Distinct functions of JNK and c-Jun in oxidant-induced hepatocyte death

Abstract

Overactivation of c-Jun N-terminal kinase (JNK)/c-Jun signaling is a central mechanism of hepatocyte injury and death including that from oxidative stress. However, the functions of JNK and c-Jun are still unclear, and this pathway also inhibits hepatocyte death. Previous studies of menadione-induced oxidant stress demonstrated that toxicity resulted from sustained JNK/c-Jun activation as death was blocked by the c-Jun dominant negative TAM67. To further delineate the function of JNK/c-Jun signaling in hepatocyte injury from oxidant stress, the effects of direct JNK inhibition on menadione-induced death were examined. In contrast to the inhibitory effect of TAM67, pharmacological JNK inhibition by SP600125 sensitized the rat hepatocyte cell line RALA255-10G to death from menadione. SP600125 similarly sensitized mouse primary hepatocytes to menadione toxicity. Death from SP600125/menadione was c-Jun dependent as it was blocked by TAM67, but independent of c-Jun phosphorylation. Death occurred by apoptosis and necrosis and activation of the mitochondrial death pathway. Short hairpin RNA knockdowns of total JNK or JNK2 sensitized to death from menadione, whereas a jnk1 knockdown was protective. Jnk2 null mouse primary hepatocytes were also sensitized to menadione death. JNK inhibition magnified decreases in cellular ATP content and β-oxidation induced by menadione. This effect mediated cell death as chemical inhibition of β-oxidation also sensitized cells to death from menadione, and supplementation with the β-oxidation substrate oleate blocked death. Components of the JNK/c-Jun signaling pathway have opposing functions in hepatocyte oxidant stress with JNK2 mediating resistance to cell death and c-Jun promoting death.

Fig. 1

SP600125 sensitizes RALA hepatocytes to menadione-induced death despite inhibiting JNK. A: Cells were infected with control virus Ad5LacZ or Ad5TAM and treated with menadione for the indicated number of hours. JNK activity was determined by an in vitro kinase assay with c-Jun as the substrate. Activity was assayed by immunoblots for phosphorylated c-Jun (P-c-Jun), and membranes were reprobed for total c-Jun to assess loading equivalency. B: Proteins from Ad5LacZ- and Ad5TAM-infected cells treated with menadione for the indicated hours and immunoblotted with an amino-terminal anti-c-Jun antibody for wild-type c-Jun (c-Jun) and a carboxy-terminal anti-c-Jun antibody that detects the truncated c-Jun TAM67 (TAM67). β-actin is a loading control. C: JNK assay of cells pretreated with vehicle dimethyl sulfoxide (DMSO) or SP600125 alone, or in combination with 30 or 50 μM menadione (Men), TNF, or phorbol myristate acetate (PMA). D: Percentage cell death in RALA hepatocytes treated with menadione alone (Men) or with SP600125 (SP+Men) at the indicated μM concentrations of menadione for 24 h (*P<0.0002 as compared to menadione alone; n=5). E: Percentages of untreated control (Con) cells, or cells treated for 12 h with SP600125 (SP) or 50 μM menadione (Men) alone, or in combination (SP/Men), that were positive by acridine orange/ethidium bromide costaining for necrosis or apoptosis under fluorescence microscopy (*P<0.0001 as compared to menadione alone; n=6).

Fig. 2

Death from SP600125/menadione occurs through mitochondrial death pathway-mediated caspase activation. A: RALA hepatocytes were treated with 50 μM menadione and SP600125 alone or in combination for 8 h. Cytosolic and mitochondrial protein fractions were immunoblotted with antibodies for cytochrome c (Cyt c), and cytochrome oxidase (Cyt ox) and tubulin to demonstrate fraction purity. B: Percentage cell death by MTT assay in cells infected with Ad5LacZ (LacZ) or Bcl-2- or Bcl-X_L-expressing adenoviruses and treated with SP600125 and 50 or 60 μM menadione for 24 h (*P<0.005 as compared to Ad5LacZ-infected cells; n=6). C: Percentage cell death 24 h after treatment with SP600125 and 40 or 50 μM menadione in the absence (SP) or presence (SP+QVD) of Q-VD-OPh (*P<0.001 as compared to cells without Q-VD-OPh; n=5). D: Percentage cell death in primary hepatocytes treated with the indicated μM concentration of menadione alone (Men) or with SP600125 pretreatment (SP+Men) for 24 h (*P<0.05, **P<0.001 as compared to menadione alone; n=5). E: Relative ROS levels in the absence (SP−) or presence (SP+) of SP600125 and left untreated (No tx) or treated with 50 μM menadione (Men 50) for 1 h (n=4–5).

