SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Drug-induced liver injury occurs in general after several weeks and is often unpredictable. It is characterized by a large spectrum of lesions that includes steatosis and phospholipidosis. Many drugs such as amiodarone and tetracycline have been reported to cause phospholipidosis and/or steatosis. In this study, acute and chronic hepatic effects of these two drugs were investigated using well-differentiated human hepatoma HepaRG cells. Accumulation of typical lipid droplets, labeled with Oil Red O, was observed in hepatocyte-like HepaRG cells after repeat exposure to either drug. Amiodarone caused the formation of additional intracytoplasmic vesicles that did not stain in all HepaRG cells. At the electron microscopic level, these vesicles appeared as typical lamellar bodies and were associated with an increase of phosphatidylethanolamine and phosphatidylcholine. A dose-dependent induction of triglycerides (TG) was observed after repeat exposure to either amiodarone or tetracycline. Several genes known to be related to lipogenesis were induced after treatment by these two drugs. By contrast, opposite deregulation of some of these genes (FASN, SCD1, and THSRP) was observed in fat HepaRG cells induced by oleic acid overload, supporting the conclusion that different mechanisms were involved in the induction of steatosis by drugs and oleic acid. Moreover, several genes related to lipid droplet formation (ADFP, PLIN4) were up-regulated after exposure to both drugs and oleic acid. Conclusion: Our results show that amiodarone causes phospholipidosis after short-term treatment and, like tetracycline, induces vesicular steatosis after repeat exposure in HepaRG cells. These data represent the first demonstration that drugs can induce vesicular steatosis in vitro and show a direct relationship between TG accumulation and enhanced expression of lipogenic genes. (HEPATOLOGY 2011;)

Drug-induced liver injury occurs infrequently after several weeks or months of treatment and usually requires metabolism of the drug to form reactive metabolites and free radicals. It is challenging to investigate because of its rarity and the lack of experimental models; consequently, its pathogenesis is poorly understood. Drug-induced liver injury encompasses a large spectrum of lesions that include steatosis and phospholipidosis resulting from disruption of lipid homeostasis.

Hepatic steatosis results from an accumulation of triglycerides (TG) in hepatocytes. It represents a reversible state of metabolic dysfunction that can possibly progress to inflammatory steatohepatitis, irreversible liver damage, fibrosis, cirrhosis, and even hepatocellular carcinoma.1, 2 Many drugs have been classified as steatogenic. Phospholipidosis is characterized by an excessive intracellular accumulation of phospholipids typified as lamellar bodies. It does not per se constitute frank toxicity,3 but it is reportedly predictive of drug or metabolite accumulation in several target tissues, and as such may be associated with toxicities. Indeed, accumulated phospholipids can interfere with cellular functions, sometimes with fatal results.4 More than 50 cationic amphiphilic drugs, including cholesterol-lowering agents, have been reported to induce phospholipidosis.5 Some of these drugs (such as amiodarone) also cause steatosis. Phospholipidosis can be predicted from in vivo and in vitro experimental models, but studies using human liver cell cultures have not demonstrated accumulation of typical lipid vesicles, a hallmark of microvesicular and macrovesicular steatosis, after treatment with steatogenic drugs. Only increased TG content has been reported.6, 7 This finding could be explained by early phenotypic changes and the short lifespan of primary normal hepatocytes and the loss of several key functions in hepatoma cell lines.8

In the present study, we used human hepatoma HepaRG cells to demonstrate the occurrence of typical features of steatosis and/or phospholipidosis after drug treatment and to identify mechanisms involved in the initiation and progression of these lesions. Differentiated HepaRG cells possess the unique ability to stably express most liver-specific functions for several weeks at confluence.9, 10 These cells were treated by tetracycline and amiodarone for either 24 hours or 14 days. Tetracycline is a well-known antibiotic that was first described as inducing steatosis in humans and rodents in 1951,11 and amiodarone, an antiarrhythmic drug, has been reported to cause both phospholipidosis12 and liver steatosis.13 Whereas intracytoplasmic lamellar bodies were observed after acute treatment with amiodarone, typical lipid droplets were found to accumulate after repeat exposure to either drug. Moreover, a number of genes known to be related to lipogenesis were found to be overexpressed after amiodarone and tetracycline treatments.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Chemicals.

Amiodarone, tetracycline, oleic acid, and Oil Red O were purchased from Sigma (St. Quentin Fallavier, France). Williams' E medium was obtained from Eurobio (Les Ulis, France). Fetal bovine serum was supplied by Perbio (Brebieres, France). [U-14C]-palmitic acid was obtained from PerkinElmer (Boston, MA).

Cell Cultures.

Human hepatoma HepaRG cells were usually seeded at a density of 2.6 × 104 cells/cm2 in Williams' E medium supplemented with 10% fetal bovine serum, 100 U/mL penicillin, 100 μg/mL streptomycin, 5 μg/mL insulin, 2 mM glutamine, and 50 μM hydrocortisone hemisuccinate.14 After 2 weeks, HepaRG cells were shifted to the same medium supplemented with 2% dimethylsulfoxide for a further 2 weeks to obtain confluent differentiated cultures containing both hepatocyte-like and biliary-like cells (around 50%–50%).15 The experimental design for acute and chronic treatments is shown in Supporting Fig. 1.

Toxicity Studies.

Experimental conditions for MTT test, apoptosis, and DCFDA assays are described in the Supporting Materials and Methods.

Oil Red O Staining.

