Fig 1.
Activation of AMPK prevented drug-induced hepatocellular injury and hepatocyte depolarization.
Using LIVE/DEAD viability/cytotoxicity Kit (Invitrogen) and confocal microscope images, hepatocyte viabilities were analysed (see details in Methods). After hepatocytes were treated with various concentrations of acetaminophen (APAP) or diclofenac (Diclo), dose-viability response curves were obtained for (A) acetaminophen (1–20mM, 24hr) in rat hepatocytes, (B) diclofenac (50–800μM, 24hr) in rat hepatocytes, (C) acetaminophen (1–50mM, 24hr) in human hepatocytes, and (D) diclofenac (100–3000μM, 24hr) in human hepatocytes. (E) Western blot was used to detect the expressions of AMPK, phos-AMPK (Thr172), ACC and phos-ACC (Ser79). Quantitative analyses were conducted by calculating the relative densities compared to respective controls. 200μM AICAR significantly activated AMPK measured by the ratio of phos-AMPK (Thr172) to total AMPK (195% of that in the control), and (F) AICAR significantly increased the ratio of phos-ACC (ser79) to total ACC (140% of that in the control). (G) Rat hepatocytes were treated with acetaminophen (10mM, 24h) or diclofenac (250μM, 24h) in the presence or the absence of AICAR (200μM). Viabilities were measured and percentages of viability relative to the control were calculated. Both drugs significantly decreased viability, and AICAR prevented drug-induced decrease in viability (relative viabilities were: 61% in APAP; 88% in APAP + AICAR; 86% in Diclo, 99% in Diclo +AICAR and 99% in AICAR). (H) Human hepatocytes were treated with acetaminophen (25mM, 24h) or diclofenac (1000μM, 24h) in the presence or the absence of AICAR (200μM). Addition of AICAR prevented drug-induced decrease in viability (relative viabilities were: 72% in APAP; 95% in APAP + AICAR; 70% in Diclo, 93% in Diclo +AICAR and 99% in AICAR). (I) Immunofluorescence and confocal microscope were used to examine polarized morphology after various treatments in hepatocytes (see details in Methods). Representative images showed that acetaminophen and diclofenac depolarized hepatocytes resulting in less branched canalicular network in cells. AICAR prevented drug-induced depolarization in rat hepatocytes treated with acetaminophen or diclofenac. The white arrows indicate the representative canalicular network, which is the orange-yellow tubular structure within the hepatocytes. (J) Quantitative analyses of polarization were performed by measuring the canalicular lengths. Percentages of canalicular lengths relative to the control were calculated. Both drugs reduced the canalicular length. AICAR prevented the decrease in canalicular length in rat hepatocytes treated with acetaminophen (10mM, 24h) or diclofenac (250μM, 24h) (relative canalicular lengths were: 41% in APAP; 92% in APAP + AICAR; 74% in Diclo; 98% in Diclo + AICAR and 103% in AICAR). (K) AICAR prevented the decrease in canalicular length in human hepatocytes treated with acetaminophen (25mM, 24h) or diclofenac (1000μM, 24h) (relative canalicular lengths were: 48% in APAP; 71% in APAP + AICAR; 44% in Diclo; 68% in Diclo + AICAR and 94% in AICAR) (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 2.
Activation of AMPK prevented drugs-induced mitochondrial dysfunction.
Rat or human hepatocytes were treated with acetaminophen (APAP) or diclofenac (Diclo) in the presence or the absence of AMPK activator AICAR (200μM). Cellular ATP levels were measured using ATP determination kit (Invitrogen). Percentages relative to respective controls were calculated. (A) AICAR prevented the decrease in cellular ATP in rat hepatocytes treated with acetaminophen (10mM, 24h) or diclofenac (250μM, 24h). Treatment with AICAR alone significantly increased ATP level. The relative ATP were 61% in APAP; 100% in APAP + AICAR; 65% in Diclo; 104% in Diclo + AICAR and 130% in AICAR, and (B) AICAR prevented the decrease in ATP in human hepatocytes treated with acetaminophen (25mM, 24h) or diclofenac (1000μM, 24h). The relative ATP levels were 28% in APAP; 92% in APAP + AICAR; 59% in Diclo; 100% in Diclo + AICAR and 147% in AICAR. Treatment with AICAR alone significantly increased ATP level. (C) Mitochondrial potentials were measured and quantitatively analysed using Image J (see details in Methods). Percentages relative to respective controls were calculated. AICAR prevented the decrease in mitochondrial potential in rat hepatocytes treated with acetaminophen (10mM, 24h) or diclofenac (250μM, 24h). Treatment with AICAR alone significantly increased mitochondrial potential. FCCP treatment was used as negative control. The relative mitochondrial potentials were 62% in APAP; 96% in APAP + AICAR; 72% in Diclo; 97% in Diclo + AICAR; 113% in AICAR and 27% in FCCP. (D) AICAR also prevented the decreases of mitochondrial potential in human hepatocytes treated with acetaminophen (25mM, 24h) or diclofenac (1000μM, 24h). The relative mitochondrial potentials were 68% in APAP; 96% in APAP + AICAR; 44% in Diclo; 91% in Diclo + AICAR; 136% in AICAR and 30% in FCCP (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 3.
