Drug-induced hepatic steatosis in absence of severe mitochondrial dysfunction in HepaRG cells: proof of multiple mechanism-based toxicity.

Steatosis is a liver lesion reported with numerous pharmaceuticals. Prior studies showed that severe impairment of mitochondrial fatty acid oxidation (mtFAO) constantly leads to lipid accretion in liver. However, much less is known about the mechanism(s) of drug-induced steatosis in the absence of severe mitochondrial dysfunction, although previous studies suggested the involvement of mild-to-moderate inhibition of mtFAO, increased de novo lipogenesis (DNL), and impairment of very low-density lipoprotein (VLDL) secretion. The objective of our study, mainly carried out in human hepatoma HepaRG cells, was to investigate these 3 mechanisms with 12 drugs able to induce steatosis in human: amiodarone (AMIO, used as positive control), allopurinol (ALLO), D-penicillamine (DPEN), 5-fluorouracil (5FU), indinavir (INDI), indomethacin (INDO), methimazole (METHI), methotrexate (METHO), nifedipine (NIF), rifampicin (RIF), sulindac (SUL), and troglitazone (TRO). Hepatic cells were exposed to drugs for 4 days with concentrations decreasing ATP level by less than 30% as compared to control and not exceeding 100 × Cmax. Among the 12 drugs, AMIO, ALLO, 5FU, INDI, INDO, METHO, RIF, SUL, and TRO induced steatosis in HepaRG cells. AMIO, INDO, and RIF decreased mtFAO. AMIO, INDO, and SUL enhanced DNL. ALLO, 5FU, INDI, INDO, SUL, RIF, and TRO impaired VLDL secretion. These seven drugs reduced the mRNA level of genes playing a major role in VLDL assembly and also induced endoplasmic reticulum (ER) stress. Thus, in the absence of severe mitochondrial dysfunction, drug-induced steatosis can be triggered by different mechanisms, although impairment of VLDL secretion seems more frequently involved, possibly as a consequence of ER stress.

Endotoxin regulates matrix genes increasing reactive oxygen species generation by intercellular communication between palmitate-treated hepatocyte and stellate cell.

Previous studies have shown that gut-derived bacterial endotoxins contribute in the progression of simple steatosis to steatohepatitis, although the mechanism(s) remains inaccurate to date. As hepatic stellate cells (HSC) play a pivotal role in the accumulation of excessive extracellular matrix (ECM), leading to collagen deposition, fibrosis, and perpetuation of inflammatory response, an in vitro model was developed to investigate the crosstalk between HSC and hepatocytes (human hepatoma cell) pretreated with palmitate. Bacterial lipopolysaccharide (LPS) stimulated HSC with phosphorylation of the p38 mitogen-activated protein kinase/NF-κB pathway, while several important pro-inflammatory cytokines were upregulated in the presence of hepatocyte-HSC. Concurrently, fibrosis-related genes were regulated by palmitate and the inflammatory effect of endotoxin where cells were more exposed or sensitive to reactive oxygen species (ROS). This interaction was accompanied by increased expression of the mitochondrial master regulator, proliferator-activated receptor gamma coactivator alpha, and a cytoprotective effect of the agent N-acetylcysteine suppressing ROS production, transforming growth factor-ß1, and tissue inhibitor of metalloproteinase-1. In summary, our results demonstrate that pro-inflammatory mediators LPS-induced promote ECM rearrangement in hepatic cells transcriptionally committed to the regulation of genes encoding enzymes for fatty acid metabolism in light of differences that might require an alternative therapeutic approach targeting ROS regulation.

Drug-Induced Alterations of Mitochondrial DNA Homeostasis in Steatotic and Nonsteatotic HepaRG Cells.