Lipid accumulation was determined by way of Oil Red O staining, which allows detection of TG and cholesterol esters. Oil Red O was dissolved in isopropanol (0.5:100) for stock solution. After treatments, cells were washed with phosphate-buffered saline, incubated for 1 hour with Oil Red O–saturated solution (isopropanol:water, 3:2), washed in water, and observed under a phase-contrast microscope.

Transmission Electron Microscopy.

After treatment, HepaRG cells were fixed by addition of 2.5% glutaraldehyde for 30 minutes. After fixation, the specimens were rinsed with 0.2 M Na cacodylate buffer and post-fixed with 2% osmium tetroxide for 30 minutes. After further rinsing, the samples were dehydrated, infiltrated by a mixture of acetone-eponate (50/50), and embedded in DMP30-Eponate. Ultrathin sections were examined with a JEOL 100CXII electron microscope.

Lipid and Phospholipid Analyses.

Cells were homogenized in 2 mL methanol/5 mM ethylene glycol tetraacetic acid (2:1, vol/vol). Lipids were extracted in chloroform/methanol/water (2.5:2.5:2.1, vol/vol/vol). Chloroform and organic phases were evaporated to dryness. Cholesterol, cholesterol ester, and TG were analyzed by way of gas/liquid chromatography on a Focus Thermo Electron system using Zebron-1 Phenomenex–fused silica capillary columns (5 m × 0.32 mm internal diameter (i.d.), 0.50 μm film thickness).16 Oven temperature was programmed from 200°C to 350°C at a rate of 5°C per minute, and the carrier gas was hydrogen (0.5 bar). The injector and the detector were set at 315°C and 345°C, respectively. Phospholipids were analyzed by way of high-performance liquid chromatography (HPLC) on an Uptisphere 6OH analytical column (5 μm particle size, 250 × 2.1 mm) fitted with a DIOL guard column cartridge (10 × 2.1 mm) and coupled to a light scattering detector (Polymer Laboratory ELS 2100, nitrogen flow 1.8 mL/minute, evaporating temperature 50°C, and nebulizer temperature 80°C). Separation was achieved at a flow rate of 0.25 mL/minute using a gradient of B (isopropanol/water/triethylamine/acetic acid [85:15:0.014:0.5, vol/vol/vol/vol]) in A (hexane/isopropanol/triethylamin/acetic acid [82:18:0.014:0.5, vol/vol/vol/vol]) from 5% to 35% of B in 35 minutes. The variability of these methods was low, not exceeding 3% and 6.5% for gas/liquid chromatography and HPLC analyses, respectively.

Fatty Acid Oxidation.

HepaRG cells were seeded in 60-mm petri dishes. The culture medium was removed and replaced with a fresh medium containing 0.5 mM L-carnitine and 10% fat-free bovine serum albumin. [U-14C]-palmitic acid (final concentration, 1 mM; 0.05 μCi/mL) was added, and the reaction was carried out for 90 minutes at 37°C. After addition of perchloric acid (final concentration, 3%) and centrifugation at 4,000g for 10 minutes, an aliquot of the supernatant was counted for [14C]-labeled acid-soluble β-oxidation products.17

Reverse-Transcription Quantitative Polymerase Chain Reaction Analysis.

Total RNA was extracted from 106 HepaRG cells with the SV total RNA isolation system (Promega, Madison, WI). Reverse-transcription quantitative polymerase chain reaction (RT-qPCR) was performed using an SYBR Green mix.10 Primer sequences are listed in Table 1.

Table 1. Primer Sequences
GeneNameForward PrimerReverse Primer
18S18SCGCCGCTAGAGGTGAAATTCTTGGCAAATGCTTTCGCTC
ACADLLong-chain specific acyl-CoA dehydrogenaseGTCCAAACGTTTCGCTTCATTTTGGCAAAACAGTTGCTCA
ACLYATP-citrate synthaseAAGGAGTTCTTTGCCCGTCTGATTTTGCGGGGTTCGTC
ADFPAdipose differentiation-related proteinCTCATGGGTAGAGTGGAAAAGGAGCATTGGTTGGATGTTGGACAGGAGGGTGTGGCACGT
ALBAlbuminTGCTTGAATGTGCTGATGACAGGAAGGCAAGTCAGCAGGCATCTCATC
ALDBAldolase BGCATCTGTCAGCAGAATGGATAGACAGCAGCCAGGACCTT
APOBApolipoprotein BCCTCCGTTTTGGTGGTAGAGCCTAAAAGCTGGGAAGCTGA
ASAH1N-acylsphingosine amidohydrolase 1TGGTCCTGAAGGAGGATAGGTCTCCTACCCAAGTCTCAGCG
ASML3AAcid sphingomyelinase-like phosphodiesterase 3aCAGAACATCTCCAAAAGGGCAATCCTCCTCCGGCGATAG
CHREBPCarbohydrate-responsive element-binding proteinGTCACGAAGCCACACACGGAGACAAGATCCGCCTGAAC
CPT1ACarnitine O-palmitoyltransferase 1GCCTCGTATGTGAGGCAAAATCATCAAGAAATGTCGCACG
CYP2E1Cytochrome P450 2E1TTGAAGCCTCTCGTTGACCCCGTGGTGGGATACAGCCAA
CYP3A4Cytochrome P450 3A4CTTCATCCAATGGACTGCATAAATTCCCAAGTATAACACTCTACACAGACAA
ELOVL6Elongation of very long chain fatty acids protein 6ATTCATTAGGTGCCGACCACTTCGAAAAGCAGTTCAACGA
FABP1Fatty acid-binding protein 1CACCCCCTTGATATCCTTCCTTCTCCGGCAAGTACCAACT
FASNFatty acid synthaseAACTCCTGCAAGTTCTCCGAGCTCCAGCCTCGCTCTC
GDPD3Glycerophosphodiester phosphodiesterase domain containing 3GCTGAAGGCTGCTTCAAAATGTGGTTTCGAAATGGCTGAT
LPIN1Lipin-1GATGTCAATGCACCCTGAGAGTGTTTGCAATACAAAGGCG
LPLLipoprotein lipaseAATGAGGTGGCAAGTGTCCTCTCCAGAGTCTGACCGCCT
LSSLanosterol synthaseTATTTCCACAAGCGTTTCCCTGAAGCAAACTCCCCAGG
MTPMicrosomal triglyceride transferGAGCTTGTACAGCCGGTCATCAGTTGAGGATTGCTGGTCA
PLIN4Perilipin-4CAGATGCAGGAAGCATCAAAGCGACTAAAAGGCACTCTGG
PPARAPeroxisome proliferator-activated receptor αCATTACGGAGTCCACGCGTACCAGCTTGAGTCGAATCGTT
PPARGPeroxisome proliferator-activated receptor γGATGACAGCGACTTGGCAACTTCAATGGGCTTCACATTCA
PPARGC1APeroxisome proliferator-activated receptor γ coactivator 1-αCTGCTAGCAAGTTTGCCTCAAGTGGTGCAGTGACCAATCA
SCDAcyl-CoA desaturaseGACGATGAGCTCCTGCTGTTCTCTGCTACACTTGGGAGCC
SLC27A4Long-chain fatty acid transport protein 4CCTCCTTCCGTAGCTCTGTCGAAGGAACTGCCCCTGTATG
SOAT1Sterol O-acyltransferase 1ATTCCTCTGCCTCTGCTGTCGCTGTCAAAGTCCAGGGAAA
SREBP1Sterol regulatory element-binding protein 1AGGGAAGTCACTGTCTTGGTTGCTGCTGACCGACATCGAA
THRSPThyroid hormone-inducible hepatic proteinAGGCCTTTCTGCTCTCATCAAAATGACGGGACAAGTTTGG

