J.V. Castell

Evaluation of the cytotoxicity of ten chemicals on human cultured hepatocytes: Predictability of human toxicity and comparison with rodent cell culture systems

The cytotoxic effect of the first 10 chemicals on the MEIC list (evaluated in the Multicentre Evaluation of In Vitro Cytotoxicity organized by the Scandinavian Society of Cell Toxicology) was evaluated on human and rat cultured hepatocytes and in the non-hepatic murine 3T3 cell line. The MTT test was used as an endpoint to evaluate cytotoxicity after 24 hr of exposure to the chemicals. The predictability of human toxicity using human hepatocytes was analysed and compared with the results using rodent cell culture systems and rat and mouse LD(50) tests. Ferrous sulphate, diazepam and isopropyl alcohol produced about the same toxicity in all three cell culture models; paracetamol and acetylsalicylic acid were more toxic to human and rat hepatocytes than to mouse 3T3 cells; amitriptyline, ethylene glycol, methanol and ethanol were more toxic to human hepatocytes than to rodent cells. Digoxin was the most cytotoxic chemical to human hepatocytes (IC(50), 4.9 nm), the alcoholic compounds (isopropanol, ethylene glycol, ethanol and methanol) were the least toxic (IC(50), 125-819 mm) and paracetamol, acetylsalicylic acid, ferrous sulphate, diazepam and amitriptyline showed intermediate cytotoxicities (IC(50), 0.05-6 mm). The data suggest that for these 10 chemicals, acute toxicity in humans was more accurately predicted using human hepatocytes than using rat hepatocytes or mouse non-hepatic 3T3 cells.

Sequential Hepatogenic Transdifferentiation of Adipose Tissue-Derived Stem Cells: Relevance of Different Extracellular Signaling Molecules, Transcription Factors Involved, and Expression of New Key Marker Genes:

Adipose tissue contains a mesenchymal stem cell (MSC) population known as adipose-derived stem cells (ASCs) capable of differentiating into different cell types. Our aim was to induce hepatic transdifferentiation of ASCs by sequential exposure to several combinations of cytokines, growth factors, and hormones. The most efficient hepatogenic protocol includes fibroblastic growth factors (FGF) 2 and 4 and epidermal growth factor (EGF) (step 1), hepatocyte growth factor (HGF), FGF2, FGF4, and nicotinamide (Nic) (step 2), and oncostatin M (OSM), dexamethasone (Dex), and insulin-tranferrin-selenium (step 3). This protocol activated transcription factors [GATA6, Hex, CCAAT/enhancer binding protein α and β (CEBPα and β), peroxisome proliferator-activated receptor-γ, coactivator 1 α (PGC1α), and hepatocyte nuclear factor 4 α (HNF4α)], which promoted a characteristic hepatic phenotype, as assessed by new informative markers for the step-by-step hepatic transdifferentiation of hMSC [early markers: albumin (ALB), α-2-macroglobuline (α2M), complement protein C3 (C3), and selenoprotein P1 (SEPP1); late markers: cytochrome P450 3A4 (CYP3A4), apolipoprotein E (APOE), acyl-CoA synthetase long-chain family member 1 (ACSL1), and angiotensin II receptor, type 1 (AGTR1)]. The loss of adipose adult stem cell phenotype was detected by losing expression of Thy1 and inhibitor of DNA binding 3 (Id3). The reexpression of phosphoenolpyruvate corboxykinase (PEPCK), apolipoprotein C3 (APOCIII), aldolase B (ALDOB), and cytochrome P450 1A2 (CYP1A2) was achieved by transduction with a recombinant adenovirus for HNF4α and finally hepatic functionality was also assessed by analyzing specific biochemical markers. We conclude that ASCs could represent an alternative tool in clinical therapy for liver dysfunction and regenerative medicine.

Effect of xenobiotics on monooxygenase activities in cultured human hepatocytes

The activity of human cytochrome P450 monooxygenases, aryl hydrocarbon hydroxylase and 7-ethoxycoumarin O-deethylase can be increased by 3-methylcholanthrene, phenobarbital and ethanol in human hepatocytes maintained in primary culture. Total cytochrome P450 content increased two-fold after 48 hr of incubation with methylcholanthrene or phenobarbital and 1.5-fold after incubation with ethanol. The three chemicals elicited different effects on cytochrome P450 dependent activities. Addition of 3-methylcholanthrene caused a time- and concentration-dependent increase in both monooxygenase activities, aryl hydrocarbon hydroxylase and 7-ethoxycoumarin O-deethylase, while phenobarbital and ethanol increased 7-ethoxycoumarin O-deethylase activity but had no effect on aryl hydrocarbon hydroxylase. Dexamethasone per se had little or no effect on either monooxygenase activities, but potentiated the effect of the three chemicals on 7-ethoxycoumarin O-deethylase.