Activation of AMPK prevented drugs-induced mitochondrial fragmentation.
(A) Rat hepatocytes were treated with acetaminophen (10mM, 24h) or diclofenac (250μM, 24h) with or without AICAR (200μM), and stained with mito-Tracker Green and TMRE (red color). Merged images were taken by confocal microscope (see details in Methods). Both acetaminophen (APAP) and diclofenac (Diclo) caused mitochondrial fragmentation, and AICAR prevented the drug-induced fragmentation. Zoomed areas were indicated. (B) Immunofluorescence of Tom20 (green color) and confocal microscope images showed that acetaminophen (25mM, 24h) and diclofenac (1000μM, 24h) caused mitochondrial fragmentation in human hepatocytes, and AICAR prevented the fragmentation. Zoomed areas were indicated. (C-D) The numbers of hepatocytes with fused mitochondria were counted and percentages of hepatocytes with fused mitochondria were calculated (see details in Methods). Addition of AICAR to acetaminophen or diclofenac prevented the decrease in percentage of hepatocytes with fused mitochondria. The percentages of rat hepatocytes with fused mitochondria were 94% in Control; 29% in APAP; 74% in APAP + AICAR; 63% in Diclo; 80% in Diclo + AICAR and 94% in AICAR. In human hepatocytes, they were 83% in Control; 37% in APAP; 69% in APAP + AICAR; 48% in Diclo; 68% in Diclo + AICAR and 87% in AICAR (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 4.
Effects of acetaminophen and diclofenac on expressions of mitochondrial fusion and fission proteins.
Total proteins were extracted after rat hepatocytes were treated with acetaminophen (APAP, 10mM, 24h) or diclofenac (Diclo, 250μM, 24h) in the presence or the absence of 200μM AICAR. Western blots were performed to detect protein expressions. Quantitative analyses were obtained by calculating the relative densities compared to respective controls. (A) Treatment with acetaminophen or diclofenac significantly decreased Mfn1 expression. Addition of AICAR significantly promoted Mfn1 expression in the presence of both drugs. The relative levels of Mfn1 were 34% in APAP; 66% in APAP + AICAR; 75% in Diclo and 88% in Diclo + AICAR. (B) Treatment with acetaminophen, but no diclofenac, decreased Mfn2. AICAR prevented acetaminophen-induced decrease in Mfn2. The relative levels of Mfn2 were 53% in APAP; 68% in APAP + AICAR; 93% in Diclo and 107% in Diclo + AICAR. (C) Treatment with acetaminophen or diclofenac reduced Opa1expression. AICAR prevented drug-induced decrease in Opa1. The relative levels of Opa1 were 65% in APAP; 107% in APAP + AICAR; 86% in Diclo and 113% in Diclo + AICAR. (D) Treatment with acetaminophen or diclofenac in the presence or the absence of AICAR did not have significant effects on activity of Drp1, measured by the ratio of phos-Drp1 (616) to total Drp1, and the expressions of mitochondrial structure proteins, such as (E) Hsp60, (F) Cox4 and (G) Tom20 (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 5.
Effect of AMPK activator AICAR on expressions of mitochondrial fusion and fission proteins.
Total proteins were extracted after hepatocytes were treated with AICAR (200μM, 24h). Western blots were performed to detect protein expressions. Quantitative analyses were conducted by calculating the relative densities compared to the respective controls. (A) AICAR significantly increased Mfn1 level to 145% of that in the control. (B) AICAR did not affect Mfn2 expression (95% of that in the control). (C) AICAR significantly increased Opa1 level to 132% of that in the control. (D) AICAR did not affect activity of fission protein Drp1 measured by ratio phos-Drp1 (ser 616) to total Drp1 (100% of that in the control). (E-G) AICAR did not affect expressions of mitochondrial structure proteins Hsp60, Cox4 and Tom20 (* p<0.05 and ** p<0.01). (H) Representative images of mitochondrial staining of TMRE (red) and MitoTracker Green showed that more interconnected mitochondrial network were presented in hepatocytes treated with AICAR (indicated by the white-solid arrows). In the control hepatocytes, there were less interconnected mitochondrial network, and mainly contained a mixture of fused and fragmented mitochondria (indicated by the white-dots arrows).
Fig 6.
Inhibition of mitochondrial fission by Drp1 inhibitor MDIVI1 did not prevent drug-induced mitochondrial fragmentation.