Microarray Analysis.

Experimental conditions are given in the Supporting Materials and Methods as described.18

Western Blot Analysis.

Total cellular protein extracts were obtained by way of cell lysis. Fifty micrograms of protein underwent electrophoresis, and immunoblotting was performed with anti-ADFP (Abcam, Cambridge), anti-PPARG (Dharmacon), anti-CYP3A4 (Millipore), anti-CYP2E1 (Oxford Biomedical Research), and anti-HSC70 (Tebu-bio) antibodies.

Statistical Analysis.

Results are expressed as the mean ± SD of three independent experiments. The Mann-Whitney U test was applied to compare data between drug-treated and corresponding control cultures. Data were considered significantly different at P < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Demonstration of Lipid Accumulation by Way of Oil Red O Staining.

Preliminary experiments were performed to compare toxic effects of tetracycline and amiodarone after acute and repeat treatments to select several nontoxic and subtoxic concentrations of each drug for further studies (Supporting Results and Supporting Fig. 2). Thus, HepaRG cells were exposed to 50 μM tetracycline, 20 μM amiodarone, and 500 μM oleic acid (a positive in vitro steatosis inducer19, 20) for 24 hours or 14 days and stained with Oil Red O to detect intracytoplasmic lipid droplets. Oleic acid induced formation of droplets in hepatocyte-like cells after both acute and repeat incubation, with the vesicles being enlarged after 14 days (Fig. 1c,d). Similarly, lipid vesicles were also observed in hepatocyte-like cells treated with tetracycline (Fig. 1e,f), but staining was more important after 14 days than 24 hours. Amiodarone also caused accumulation of Oil Red O–stained vesicles in hepatocyte-like cells, but only after 14 days (Fig. 1h). In addition, numerous small unstained vesicles were observed in both HepaRG cell types after either 24-hour or 14-day amiodarone treatment, suggesting phospholipid accumulation (Fig. 1g,h).

thumbnail image

Figure 1. Determination of lipid accumulation by way of Oil Red O staining in HepaRG cells treated for 24 hours or 14 days with tetracycline and amiodarone. HepaRG cells were exposed to (A,B) solvent, (C,D) 500 μM oleic acid, (E,F) 50 μM tetracycline, or (G,H) 20 μM amiodarone for 24 hours (A,C,E,G) or 14 days (B,D,F,H). Lipid accumulation was then visualized by way of Oil Red O staining, which allowed detection of TG and cholesterol esters. HepaRG cells were observed and photographed under a phase-contrast microscope (A-F, original magnification ×20; G and H, original magnification ×40). Arrows indicate unstained vesicles.

Download figure to PowerPoint

Electron Microscopic Examination of HepaRG Cells.

Electron microscopic detection of intracytoplasmic lamellar bodies is thought to be the most reliable method for the identification of phospholipidosis. Exposure of HepaRG cells to 20 μM amiodarone for either 24 hours or 14 days led to the formation of typical concentric lamellar structures corresponding to excessive intracellular accumulation of phospholipids in lysosomes of both hepatocyte-like and biliary-like cells. These lamellar bodies appeared larger after 14-day repeat treatments and mixed with clear vesicles typical of lipid droplets (Fig. 2c,d). As expected, no lamellar bodies were found in tetracycline-treated HepaRG cells (Fig. 2b). Only lipid droplets were observed; they were already detected after 24 hours and were enlarged after 14 days (Fig. 2b). Interestingly, mitochondria, nuclei, and endoplasmic reticulum remained morphologically unchanged.

thumbnail image

Figure 2. Electron microscopic analysis of HepaRG cells after chronic exposure to tetracycline and amiodarone. HepaRG cells were incubated with (A) solvent, (B) 50 μM tetracycline, or (C,D) 20 μM amiodarone for 14 days and analyzed by way of electron microscopy (A-C, original magnification ×2,700; D, original magnification ×10,000). BC, bile canaliculus; L, lysosome; LB, lamellar body; LD, lipid droplet; M, mitochondrion; N, nucleus.