Intracellular glutathione in human hepatocytes incubated with S-adenosyl-L-methionine and GSH-depleting drugs

The present study was undertaken to investigate (a) whether S-adenosyl-L-methionine (SAMe) added to culture medium can increase intracellular glutathione (GSH) levels in human hepatocytes and (b) whether SAMe can prevent the GSH depletion found in human hepatocytes incubated with GSH-depleting drugs (paracetamol, opiates, ethanol). Incubation of hepatocytes with increasing concentrations of SAMe resulted in a dose-dependent elevation of intracellular GSH content, which reached its maximum (35% increase) at 30 microM after 20 h. SAMe, as the only sulfur source in the medium, was efficient in repleting GSH-depleted hepatocytes following treatment with diethyl maleate. Incubation of human hepatocytes with SAMe attenuated the GSH depletion of cells incubated with toxic concentrations of paracetamol (2 mM), heroin (0.5 mM) and methadone (0.2 mM). A decrease in GSH due to exposure of hepatocytes to 50 mM ethanol was prevented when SAMe was simultaneously added to ethanol, and human hepatocytes maintained their GSH levels like non ethanol-treated cells. The experimental results of our work give the first direct evidence of the ability of exogenously administered SAMe to increase intracellular GSH levels in human hepatocytes and to prevent the GSH depletion caused by paracetamol, opiates and ethanol.

Functionality of cultured human hepatocytes from elective samples, cadaveric grafts and hepatectomies.

The major possible sources of human liver for hepatocyte isolation are elective liver biopsies, cadaveric liver grafts and therapeutic liver resections. The suitability in terms of metabolic-competent hepatocyte cultures and risk/benefit of these resources has been comparatively studied. To this end, viability of isolated hepatocytes, yield of isolation procedure, hepatocyte survival during culture and CYP activities were the parameters analysed. The best results were found in hepatocytes prepared from elective biopsies, whereas a marked reduction in viability and functional competence was seen in hepatocytes from hepatectomy samples. Metabolic differences were observed in total CYP oxidative metabolism (7-ethoxycoumarin O-deethylation, total testosterone hydroxylation), as well as in CYP3A4, CYP2C9 or CYP2C19 activities (testosterone oxidations at 6beta-, 16beta- and 17-positions, respectively). Vascular control during the hepatectomy procedure influenced hepatocyte functionality: higher CYP activities were found in hepatocytes isolated from samples obtained under non-ischemic conditions or continuous vascular clamping than in those obtained under intermittent vascular clamping. In addition to cellular functionality, other criteria such as sample availability or ethical aspects should be considered. Elective biopsies have low, but not absent, surgical risk. However, the better functionality and the higher accessibility of elective liver samples in comparison to the other groups suggest this source of liver tissue as the most appropriate for cell harvesting purposes.

POTENTIATION OF COCAINE HEPATOTOXICITY BY ETHANOL IN HUMAN HEPATOCYTES

The hepatotoxic effects of cocaine on the human liver and the effect of ethanol on cocaine-induced hepatotoxicity have been examined in adult human hepatocytes cultured in chemically defined conditions. Cultures were exposed to concentrations of cocaine ranging from 10(-2) to 10(-5) M. Cytotoxicity was evaluated after 24 hr of continuous exposure to cocaine by measuring the leakage of intracellular LDH and the ability of cells to reduce MTT. According to these end-point parameters, half-maximal cytotoxic concentrations of cocaine for human hepatocytes (IC50) were 6.8 and 7.8 mM, respectively. Lower concentrations of cocaine, however, impaired basic metabolic functions of human hepatocytes. Exposure of cells to 2 mM cocaine for 24 hr resulted in a 50% decrease in hepatic glycogen, a 40% decrease in cellular glutathione content, and a 40% decrease in urea synthesis with respect to control values. For most of the metabolic parameters assayed, significant alterations were observed at 0.5 mM cocaine. Glycogen reloading of hepatocytes began to be inhibited in the presence of 0.60 mM cocaine (IC10). Ethanol greatly potentiated cocaine-induced hepatotoxicity. After a 48 hr pretreatment of human hepatocytes with 50 mM ethanol, low concentration of cocaine (0.25 mM) that had no effects on hepatocyte metabolism in the absence of ethanol caused a 20% inhibition of the urea synthesis rate, a 40% degradation of glycogen stores, and a 30% reduction in glutathione content. The results of our work show that ethanol increases the effects of cocaine on human hepatocytes by a factor of 10.

Molecular mechanism of diclofenac hepatotoxicity: Association of cell injury with oxidative metabolism and decrease in ATP levels.