Hepatocytes were treated with acetaminophen (APAP) or diclofenac (Diclo) with or without Drp1 inhibitor, MDIVI1 (25μM). (A) Mito-staining and confocal microscope images showed that MDIVI1 did not prevent acetaminophen- and diclofenac-induced mitochondrial fragmentation in rat hepatocytes. (B) Percentages of hepatocytes with fused mitochondria were calculated and analysed (see details in Methods). MDIVI1 did not have effect on the percentages of hepatocytes with fused mitochondria in the presence of acetaminophen or diclofenac. The percentages of cells with fused mitochondria were 86% in Control; 35% in APAP; 30% in APAP + MDIVI1; 53% in Diclo; 39% in Diclo + MDIVI1 and 75% in MDIVI1. (C) Hepatocytes were treated with or without MDIVI1 (25μM, 24hr). Western blot were performed to detect Drp1 and phos-Drp1(616). Quantitative analyses were obtained by calculating the relative densities compared to respective controls. (D) Quantitative analysis for expression of Drp1. (E) Quantitative analysis for expression of phos-Drp1(616). (F) Quantitative analysis for activity of Drp1 (ratio of phos-Drp1/Drp1) (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 7.
Activation of autophagy had differentially effects on prevention of drug-induced hepatocellular injury.
Rat hepatocytes were treated with acetaminophen (APAP, 10mM, 24h) or diclofenac (Diclo, 250μM, 24h) with or without AICAR (200μM). Western blots for autophagic marker LC3II and mitophagy marker Parkin were performed. Quantitative analyses were conducted by calculating the relative densities compared to the respective controls. (A) Treatment with 200μM AICIAR alone in hepatocytes significantly increased LC3II expression to 125% of that in the control, confirming that activation of AMPK promotes autophagy activity. (B) Acetaminophen only mildly decreased LC3II expression. AICAR significantly increased LC3II level in the presence of acetaminophen. Diclofenac significantly increased LC3II levels, and addition of AICAR to diclofenac did not elevate the LC3II significantly. The relative levels of LC3II were 88% in APAP; 270% in APAP + AICAR; 338% in Diclo and 456% in Diclo + AICAR. (C) Acetaminophen significantly decreased Parkin expression to 62% of that in the control, and addition of AICAR prevented the decrease. Diclofenac significantly increased Parkin level to 132% of that in the control, and addition of AICAR to diclofenac did not elevate the Parkin significantly. The relative levels of Parkin were 62% in APAP; 120% in APAP + AICAR; 132% in Diclo and 151% in Diclo + AICAR. (D) Treatment with 200μM AICIAR alone in hepatocytes mildly, but not significantly, increased Parkin expression (119% of that in the control). (E) Autophagy activator rapamycin (2μM, 24hr) significantly improved viability in acetaminophen-treated rat hepatocytes, but not in diclofenac-treated rat hepatocytes. Treatment of rapamycin (Rap) alone did not significantly change viability. The relative viabilities were 63% in APAP; 82% in APAP + Rap; 84% in Diclo; 76% in Diclo + Rap and 97% in Rap. (F) Human hepatocytes were also treated with acetaminophen (25mM, 24h) or diclofenac (1000μM, 24h) in the presence or absence of rapamycin (2μM). Similarly, rapamycin prevented acetaminophen-induced, but not diclofenac-induced, decrease in viability. Treatment of rapamycin alone did not significantly change viability. The relative viabilities were 69% in APAP; 92% in APAP + Rap; 78% in Diclo; 79% in Diclo + Rap and 98% in Rap. (G) Representative images from confocal microscope showed that rapamycin prevented acetaminophen-induced, but not diclofenac-induced, depolarization. (H) Quantitative analyses of polarization by measuring canalicular lengths showed the rapamycin prevented acetaminophen-induced, but not diclofenac-induced, decrease in canalicular length in hepatocytes. The relative canalicular lengths were 52% in APAP; 87% in APAP + Rap; 67% in Diclo; 53% in Diclo + Rap and 106% in Rap. (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 8.
Acetaminophen and diclofenac had differential effects on AMPK activation.