Download figure to PowerPoint

Lipid Quantification.

Cholesterol and neutral lipids (TG and cholesterol esters) were quantified by way of gas/liquid chromatography (Fig. 3). Whereas tetracycline caused no significant changes in TG content after 24 hours, a six-fold increase was induced by a 50 μM concentration after 14 days. By contrast, cholesterol and cholesterol esters content remained unchanged. A dose-dependent increase in TG content was also observed, and cholesterol esters were slightly augmented in HepaRG cells treated by amiodarone for 14 days. In addition, phospholipids (phosphatidylethanolamine, phosphatidylcholine, sphingomyelin, phosphatidylserine, and phosphatidylinositol) were measured by way of HPLC in HepaRG cells treated with 20 μM amiodarone for 24 hours or 14 days (Fig. 4). Whereas no significant change was observed in phospholipid content after acute exposure, phosphatidylethanolamine and phosphatidylcholine levels were strongly enhanced, and sphingomyelin, phosphatidylserine, and phosphatidylinositol levels were slightly augmented after 14 days.

thumbnail image

Figure 3. Quantification of TG, cholesterol esters, and cholesterol in HepaRG cells after treatment by various concentrations of tetracycline and amiodarone. HepaRG cells were exposed for (A,C) 24 hours or (B,D) 14 days to solvent (control) or different concentrations of (A,B) tetracycline or (C,D) amiodarone. Lipids were extracted (chloroform/methanol/water in the presence of internal standards) and TG, cholesterol, and cholesterol esters were analyzed by way of gas/liquid chromatography. Results are expressed as the mean ± SD of three independent experiments. *P < 0.05 versus control.

Download figure to PowerPoint

thumbnail image

Figure 4. Quantification of phospholipids in HepaRG cells after acute and chronic exposure to amiodarone. HepaRG cells were exposed to solvent (control) or 20 μM amiodarone for 24 hours or 14 days. Phospholipids were extracted (chloroform/methanol/water) and phosphatidylethanolamine (PE), phosphatidylcholine (PC), sphingomyelin (SM), phosphatidylserine (PS), and phosphatidylinositol (PI) were measured by way of HPLC. Results are expressed as the mean ± SD of three independent experiments. *P < 0.05 versus control.

Download figure to PowerPoint

Fatty Acid Oxidation Inhibition.

Impairment of mitochondrial fatty acid oxidation (FAO) is considered one of the major mechanisms of liver steatosis.21 FAO was evaluated by measuring [14C]-labeled acid-soluble β-oxidation products in HepaRG cells after 24-hour and 14-day treatments using either 20 μM tetracycline or 50 μM amiodarone (Fig. 5). A 20% diminution of FAO was observed after both acute and chronic amiodarone treatments, and only after chronic tetracycline exposure.

thumbnail image

Figure 5. Determination of FAO in HepaRG cells after acute and chronic exposure to tetracycline and amiodarone. HepaRG cells were incubated with solvent (control), (A) 50 μM tetracycline, or (B) 20 μM amiodarone for 24 hours or 14 days. FAO was evaluated by measuring [14C]-labeled acid-soluble β-oxidation products generated by cells after 90 minutes of incubation with [U-14C]-palmitic acid. Basal FAO was estimated to represent 626.8 nmol/mg protein/hour. Results are expressed as the mean ± SD of three independent experiments. *P < 0.05 versus control.

Download figure to PowerPoint

Modulation of Messenger RNA Levels of Genes Involved in Lipid Metabolism.

To characterize gene expression changes associated with induction of phospholipidosis and steatosis, the transcriptome of HepaRG cells was analyzed after 24-hour and 14-day treatments with 20 μM amiodarone using pangenomic oligonucleotide microarrays. Significantly modulated genes were extracted with a fold change >1.5 or <−1.5 and P ≤ 0.01 as filters. Their total numbers reached 547 and 594 with up-regulated genes representing 48% and 44%, after 24-hour and 14-day exposure, respectively (Supporting Tables 1 and 2); 176 genes were in common at the two time points. Functional analysis revealed that expression of many genes involved in the regulation of lipid metabolism (including ACOT12, ADFP, ALDH3A1, APOA2, FASN, MOGAT1, SREBP1, and THRSP) or related to phospholipidosis (such as LSS, LPIN1, ASML3A, and GDPD3) was significantly altered. Various genes regulating growth/proliferation, cell death, assembly/organization, and inflammation were also substantially deregulated.