A certain number of case reports of adverse hepatic reactions to diclofenac are known, suggesting that diclofenac-associated hepatitis may be more common than previously recognized. In order to discriminate among possible molecular mechanisms of toxicity, the following were investigated: (a) cytotoxicity of diclofenac on metabolizing (rat hepatocytes) and non-metabolizing hepatic cells (HepG2, FaO); (b) changes in calcium homoeostasis, glutathione (GSH), lipid peroxidation and ATP levels, and (c) diclofenac metabolism in relation to cytotoxicity. The results indicate that toxicity is associated with the oxidative metabolism of the drug, and correlated with the formation of a minor oxidation metabolite. Inhibitors of diclofenac metabolism concomitantly reduced the toxicity of the drug. Hepatocyte injury was preceded by a decrease in ATP levels. No oxidative stress (no changes in GSH, no lipid peroxidation) could be demonstrated at this early stage. Cytotoxicity was prevented when cells were incubated with fructose, suggesting that the inability of mitochondria to produce ATP is the probable cause of diclofenac hepatotoxicity.

Evaluation of the cytotoxicity of 10 chemicals in human and rat hepatocytes and in cell lines : correlation between in vitro data and human lethal concentration

The cytotoxicity of 10 chemicals from the Multicentre Evaluation of In vitro Cytotoxicity (MEIC) list (nos 21-30) was evaluated in human and rat cultured hepatocytes and in two established cell lines (HepG2 and 3T3) according to the MEIC programme organized by the Scandinavian Society of Cell Toxicology. The MTT test was used as the endpoint of cytotoxicity after 24hr of exposure to the chemicals. Theophylline, phenobarbital and paraquat were the least cytotoxic compounds in the cellular systems (IC(50) = 450-17,000 mum) except for the 3T3 cells. The seven remaining chemicals (dextropropoxyphene, propranolol, arsenic trioxide, cupric sulfate, mercuric chloride, thioridazine and thallium sulfate) showed a similar relative cytotoxic ranking in the four in vitro systems in the lower range of concentrations (IC(50) = 2-350 mum). The data suggest that these 10 chemicals have a basal cytotoxic effect common to the four in vitro systems, and probably none of these compounds could be considered either hepatotoxic or species specific. The correlation between in vitro data and human lethal blood concentrations showed that the predictability of the in vitro systems was similar to that of in vivo rodent tests (LD(50)) only when low cytotoxic concentrations (IC(10)) were used for correlation.

Inhibition of monooxygenase activities in human hepatocytes by interferons.

The effects of human recombinant interferons alpha and gamma on monooxygenase activities in cultured human hepatocytes have been investigated. Dose-response and time course studies showed that interferons reduced 7-ethoxyresorufin O-deethylase activity of human hepatocytes after a 12-hr incubation with 300 U/ml interferons alpha and gamma (52% and 38% decrease, respectively). A reduction in 7-ethoxyresorufin O-deethylase activity was also observed in HepG2 cells, although in these cells maximal inhibition was observed after 24 hr of treatment with 1000 U/ml (a 41% and 28% decrease with interferon alpha and gamma, respectively). A decrease in activity was also observed in 7-pentoxyresorufin O-depentylase, 7-ethoxycoumarin O-deethylase and testosterone 2alpha- and 6beta-hydroxylase. It is noteworthy that the marked increase in 7-ethoxyresorufin O-deethylase activity detected in human hepatocytes after incubation with 2 mum-3-methylcholanthrene (10-fold over non-treated cells) was reduced by 40% in the presence of interferons (300 U/ml), thus indicating that the inducibility of monooxygenases could be altered by interferon treatment. The inhibitory effect of interferons on 7-ethoxyresorufin O-deethylase was transient and HepG2 cells recovered their normal activity 24 hr after interferon removal from culture medium. This study provides the first direct evidence that interferons down-regulate the level of monooxygenases in human hepatic cells and prevent, in part, their induction by xenobiotics.

Acute cytotoxicity of ten chemicals in human and rat cultured hepatocytes and in cell lines: Correlation between in vitro data and human lethal concentrations

The cytotoxicity of ten chemicals from the MEIC list (nos 11-20) was evaluated in human and rat cultured hepatocytes and in two established cell lines (HepG2 and 3T3) according to the Multicentre Evaluation of In Vitro Cytotoxicity programme organized by the Scandinavian Society of Cell Toxicology. The lactate dehydrogenase intracellular activity and the MTT test were used as endpoints of cytotoxicity after 24 hr of exposure to the chemicals. Sodium chloride and lithium sulphate were the least cytotoxic compounds in all of the cellular systems (IC(50), 25-150 mm). The eight remaining chemicals (1,1,1-trichloroethane, phenol, sodium fluoride, malathion, 2,4-dichlorophenoxyacetic acid, xylene, nicotine and potassium cyanide) showed a similar cytotoxic potential in the four in vitro systems in a narrow range of concentrations (IC(50), 1-30 mm). The data suggest that these ten chemicals have a basal cytotoxic effect common to the four in vitro systems, and probably none of these compounds could be considered either hepatotoxic or to exert species-specific toxicity. The correlation between in vitro data and human lethal blood concentrations showed a relatively low predictability for the toxicity of six compounds with important lethal effects on the CNS. The predictability of the in vitro systems was similar to that of in vivo rodent tests (LD(50)) only when low cytotoxic concentrations (IC(10)) were used for the correlation.