Although mitochondriotoxicity plays a major role in drug-induced hepatotoxicity, alteration of mitochondrial DNA (mtDNA) homeostasis has been described only with a few drugs. Because it requires long drug exposure, this mechanism of toxicity cannot be detected with investigations performed in isolated liver mitochondria or cultured cells exposed to drugs for several hours or a few days. Thus, a first aim of this study was to determine whether a 2-week treatment with nine hepatotoxic drugs could affect mtDNA homeostasis in HepaRG cells. Previous investigations with these drugs showed rapid toxicity on oxidative phosphorylation but did not address the possibility of delayed toxicity secondary to mtDNA homeostasis impairment. The maximal concentration used for each drug induced about 10% cytotoxicity. Two other drugs, zalcitabine and linezolid, were used as positive controls for their respective effects on mtDNA replication and translation. Another goal was to determine whether drug-induced mitochondriotoxicity could be modulated by lipid overload mimicking nonalcoholic fatty liver. Among the nine drugs, imipramine and ritonavir induced mitochondrial effects suggesting alteration of mtDNA translation. Ritonavir toxicity was stronger in nonsteatotic cells. None of the nine drugs decreased mtDNA levels. However, increased mtDNA was observed with five drugs, especially in nonsteatotic cells. The mtDNA levels could not be correlated with the expression of key factors involved in mitochondrial biogenesis, such as peroxisome proliferator-activated receptor-γ coactivator 1α (PGC1α), PGC1ß, and AMP-activated protein kinase α-subunit. Hence, drug-induced impairment of mtDNA translation might not be rare, and increased mtDNA levels could be a frequent adaptive response to slight energy shortage. Nevertheless, this adaptation could be impaired by lipid overload.

Bisphenol a induces steatosis in HepaRG cells using a model of perinatal exposure.

Human exposure to bisphenol A (BPA) could favor obesity and related metabolic disorders such as hepatic steatosis. Investigations in rodents have shown that these deleterious effects are observed not only when BPA is administered during the adult life but also with different protocols of perinatal exposure. Whether perinatal BPA exposure could pose a risk in human is currently unknown, and thus appropriate in vitro models could be important to tackle this major issue. Accordingly, we determined whether long-term BPA treatment could induce steatosis in human HepaRG cells by using a protocol mimicking perinatal exposure. To this end, the kinetics of expression of seven proteins differentially expressed during liver development was determined during a 4-week period of cell culture required for proliferation and differentiation. By analogy with data reported in rodents and humans, our results indicated that the period of cell culture around day 15 and day 18 after seeding could be considered as the "natal" period. Consequently, HepaRG cells were treated for 3 weeks with BPA (from 0.2 to 2000 nM), with a treatment starting during the proliferating period. BPA was able to induce steatosis with a nonmonotonic dose response profile, with significant effects on neutral lipids and triglycerides observed for the 2 nM concentration. However, the expression of many enzymes involved in lipid and carbohydrate homeostasis was unchanged in exposed HepaRG cells. The expression of other potential BPA targets and enzymes involved in BPA biotransformation was also determined, giving answers as well as new questions regarding the mechanisms of action of BPA. Hence, HepaRG cells provide a valuable model that can prove useful for the toxicological assessment of endocrine disruptors on hepatic metabolisms, in particular in the developing liver. © 2016 Wiley Periodicals, Inc. Environ Toxicol 32: 1024-1036, 2017.

A cellular model to study drug-induced liver injury in nonalcoholic fatty liver disease: Application to acetaminophen.