To validate and complete this microarray analysis, changes in the expression of 29 genes, which are key players in lipid metabolism and/or liver-specific functions, were further examined by way of RT-qPCR in HepaRG cells exposed to several concentrations of amiodarone (5-20 μM), tetracycline (10-100 μM), and oleic acid (100-500 μM) for 24 hours or 14 days. The data are displayed in Table 2. Importantly, RT-qPCR data obtained with 20 μM amiodarone fully confirmed corresponding microarray results. It is noteworthy that many deregulated genes were common to 20 μM amiodarone after 24-hour and 14-day treatments and to 100 μM tetracycline after 24-hour treatment. Two genes involved in fatty acid transport were up-regulated, SLC27A4 by both drugs and FABP1 by tetracycline after repeat treatments. Only one gene involved in mitochondrial biogenesis, PPARGC1A, was overexpressed by both amiodarone and tetracycline. By contrast, several genes involved in de novo lipogenesis were modulated by the two drugs. Transcripts of SREBP1, THRSP, ACLY, FASN, and SCD1 were significantly augmented after 24-hour and/or 14-day treatments by amiodarone. SREBP1 and PPARG were also up-regulated, whereas THRSP was down-regulated by 100 μM tetracycline after 24-hour treatment. However, THRSP was overexpressed after 14-day exposure to 10 μM tetracycline. Expression of genes involved in cholesterol metabolism was also altered; thus, transcript levels of LSS were increased after 24-hour amiodarone and 14-day tetracycline treatments. In addition, SOAT1 and LPIN1 were induced by 20 μM amiodarone after both short- and long-term treatments. Moreover, genes involved in the formation of lipid droplets, particularly PLIN4 and ADFP, were overexpressed by high concentrations of both drugs, regardless of the duration of treatment. In addition, LPL as well as GDPD3 and ASML3A, two genes involved in phospholipids degradation, were up-regulated after long-term exposure to amiodarone. Finally, the two test CYP genes were also modulated: transcripts of CYP2E1 were decreased by both drugs, whereas those of CYP3A4 were induced only by amiodarone. No changes were noticed in ALB or ALDB transcripts regardless of the drug treatment.

Table 2. Effects of Acute (24 Hours) and Repeat (14 Days) Exposure of HepaRG Cells to Amiodarone, Tetracycline, and Oleic Acid on Expression of Messenger RNAs Encoding Lipid Metabolism and Liver-Specific Genes
 AmiodaroneTetracyclineOleic Acid
24 Hours14 Days24 Hours14 Days24 Hours14 Days
10 μM20 μM5 μM10 μM20 μM50 μM100 μM10 μM50 μM250 μM500 μM250 μM500 μM
  1. Results are expressed as fold of the value found in control cells arbitrarily set at 1. Data are expressed as the mean ± SD of three independent experiments. Data with P < 0.05 and fold change >1.5 (up-regulated genes) or <0.67 (down-regulated genes) are indicated in bold. No significant modulation of any gene was found with 5 μM amiodarone and 10 μM tetracycline after 24-hour exposure and with 100 μM oleic acid after 24-hour and 14-day exposures.