Rat hepatocytes were treated with acetaminophen (APAP, 10mM, 24h) or diclofenac (Diclo, 250μM, 24h) in the presence or the absence of AICAR (200μM). Western blots of AMPK, phos-AMPK (Thr172), ACC and phos-ACC (Ser79) were performed. Ratios of phosphorylated protein to the total protein were used to measure the activation of AMPK or ACC, and percentages of the relative ratios were calculated by comparing to the respective controls. (A) Acetaminophen significantly deactivated AMPK, and addition of AICAR prevented the deactivation. The relative activities of AMPK were 53% in APAP and 120% in APAP + AICAR. (B) Acetaminophen also significantly reduced phos-ACC (ser79) levels confirming the deactivation of AMPK. AICAR prevented acetaminophen-induced decrease in phos-ACC. The relative levels of ACC were 61% in APAP and 133% in APAP + AICAR. (C) Diclofenac activated AMPK, and addition of AICAR significantly promoted the activation of AMPK. The relative activities of AMPK were 142% in Diclo and 211% in Diclo + AICAR. (D) Diclofenac also significantly increased phos-ACC (ser 79) levels confirming the activation of AMPK. Addition of AICAR increased, but not significantly, the level of phos-ACC in the presence of diclofenac. The relative levels of ACC were 132% in Diclo and 165% in Diclo + AICAR (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 9.
AMPK activation reversed drug-induced hepatocellular injury.
Rat hepatocytes were treated with 10mM acetaminophen (APAP) or 250μM diclofenac (Diclo) for a period of 8 hours. Viabilities were measured every hour and time-courses of viability were analysed for (A) acetaminophen, and (B) diclofenac. (C) ATP levels were measured after rat hepatocytes were treated with 10mM acetaminophen or 250μM diclofenac for 5 hours. Both drugs significantly decreased ATP levels to 70% and 77% of that in the control, respectively. (D-E) To investigate the reversal effect of AICAR, hepatocytes were first treated with acetaminophen (10mM for rat hepatocytes; 25mM for human hepatocytes) or diclofenac (250μM for rat hepatocytes; 1000μM for human hepatocytes) for 5 hours, hepatocytes were further incubated with or without 200μM AICAR for additional 18 hours. Viabilities were examined. After rat or human hepatocytes were incubated with acetaminophen or diclofenac for 5hr, the viabilities were significantly lower than that in respective controls. The administration of AICAR significantly improved viabilities. The relative viabilities in rat hepatocytes were 58% in APAP; 86% in APAP + AICAR; 86% in Diclo; 96% in Diclo +AICAR and 99% in AICAR. In human hepatocytes, the relative viabilities were 69% in APAP; 93% in APAP + AICAR; 78% in Diclo; 98% in Diclo +AICAR and 100% in AICAR. (F) Immunofluorescence and confocal microscope were used to examine polarized morphology with the same treatments stated in D-E. Representative images showed that addition of AICAR at 5hr post exposure of acetaminophen or diclofenac reversed drug-induced depolarization. (G) Quantitative analyses of polarization by measuring canalicular lengths showed that AICAR reversed the acetaminophen- or diclofenac-induced decrease in canalicular lengths. The relative canalicular lengths were 45% in APAP; 86% in APAP + AICAR; 47% in Diclo; 81% in Diclo +AICAR and 105% in AICAR (* p<0.05, ** p<0.01 and *** p<0.001).
Fig 10.
(A) Rat hepatocytes were treated with acetaminophen (APAP, 10mM, 24hr) alone or in addition of metformin (100μM) or 2-DG (2-deoxy-D-glucose) (200mM). Metformin and 2-DG were unable to prevent acetaminophen-induced decrease in viability. The relative viabilities were 69% in APAP; 27% in APAP + Metformin and 57% in APAP + 2-DG. (B) Metformin and 2-DG were unable to prevent acetaminophen-induced decrease in cellular ATP. The relative ATP levels were 72% in APAP; 19% in APAP + Metformin and 42% in APAP + 2-DG (* p<0.05, ** p<0.01 and *** p<0.001). (C) Model of AMPK effects on drug-induced mitochondrial and hepatocyte injury. Green lines indicate mitochondrial biogenesis, fusion/fission dynamic and autophagic degradation (mitophagy). Mitochondria undergo fusion/fission, the fused mitochondria have higher oxidative capacity. Mitochondria process fragmentation when they are damaged so as to separate the damaged mitochondria. The damaged mitochondria are sent to autophagy for degradation (mitophagy) and the undamaged ‘daughter’ mitochondria can be fused again. The biomaterial from mitophagy will be used for synthesis new mitochondria. Red lines indicate the effects of hepatotoxic drugs on mitochondria and AMPK. Acetaminophen and diclofenac cause mitochondrial dysfunction and damage. Both drugs decrease expression of mitofusion proteins resulting in mitochondrial fragmentation and decreased function. Hepatotoxic drugs may also inhibit AMPK activation and affect its downstream (i.e. mito-biogenesis and autophagy). Hepatotoxic drugs can also directly inhibit autophagy/mitophagy (red dots line). Blue lines indicate the regulatory effects of AMPK on mitochondrial function and quality control, including induction of autophagy which is responsible for removal of damaged mitochondria; increases in mito-biogenesis which is essential for maintain the quantity of mitochondria during cellular stress. The current study reveals that activation of AMPK increases mito-fusion protein level and promotes fusion to maximise the oxidative capacity during stress.