Fig. 3

Cell death from SP600125/menadione treatment occurs from a block in early JNK activation and is independent of c-Jun phosphorylation. A: RALA hepatocytes were treated with dimethyl sulfoxide (DMSO) or SP600125 alone or with 50 μM menadione for the indicated hours. Total protein was isolated and immunoblotted for phospho- (P-c-Jun) and total (c-Jun) c-Jun, phospho- (P-JNK) and total (JNK) JNK, phospho- (P-ERK1/2) and total (ERK1/2) ERK1/2 and β-actin as a loading control. B: Cells were infected with Ad5LacZ or Ad5TAM and treated with 50 μM menadione. Total protein from these cells was immunoblotted with the same antibodies as in (A). C: Percentage cell death at 24 h in untreated control (Con) RALA hepatocytes, and cells treated with 50 μM menadione together with SP600125 1 h before (−1h), or 1 (+1h), 2 (+2h) or 4 (+4h) h after menadione treatment (*P<0.01, **P<0.001 as compared to cells treated with SP600125 1 h before menadione; n=4).

Fig. 4

SP600125/menadione death is c-Jun dependent. A: RALA hepatocytes were infected with Ad5LacZ or Ad5TAM, pretreated with SP600125 and treated with 40 or 50 μM menadione. Percentage cell death was determined at 24 h by MTT assay (*P<0.00004 as compared to Ad5LacZ-infected cells; n=4–8). B: Immunoblots for total JNK and β-actin in jnk knockdown cells. C: Percentage cell death 24 h after treatment with the indicated menadione concentrations in knockdown cells (*P<0.002, ^#P<0.01 as compared to VEC cells; n=5–7). D: Percentages of apoptotic and necrotic cells by fluorescence microscopy 12 h after 50 μM menadione treatment (*P<0.01 as compared to menadione-treated VEC cells; n=4). E: Percentage cell death in wild-type (WT), jnk1^−/− and jnk2^−/− mouse primary hepatocytes by MTT assay after 24 h treatment with the menadione concentrations shown (*P<0.03, **P<0.005 as compared to menadione-treated wild-type cells; n=3–5).

Fig. 5

SP600125/menadione leads to ATP depletion and decreased β-oxidation. A: ATP levels in RALA hepatocytes treated with 50 μM menadione for 2 h in the absence (SP−) or presence (SP+) of SP600125 (*P<0.001 as compared to SP-/menadione-treated cells; n=4). B: ATP levels in identically treated cells at 4 h (*P<0.00004; n=3). C, D: Lactate production in similarly treated cells at 2 h (n=3–4) (C), and 4 h (n=4) (D) after menadione. E, F: Rates of β-oxidation at 2 h (E) and 4 h (F) after menadione (*P< 0.02; n=5–6).

Fig. 6

Total jnk knockdown decreases ATP levels and rates of β-oxidation. A, B: ATP levels in VEC and jnk knockdown cells at 2 (A) and 4 (B) h after treatment with 50 μM menadione (*P<0.004 as compared to VEC cells; n=3–5). C, D: Cumulative lactate production over 2 (C) and 4 (D) h (n=3). E, F: Rates of β-oxidation after 2 h (*P<0.03 as compared to VEC cells; n=3) (E), and 4 h (*P<0.002 as compared to VEC cells; n=3) (F) of menadione.

Fig. 7

Inhibition of β-oxidation sensitizes to death from menadione. A: Percentage cell death in wild-type cells 24 h after treatment with etomoxir (Eto) or 50 μM menadione (Men) alone, or in combination (Men+Eto) (*P<0.00001 as compared to menadione-treated cells; n=4–6). B: ATP levels 4 h after menadione in the absence (Eto−) or presence (Eto+) of etomoxir (*P<0.001 as compared to menadione-treated cells; n=4). C: Percentage cell death in siJNK cells 24 h after 50 or 60 μM menadione treatment without (OL−) or with (OL+) oleate supplementation (*P<0.0005 as compared to OL- cells; n=3). D: ATP levels in siJNK cells treated with 60 μM menadione for 4 h in the absence (OL−) or presence (OL+) of oleate (*P<0.001 as compared to OL- cells; n=3). E: Percentage cell death at 24 h in wild-type RALA hepatocytes treated with the indicated concentrations of menadione and SP600125 alone (SP) or with oleate (SP+OL) (*P<0.00003 as compared to SP cells; n=5).