Fatty acid transport and very low-density lipoprotein synthesis
 APOB0.90 ± 0.020.98 ± 0.081.14 ± 0.180.93 ± 0.200.98 ± 0.210.97 ± 0.120.96 ± 0.141.11 ± 0.111.01 ± 0.101.06 ± 0.281.10 ± 0.281.06 ± 0.241.01 ± 0.25
 FABP11.07 ± 0.230.91 ± 0.211.01 ± 0.190.81 ± 0.080.73 ± 0.050.89 ± 0.140.91 ± 0.151.70 ± 0.481.55 ± 0.640.93 ± 0.041.05 ± 0.101.27 ± 0.151.12 ± 0.17
 MTP0.94 ± 0.170.86 ± 0.211.02 ± 0.181.05 ± 0.301.02 ± 0.220.99 ± 0.110.86 ± 0.161.07 ± 0.100.98 ± 0.061.00 ± 0.260.97 ± 0.261.06 ± 0.300.99 ± 0.18
 SLC27A41.04 ± 0.111.53 ± 0.171.10 ± 0.201.05 ± 0.041.56 ± 0.301.12 ± 0.051.66 ± 0.331.06 ± 0.091.12 ± 0.080.99 ± 0.11.02 ± 0.230.97 ± 0.211.10 ± 0.11
Fatty acid oxidation and mitochondrial biogenesis
 ACADL0.79 ± 0.140.79 ± 0.121.10 ± 0.140.94 ± 0.151.25 ± 0.021.08 ± .071.05 ± 0.121.02 ± 0.060.89 ± 0.081.01 ± 0.201.05 ± 0.261.07 ± 0.311.10 ± 0.30
 CPT1A0.74 ± 0.150.99 ± 0.211.27 ± 0.101.13 ± 0.361.27 ± 0.261.21 ± 0.091.37 ± 0.151.25 ± 0.171.34 ± 0.121.32 ± 0.301.58 ± 0.221.19 ± 0.341.52 ± 0.20
 PPARA1.10 ± 0.061.14 ± 0.031.02 ± 0.251.13 ± 0.311.28 ± 0.261.01 ± 0.010.97 ± 0.071.03 ± 0.070.98 ± 0.110.95 ± 0.301.00 ± 0.280.94 ± 0.170.99 ± 0.26
 PPARGC1A1.20 ± 0.261.81 ± 0.221.64 ± 0.361.54 ± 0.202.18 ± 0.441.08 ± 0.071.72 ± 0.070.84 ± 0.140.88 ± 0.080.97 ± 0.080.95 ± 0.110.96 ± 0.071.14 ± 0.24
De novo lipogenesis
 ACLY1.28 ± 0.221.63 ± 0.151.14 ± 0.241.00 ± 0.061.19 ± 0.280.87 ± 0.040.93 ± 0.121.23 ± 0.061.18 ± 0.210.88 ± 0.110.92 ± 0.181.11 ± 0.270.94 ± 0.19
 CHREBP0.87 ± 0.070.80 ± 0.091.23 ± 0.191.10 ± 0.171.17 ± 0.161.14 ± 0.261.25 ± 0.241.21 ± 0.121.24 ± 0.160.93 ± 0.111.07 ± 0.090.94 ± 0.221.11 ± 0.26
 ELOVL60.91 ± 0.101.17 ± 0.131.09 ± 0.120.96 ± 0.021.01 ± 0.070.99 ± 0.071.18 ± 0.061.36 ± 0.081.28 ± 0.050.92 ± 0.120.93 ± 0.151.08 ± 0.160.99 ± 0.22
 FASN0.99 ± 0.101.21 ± 0.091.10 ± 0.100.98 ± 0.031.50 ± 0.051.00 ± 0.061.02 ± 0.121.15 ± 0.090.98 ± 0.050.66 ± 0.040.67 ± 0.040.96 ± 0.061.17 ± 0.05
 PPARG1.07 ± 0.061.35 ± 0.201.10 ± 0.311.05 ± 0.061.40 ± 0.521.16 ± 0.041.97 ± 0.381.24 ± 0.081.24 ± 0.091.06 ± 0.111.05 ± 0.151.06 ± 0.261.08 ± 0.25
 SCD11,47 ± 0,031.67 ± 0.141.20 ± 0.221.18 ± 0.101.32 ± 0.180.98 ± 0.081.05 ± 0.071.40 ± 0.171.09 ± 0.150.79 ± 0.110.74 ± 0.091.08 ± 0.120.93 ± 0.07
 SREBP11.19 ± 0.051.56 ± 0.081.30 ± 0.231.30 ± 0.271.60 ± 0.221.22 ± 0.061.80 ± 0.211.12 ± 0.051.04 ± 0.011.01 ± 0.231.02 ± 0.241.03 ± 0.021.08 ± 0.09
 THRSP2.37 ± 0.282.70 ± 0.541.33 ± 0.071.45 ± 0.261.77 ± 0.230.72 ± 0.100.41 ± 0.231.69 ± 0.321.38 ± 0.190.74 ± 0.110.71 ± 0.060.64 ± 0.100.33 ± 0.02
Cholesterol and glycero-lipid metabolism
 LPIN11.41 ± 0.371.62 ± 0.101.35 ± 0.151.18 ± 0.171.63 ± 0.261.05 ± 0.101.40 ± 0.421.19 ± 0.241.11 ± 0.300.87 ± 0.090.89 ± 0.230.97 ± 0.210.94 ± 0.10
 LSS1.21 ± 0.272.04 ± 0.241.12 ± 0.301.02 ± 0.161.16 ± 0.181.00 ± 0.151.03 ± 0.251.41 ± 0.061.50 ± 0.211.03 ± 0.100.90 ± 0.071.08 ± 0.201.06 ± 0.20
 SOAT11.09 ± 0.211.37 ± 0.061.27 ± 0.351.13 ± 0.171.75 ± 0.331.10 ± 0.081.23 ± 0.201.16 ± 0.081.15 ± 0.080.97 ± 0.130.93 ± 0.241.06 ± 0.250.92 ± 0.08
Lipid hydrolysis and formation of droplets
 ADFP1.12 ± 0.101.83 ± 0.481.31 ± 0.061.61 ± 0.292.36 ± 0.551.15 ± 0.301.60 ± 0.181.42 ± 0.141.60 ± 0.291.85 ± 0.152.43 ± 0.421.57 ± 0.201.86 ± 0.27
 LPL1.07 ± 0.151.37 ± 0.051.20 ± 0.160.99 ± 0.031.62 ± 0.130.96 ± 0.020.97 ± 0.101.18 ± 0.240.98 ± 0.080.88 ± 0.020.85 ± 0.190.92 ± 0.140.85 ± 0.13
 PLIN41.44 ± 0.702.60 ± 0.651.35 ± 0.141.25 ± 0.201.62 ± 0.231.21 ± 0.122.06 ± 0.651.22 ± 0.181.47 ± 0.111.41 ± 0.221.82 ± 0.351.16 ± 0.181.53 ± 0.16
Degradation of phospholipids
 ASAH11.05 ± 0.081.16 ± 0.161.06 ± 0.221.05 ± 0.141.20 ± 0.151.05 ± 0.030.97 ± 0.051.08 ± 0.011.14 ± 0.010.98 ± 0.140.99 ± 0.250.99 ± 0.220.95 ± 0.14
 ASML3A1.04 ± 0.081.09 ± 0.021.25 ± 0.111.19 ± 0.221.51 ± 0.180.94 ± 0.051.03 ± 0.090.92 ± 0.040.92 ± 0.050.86 ± 0.220.79 ± 0.180.94 ± 0.250.87 ± 0.14
 GDPD31.23 ± 0.401.31 ± 0.201.58 ± 0.141.53 ± 0.111.61 ± 0.081.23 ± 0.521.25 ± 0.631.06 ± 0.170.95 ± 0.010.98 ± 0.071.02 ± 0.120.91 ± 0.100.95 ± 0.08
Cytochromes P450
 CYP2E10.66 ± 0.100.43 ± 0.160.56 ± 0.100.36 ± 0.110.17 ± 0.070.66 ± 0.110.44 ± 0.091.19 ± 0.230.53 ± 0.121.08 ± 0.171.18 ± 0.061.58 ± 0.261.89 ± 0.41
 CYP3A41.42 ± 0.281.85 ± 0.572.13 ± 0.122.30 ± 0.392.97 ± 0.410.98 ± 0.090.94 ± 0.081.06 ± 0.131.25 ± 0.060.68 ± 0.080.62 ± 0.080.74 ± 0.050.78 ± 0.04
Liver-specific proteins
 ALB1.10 ± 0.120.88 ± 0.081.02 ± 0.150.95 ± 0.070.91 ± 0.191.10 ± 0.280.88 ± 0.211.02 ± 0.111.18 ± 0.040.90 ± 0.090.85 ± 0.200.81 ± 0.020.73 ± 0.06
 ALDB1.09 ± 0.140.77 ± 0.130.83 ± 0.250.83 ± 0.180.75 ± 0.290.88 ± 0.130.67 ± 0.161.03 ± 0.101.11 ± 0.110.73 ± 0.100.79 ± 0.140.63 ± 0.070.48 ± 0.02