Obesity and nonalcoholic fatty liver disease (NAFLD) can increase susceptibility to hepatotoxicity induced by some xenobiotics including drugs, but the involved mechanisms are poorly understood. For acetaminophen (APAP), a role of hepatic cytochrome P450 2E1 (CYP2E1) is suspected since the activity of this enzyme is consistently enhanced during NAFLD. The first aim of our study was to set up a cellular model of NAFLD characterized not only by triglyceride accumulation but also by higher CYP2E1 activity. To this end, human HepaRG cells were incubated for one week with stearic acid or oleic acid, in the presence of different concentrations of insulin. Although cellular triglycerides and the expression of lipid-responsive genes were similar with both fatty acids, CYP2E1 activity was significantly increased only by stearic acid. CYP2E1 activity was reduced by insulin and this effect was reproduced in cultured primary human hepatocytes. Next, APAP cytotoxicity was assessed in HepaRG cells with or without lipid accretion and CYP2E1 induction. Experiments with a large range of APAP concentrations showed that the loss of ATP and glutathione was almost always greater in the presence of stearic acid. In cells pretreated with the CYP2E1 inhibitor chlormethiazole, recovery of ATP was significantly higher in the presence of stearate with low (2.5mM) or high (20mM) concentrations of APAP. Levels of APAP-glucuronide were significantly enhanced by insulin. Hence, HepaRG cells can be used as a valuable model of NAFLD to unveil important metabolic and hormonal factors which can increase susceptibility to drug-induced hepatotoxicity.

Chronic exposure to low doses of pharmaceuticals disturbs the hepatic expression of circadian genes in lean and obese mice.

Drinking water can be contaminated with pharmaceuticals. However, it is uncertain whether this contamination can be harmful for the liver, especially during obesity. Hence, the goal of our study was to determine whether chronic exposure to low doses of pharmaceuticals could have deleterious effects on livers of lean and obese mice. To this end, lean and ob/ob male mice were treated for 4 months with a mixture of 11 drugs provided in drinking water at concentrations ranging from 10 to 106 ng/l. At the end of the treatment, some liver and plasma abnormalities were observed in ob/ob mice treated with the cocktail containing 106 ng/l of each drug. For this dosage, a gene expression analysis by microarray showed altered expression of circadian genes (e.g. Bmal1, Dbp, Cry1) in lean and obese mice. RT-qPCR analyses carried out in all groups of animals confirmed that expression of 8 different circadian genes was modified in a dose-dependent manner. For some genes, a significant modification was observed for dosages as low as 10²-10³ ng/l. Drug mixture and obesity presented an additive effect on circadian gene expression. These data were validated in an independent study performed in female mice. Thus, our study showed that chronic exposure to trace pharmaceuticals disturbed hepatic expression of circadian genes, particularly in obese mice. Because some of the 11 drugs can be found in drinking water at such concentrations (e.g. acetaminophen, carbamazepine, ibuprofen) our data could be relevant in environmental toxicology, especially for obese individuals exposed to these contaminants.

Relationships between radial glial progenitors and 5-HT neurons in the paraventricular organ of adult zebrafish - potential effects of serotonin on adult neurogenesis.

In non-mammalian vertebrates, serotonin (5-HT)-producing neurons exist in the paraventricular organ (PVO), a diencephalic structure containing cerebrospinal fluid (CSF)-contacting neurons exhibiting 5-HT or dopamine (DA) immunoreactivity. Because the brain of the adult teleost is known for its neurogenic activity supported, for a large part, by radial glial progenitors, this study addresses the origin of newborn 5-HT neurons in the hypothalamus of adult zebrafish. In this species, the PVO exhibits numerous radial glial cells (RGCs) whose somata are located at a certain distance from the ventricle. To study relationships between RGCs and 5-HT CSF-contacting neurons, we performed 5-HT immunohistochemistry in transgenic tg(cyp19a1b-GFP) zebrafish in which RGCs are labelled with GFP under the control of the cyp19a1b promoter. We show that the somata of the 5-HT neurons are located closer to the ventricle than those of RGCs. RGCs extend towards the ventricle cytoplasmic processes that form a continuous barrier along the ventricular surface. In turn, 5-HT neurons contact the CSF via processes that cross this barrier through small pores. Further experiments using proliferating cell nuclear antigen or 5-bromo-2'-deoxyuridine indicate that RGCs proliferate and give birth to 5-HT neurons migrating centripetally instead of centrifugally as in other brain regions. Furthermore, treatment of adult zebrafish with tryptophan hydroxylase inhibitor causes a significant decrease in the number of proliferating cells in the PVO, but not in the mediobasal hypothalamus. These data point to the PVO as an intriguing region in which 5-HT appears to promote genesis of 5-HT neurons that accumulate along the brain ventricles and contact the CSF.

