01

Fig. 5

Amiodarone-induced apoptosis, ER-stress and autophagy in primary human hepatocytes. a Treatment of PHHs (n = 8 donors) with amiodarone (200 µM) for 2 h significantly increased caspase-3/-7 activity followed by a significant decline at 4–8 h of treatment. b Western blot analysis confirmed a significant increase of activated caspase-3 after 2 h of PHH treatment with amiodarone, which then decreased at 4–8 h of treatment. c ER stress is triggered in PHHs by amiodarone, indicated by phosphorylation of IRE1α and eIF2α, which peaked at 2 h of treatment and thereafter declined. d Autophagy was induced at 4–8 h of amiodarone treatment of PHHs as indicated by decreased p62/SQSTM1. Significances indicated above the bars refer to control. *p < 0.05, **p < 0.01