Comparison with oleic acid–overloaded HepaRG cells revealed that, as observed with the two drugs, genes involved in the formation of lipid droplets (ADFP and PLIN4) were up-regulated by 500 μM oleic acid at the two time points. However, genes involved in de novo lipogenesis were markedly (FASN, THRSP) or slightly (SCD1) down-regulated, whereas CPT1A involved in FAO was increased. In addition, CYP3A4 transcripts were reduced after 24 hours and CYP2E1 levels were increased after 14 days of oleic acid overload. ALDB transcripts were also decreased after repeat oleic acid exposure.

Importantly, the expression of several genes was also analyzed at the protein level by way of western blotting (Fig. 6). For all of them (PPARG, ADFP, CYP2E1, and CYP3A4), changes in protein content followed messenger RNA (mRNA) modifications after treatment with either drug or oleic acid.

thumbnail image

Figure 6. Expression of PPARG, ADFP, CYP2E1, and CYP3A4 proteins in HepaRG cells after acute or chronic exposure to amiodarone, tetracycline, and oleic acid. HepaRG cells were incubated with solvent (control), 20 μM amiodarone, 50 or 100 μM tetracycline, and 500 μM oleic acid for either 48 hours or 14 days. Equal amounts of total protein lysates were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, and expression of PPARG, ADFP, CYP2E1, CYP3A4, and HSC70 was determined by way of immunoblotting using appropriate antibodies and developed with an electrochemiluminescence reagent. Quantification of western blots was realized and protein expression was normalized with HSC70 and expressed as fold change (FC) of the value found in control cells arbitrarily set at 1.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Liver steatosis is characterized by excessive accumulation of neutral lipids, mainly TG, into intracytoplasmic macrovesicles and microvesicles that are induced by various factors, including several drugs.13 Previous studies have reported accumulation of TG in liver cell cultures after drug treatment but have not shown formation of lipid droplets.6, 7In vitro generation of lipid droplets has been described only after medium addition of fatty acids, such as monounsaturated oleic acid.19, 20 We report accumulation of TG and formation of lipid droplets in human hepatoma HepaRG cells after repeat treatment with two prototypical steatogenic drugs: tetracycline and amiodarone. Generation of fatty liver cells was associated with increased expression of several genes involved in lipogenesis. Accumulation of numerous lipid vesicles in most hepatocyte-like HepaRG cells was associated with a nearly six-fold increase in TG content after a 14-day exposure to either 50 μM tetracycline or 20 μM amiodarone. Microvesicular steatosis has been reported in patients with high serum and liver (1-2 mM) concentrations of amiodarone22, 23 and tetracycline11, 24 after chronic use in humans. Compared with these in vivo data, it appears that steatosis can be induced in HepaRG cells at relatively low drug concentrations.

Several mechanisms have been implicated in drug-induced steatosis. Inhibition of mitochondrial FAO is considered one of the major mechanisms of hepatosteatosis and has been demonstrated with higher concentrations of tetracycline (> 250 μM) and amiodarone (> 100 μM) in isolated mitochondria in mice and humans.11, 13 Only a weak inhibition, not exceeding 20%, was observed in HepaRG cells—mainly after chronic exposure to either drug—by measuring oxidation products of palmitic acid, and no related gene was found to exhibit altered expression. Several other mechanisms can be responsible for TG accumulation in liver, including reduced mitochondrial transition pore activity, de novo lipogenesis, and alteration of fatty acid uptake.25 Our transcriptional analysis showed that expression of many genes related to lipid metabolism was altered after drug treatment. In particular, several genes known to be related to lipogenesis (the lipogenic transcription factor SREBP1, FASN, and ACLY) were up-regulated after acute and/or long-term exposure to amiodarone. Levels of SREBP1 mRNA and PPARG mRNA and protein were also enhanced after acute treatment with 100 μM tetracycline. Activation of PPARG has been described as an important mechanism of lipid deposition.7 Indeed, several ligands of PPARG have been shown to cause fat accumulation by a nuclear receptor-dependent mechanism in human hepatocytes, whereas they had no significant effects in HepG2 cells.7 In addition, an increase in THRSP mRNAs was found after short- and long-term exposure to amiodarone and after chronic exposure to tetracycline. Moreau et al.26 have recently shown that THRSP overexpression in human hepatocytes promoted an enhancement of lipogenesis through activation of PXR and/or CAR. Notably, opposite deregulation of lipogenic genes was observed in oleic acid–overloaded HepaRG cells. Indeed, FASN, SCD1, and THRSP were down-regulated, whereas CPT1A involved in FAO was up-regulated. Similarly, an inhibition of lipogenesis was observed in immortalized human hepatocytes after a 7-day overload with 50 μM oleic acid.19 Taken altogether, these results support the involvement of different mechanisms in steatosis induced by drugs and oleic acid. In oleic acid–overloaded HepaRG cells, down-regulation of lipogenic genes could be regarded as negative feedback regulation of lipid accumulation. Interestingly, the opposite effects on lipogenesis were observed in amiodarone- and/or tetracycline-treated mouse and rat livers on transcriptomic analysis; the lipogenesis pathway was induced in mouse liver as in HepaRG cells, whereas it was inhibited in rats.27–29 Furthermore, the significantly increased levels of SOAT1 transcripts could be related to the higher cholesterol esters content observed in 14-day treated cells by 20 μM amiodarone.