Cellular stress in the pathogenesis of nonalcoholic steatohepatitis and liver fibrosis.

The burden of chronic liver disease is rising substantially worldwide. Fibrosis, characterized by excessive deposition of extracellular matrix proteins, is the common pathway leading to cirrhosis, and limited treatment options are available. There is increasing evidence suggesting the role of cellular stress responses contributing to fibrogenesis. This Review provides an overview of studies that analyse the role of cellular stress in different cell types involved in fibrogenesis, including hepatocytes, hepatic stellate cells, liver sinusoidal endothelial cells and macrophages.

Modeling and therapeutic targeting of inflammation-induced hepatic insulin resistance using human iPSC-derived hepatocytes and macrophages.

Hepatic insulin resistance is recognized as a driver of type 2 diabetes and fatty liver disease but specific therapies are lacking. Here we explore the potential of human induced pluripotent stem cells (iPSCs) for modeling hepatic insulin resistance in vitro, with a focus on resolving the controversy about the impact of inflammation in the absence of steatosis. For this, we establish the complex insulin signaling cascade and the multiple inter-dependent functions constituting hepatic glucose metabolism in iPSC-derived hepatocytes (iPSC-Heps). Co-culture of these insulin-sensitive iPSC-Heps with isogenic iPSC-derived pro-inflammatory macrophages induces glucose output by preventing insulin from inhibiting gluconeogenesis and glycogenolysis and activating glycolysis. Screening identifies TNFα and IL1ß as the mediators of insulin resistance in iPSC-Heps. Neutralizing these cytokines together restores insulin sensitivity in iPSC-Heps more effectively than individual inhibition, reflecting specific effects on insulin signaling and glucose metabolism mediated by NF-κB or JNK. These results show that inflammation is sufficient to induce hepatic insulin resistance and establish a human iPSC-based in vitro model to mechanistically dissect and therapeutically target this metabolic disease driver.

The Role of STING in Liver Injury Is Both Stimulus- and Time-Dependent.

STING, Tmem173, is involved in liver injury caused by both infectious and sterile inflammatory models. Its role in toxic liver injury and non-alcoholic fatty liver disease (NAFLD), however, is less clear. While a few groups have investigated its role in NAFLD pathogenesis, results have been conflicting. The objective of this study was to clarify the exact role of STING in toxic liver injury and NAFLD models. Goldenticket mice (Tmem173gt), which lack STING protein, were subjected to either a toxic liver injury with tunicamycin (TM) or one of two dietary models of non-alcoholic fatty liver disease: high fructose feeding or Fructose-Palmitate-Cholesterol (FPC) feeding. Three days after TM injection, Tmem173gt mice demonstrated less liver injury (average ALT of 54 ± 5 IU/L) than control mice (average ALT 108 ± 24 IU/L). In contrast, no significant differences in liver injury were seen between WT and Tmem173gt mice fed either high fructose or FPC. Tmem173gt mice only distinguished themselves from WT mice in their increased insulin resistance. In conclusion, while STING appears to play a role in toxic liver injury mediated by TM, it plays little to no role in two dietary models of NAFLD. The exact role of STING appears to be stimulus-dependent.

Drug-induced liver injury in obesity and nonalcoholic fatty liver disease.