As expected, several genes encoding proteins involved in the formation of vesicles, especially PLIN4 and ADFP, were overexpressed by high concentrations of the two test drugs and oleic acid. Importantly, ADFP was also augmented at the protein level. PLIN4 transcripts have been shown to augment with increasing liver fat content30 and ADFP was found up-regulated in human and mouse fatty liver.31 A high-fat diet increased expression of ADFP through PPARG activation, followed by induction of liver steatosis.31 PLIN4 and ADFP coat lipid droplets and protect TG from the lipolytic action of LPL.32, 33 However, an increase of LPL transcripts was observed in HepaRG cells after chronic exposure to 20 μM amiodarone. Although expressed at lower levels than in adipose tissue, LPL caused hepatic steatosis and insulin resistance when specifically overexpressed in mouse liver.34 In humans, LPL expression was reported to be increased in proportion to liver fat content.35 No genes involved in the formation of such droplets were found altered in amiodarone-treated HepG2 cells.36 Only some genes related to lipid and cholesterol metabolism (including ELOVL6, SCD, LSS, and LPIN1) were induced.36

Interestingly, CYP2E1 was down-regulated in a concentration-dependent manner by both acute and repeat treatment by either drug, whereas it was increased by the addition of oleic acid after 14 days. These data were confirmed by way of western blotting analysis. Divergent effects of steatosis on CYP2E1 have been reported. Indeed, whereas CYP2E1 expression and activity were reduced in fat-overloaded hepatocytes,37 they were increased in patients with nonalcoholic steatohepatitis,38, 39 which is associated with steatosis and inflammation. Release of cytokines, rather than fat accumulation, has been suggested to be ultimately responsible for the changes in CYP2E1 expression occurring in nonalcoholic steatohepatitis.38, 40 Moreover, an increase of CYP3A4 mRNA and protein was found after short- and long-term amiodarone exposure. However, a decrease in corresponding activity (Supporting Fig. 3) was observed, suggesting the occurrence of different transcriptional and posttranscriptional regulatory mechanisms. Notably, tetracycline was ineffective for CYP3A4 expression. Previous studies have shown that the formation of the main amiodarone metabolite, the dealkylated metabolite desethylamiodarone, is catalyzed by CYP3A441 and that amiodarone, but not its metabolite, is a weak inhibitor of CYP3A4-mediated activity.42

In addition to steatosis, amiodarone, like other cationic amphiphilic drugs, induced phospholipidosis, identified as intracellular lamellar inclusion bodies formed by excessive accumulation of phospholipids. These lamellar bodies were observed in both hepatocyte-like and biliary-like HepaRG cells in agreement with the fact that phospholipidosis can be visualized in various hepatic43 and nonhepatic cell types.44 Up-regulation of the fatty acid biosynthesis-related gene SCD suggested an enhanced synthesis of phospholipids in HepaRG cells treated with amiodarone for 24 hours. Furthermore, an induction of cholesterol synthesis, supported by overexpression of LSS, was observed, representing an indirect mechanism of phospholipidosis.36 Another gene LPIN1 was specifically overexpressed in HepaRG cells after both acute and repeat amiodarone exposure. LPIN1 encodes the phosphatidate phosphatase-1 enzyme, which converts phosphatidate to diacylglycerol. The resulting diacylglycerol serves as substrate for the synthesis of triacylglycerol as well as phosphatidylethanolamine and phosphatidylcholine.45 Importantly, a strong increase of phosphatidylethanolamine and phosphatidylcholine was observed in HepaRG cells treated with amiodarone for 14 days. In addition, genes involved in phospholipid degradation (GDPD3 and ASML3A) were also up-regulated after 14 days. GDPD3 and LSS were similarly found overexpressed in amiodarone-treated HepG2 cells.36, 46 Some of these genes (ASML3A, GDPD3, LPL) were modulated specifically after repeat exposure with amiodarone; they likely corresponded to a defense mechanism to reduce phospholipid accumulation and therefore could represent potential biomarkers of drug-induced phospholipidosis.

In conclusion, our study provides the first in vitro demonstration of drug-induced vesicular steatosis after repeat treatments. This vesicular steatosis was characterized by an excessive accumulation of TG together with the appearance of Oil Red O–stained lipid vesicles and overexpression of several genes involved in lipogenesis and droplet formation. These data provide new insight into the mechanisms of drug-induced TG accumulation in human hepatocytes and suggest that the HepaRG cell model represents a unique tool for estimating the ability of new drugs to induce steatosis and/or phospholipidosis, as well as other liver injuries, during their early development stage. This cell model should also be appropriate for investigations on steatosis reversibility as well as late steatosis stages leading to steatohepatitis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

We acknowledge Agnès Burel from the Electron Microscopy platform (Rennes, France) and Justine Bertrand-Michel from the Lipidomic platform (Toulouse, France) for technical help.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  8. Supporting Information

Additional Supporting Information may be found in the online version of this article.

FilenameFormatSizeDescription
HEP_24290_sm_suppinfo.doc1227KSupporting Information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.