Obesity is commonly associated with nonalcoholic fatty liver (NAFL), a benign condition characterized by hepatic lipid accumulation. However, NAFL can progress in some patients to nonalcoholic steatohepatitis (NASH) and then to severe liver lesions including extensive fibrosis, cirrhosis and hepatocellular carcinoma. The entire spectrum of these hepatic lesions is referred to as nonalcoholic fatty liver disease (NAFLD). The transition of simple fatty liver to NASH seems to be favored by several genetic and environmental factors. Different experimental and clinical investigations showed or suggested that obesity and NAFLD are able to increase the risk of hepatotoxicity of different drugs. Some of these drugs may cause more severe and/or more frequent acute liver injury in obese individuals whereas others may trigger the transition of simple fatty liver to NASH or may worsen hepatic lipid accumulation, necroinflammation and fibrosis. This review presents the available information regarding drugs that may cause a specific risk in the context of obesity and NAFLD. These drugs, which belong to different pharmacological classes, include acetaminophen, halothane, methotrexate, rosiglitazone and tamoxifen. For some of these drugs, experimental investigations confirmed the clinical observations and unveiled different pathophysiological mechanisms which may explain why these pharmaceuticals are particularly hepatotoxic in obesity and NAFLD. Because obese people often take several drugs for the treatment of different obesity-related diseases, there is an urgent need to identify the main pharmaceuticals that may cause acute liver injury on a fatty liver background or that may enhance the risk of severe chronic liver disease.

Possible Involvement of Mitochondrial Dysfunction and Oxidative Stress in a Cellular Model of NAFLD Progression Induced by Benzo[a]pyrene/Ethanol CoExposure.

Exposure to xenobiotics could favor the transition of nonalcoholic fatty liver (NAFL) to nonalcoholic steatohepatitis in obese patients. Recently, we showed in different models of NAFL that benzo[a]pyrene (B[a]P) and ethanol coexposure induced a steatohepatitis-like state. One model was HepaRG cells incubated with stearate and oleate for 2 weeks. In the present study, we wished to determine in this model whether mitochondrial dysfunction and reactive oxygen species (ROS) overproduction could be involved in the occurrence of this steatohepatitis-like state. CRISPR/Cas9-modified cells were also used to specify the role of aryl hydrocarbon receptor (AhR), which is potently activated by B[a]P. Thus, nonsteatotic and steatotic HepaRG cells were treated with B[a]P, ethanol, or both molecules for 2 weeks. B[a]P/ethanol coexposure reduced mitochondrial respiratory chain activity, mitochondrial respiration, and mitochondrial DNA levels and induced ROS overproduction in steatotic HepaRG cells. These deleterious effects were less marked or absent in steatotic cells treated with B[a]P alone or ethanol alone and in nonsteatotic cells treated with B[a]P/ethanol. Our study also disclosed that B[a]P/ethanol-induced impairment of mitochondrial respiration was dependent on AhR activation. Hence, mitochondrial dysfunction and ROS generation could explain the occurrence of a steatohepatitis-like state in steatotic HepaRG cells exposed to B[a]P and ethanol.

Co-exposure to benzo[a]pyrene and ethanol induces a pathological progression of liver steatosis in vitro and in vivo.

Hepatic steatosis (i.e. lipid accumulation) and steatohepatitis have been related to diverse etiologic factors, including alcohol, obesity, environmental pollutants. However, no study has so far analyzed how these different factors might interplay regarding the progression of liver diseases. The impact of the co-exposure to the environmental carcinogen benzo[a]pyrene (B[a]P) and the lifestyle-related hepatotoxicant ethanol, was thus tested on in vitro models of steatosis (human HepaRG cell line; hybrid human/rat WIF-B9 cell line), and on an in vivo model (obese zebrafish larvae). Steatosis was induced prior to chronic treatments (14, 5 or 7 days for HepaRG, WIF-B9 or zebrafish, respectively). Toxicity and inflammation were analyzed in all models; the impact of steatosis and ethanol towards B[a]P metabolism was studied in HepaRG cells. Cytotoxicity and expression of inflammation markers upon co-exposure were increased in all steatotic models, compared to non steatotic counterparts. A change of B[a]P metabolism with a decrease in detoxification was detected in HepaRG cells under these conditions. A prior steatosis therefore enhanced the toxicity of B[a]P/ethanol co-exposure in vitro and in vivo; such a co-exposure might favor the appearance of a steatohepatitis-like state, with the development of inflammation. These deleterious effects could be partly explained by B[a]P metabolism alterations.