Treatment with the antidepressant nefazodone has been associated with clinical idiosyncratic hepatotoxicty. Using membranes expressing human bile salt export pump (BSEP), human sandwich hepatocytes, and intact rats, we compared nefazodone and its marketed analogs, buspirone and trazodone. We found that nefazodone caused a strong inhibition of BSEP (IC(50) = 9 microM), inhibition of taurocholate efflux in human hepatocytes (IC(50) = 14 microM), and a transient increase in rat serum bile acids 1 h after oral drug administration. Buspirone or trazodone had no effect on biliary transport system. Nefazodone produced time- and concentration-dependent toxicity in human hepatocytes with IC(50) = 18 microM and 30 microM measured by inhibition of protein synthesis after 6 h and 24 h incubation, respectively. Toxicity was correlated with the amount of unmetabolized nefazodone. Partial recovery in toxicity by 24 h has been associated with metabolism of nefazodone to sulfate and glucuronide conjugates. The saturation of nefazodone metabolism resulted in sustained decrease in protein synthesis and cell death at 50 microM. The toxicity was not observed with buspirone or trazodone. Addition of 1-aminobenzotriazole (ABT), an inhibitor of CYP450, resulted in enhancement of nefazodone toxicity at 10 microM and was associated with accumulation of unmetabolized nefazodone. In human liver microsomes, ABT also prevented metabolism of nefazodone and formation of glutathione conjugates. We suggest that inhibition of bile acid transport by nefazodone is an indicator of potential hepatotoxicity. Our findings are consistent with the clinical experience and suggest that described methodology can be applied in the selection of nonhepatotoxic drug candidates.

The importance of uridine 5'-diphosphate-glucuronosyltranferases (UGT) 2B15 and other UGT enzymes (1A1, 1A6, and 1A9) in glucuronidating acetaminophen (APAP) is demonstrated. The kinetics and contributions of various UGTs in glucuronidating APAP are presented using clinically and toxicologically relevant concentrations of the substrate. UGT 1A9 and UGT 2B15 contribute significantly toward glucuronidating APAP when incubations were conducted in either phosphate or Tris-HCl buffers at 0.1 and 1.0 mM substrate concentrations. At 10 mM APAP, UGT 1A9 is a significant enzyme responsible for metabolizing APAP in either one of the buffers. UGT 1A1 is the next most important enzyme in glucuronidating APAP at this high substrate concentration. The contribution of UGT 1A6 at 10 mM APAP concentration became obscured by similar relative activities exhibited by UGTs 1A7, 1A8, and 2B7. These observations may reflect the differences in kinetic parameters for APAP glucuronidation by the individual UGTs. UGT 1A1 demonstrated Hill kinetics while UGT 1A9 displayed Michaelis-Menten kinetics. Substrate inhibition kinetics is observed with UGT 1A6, UGT 2B15, and human liver microsomes. The substrate inhibition is confirmed by employing stable isotope-labeled APAP as the substrate, while APAP glucuronide is used to test for inhibition of d4-APAP glucuronide. The in vitro hepatotoxicity caused by APAP in combination with phenobarbital or phenytoin is demonstrated in this study. The inhibition of APAP glucuronidation by phenobarbital leads to an increase in APAP-mediated toxicity in human hepatocytes. The toxicity to hepatocytes was further increased by coadministering APAP with phenytoin and phenobarbital. This synergistic increase in toxicity is postulated to be due to inhibition of UGTs (1A6, 1A9, and 2B15) responsible for detoxifying APAP through the glucuronidation pathway.

Here we present a preclinical model to assess drug-drug interactions due to inhibition of glucuronidation. Treatment with the antiepileptics phenobarbital (PB) or phenytoin (PH) has been associated with increased incidence of acetaminophen (APAP) hepatotoxicity in patients. In human hepatocytes, we found that the toxicity of APAP (5 mM) was increased by simultaneous treatment with phenobarbital (2 mM) or phenytoin (0.2 mM). In contrast, pretreatment with PB for 48 h prior to APAP treatment did not increase APAP toxicity unless both drugs were present simultaneously. Cells treated with APAP in combination with PB or PH experienced decreases in protein synthesis as early as 1 h, ultrastructural changes by 24 h, and release of liver enzymes by 48 h. Toxicity correlated with inhibition of APAP glucuronidation. PB or PH also inhibited APAP glucuronidation in rat and human liver microsomes and expressed human UGT1A6, 1A9, and 2B15. As with intact hepatocytes, PB and PH were neither hydroxylated nor glucuronidated, suggesting the direct inhibition of UGTs. Our findings suggest that, in multiple drug therapy, an inhibitory complex between UGT and one of the drugs can lead to decreased glucuronidation and increased systemic exposure and toxicity of a coadministered drug.

The inability to predict if a metabolically bioactivated compound will cause toxicity in later stages of drug development or post-marketing is of serious concern. One approach for improving the predictive success of compound toxicity has been to compare the gene expression profile in preclinical models dosed with novel compounds to a gene expression database generated from compounds with known toxicity. While this guilt-by-association approach can be useful, it is often difficult to elucidate gene expression changes that may be related to the generation of reactive metabolites. In an effort to address this issue, we compared the gene expression profiles obtained from animals treated with a soft-electrophile-producing hepatotoxic compound against corresponding deuterium labeled analogues resistant to metabolic processing. Our aim was to identify a subset of potential biomarker genes for hepatotoxicity caused by soft-electrophile-producing compounds. The current study utilized a known hepatotoxic compound N-methylformamide (NMF) and its two analogues labeled with deuterium at different positions to block metabolic oxidation at the formyl (d(1)) and methyl (d(3)) moieties. Groups of mice were dosed with each compound, and their livers were harvested at different time intervals. RNA was prepared and analyzed on Affymetrix GeneChip arrays. RNA transcripts showing statistically significant changes were identified, and selected changes were confirmed using TaqMan RT-PCR. Serum clinical chemistry and histopathologic evaluations were performed on selected samples as well. The data set generated from the different groups of animals enabled us to determine which gene expression changes were attributed to the bioactivating pathway. We were able to selectively modulate the metabolism of NMF by labeling various positions of the molecule with a stable isotope, allowing us to monitor gene changes specifically due to a particular metabolic pathway. Two groups of genes were identified, which were associated with the metabolism of a certain part of the NMF molecule. The metabolic pathway leading to the production of reactive methyl isocyanate resulted in distinct expression patterns that correlated with histopathologic findings. There was a clear correlation between the expression of certain genes involved in the cell cycle/apoptosis and inflammatory pathways and the presence of reactive metabolite. These genes may serve as potential genomic biomarkers of hepatotoxicity induced by soft-electrophile-producing compounds. However, the robustness of these potential genomic biomarkers will need to be validated using other hepatotoxicants (both soft- and hard-electrophile-producing agents) and compounds known to cause idiosyncratic liver toxicity before being adopted into the drug discovery screening process.

Treatment with flutamide has been associated with clinical hepatotoxicty. The toxicity, metabolism,and transport of flutamide were investigated using cultured human hepatocytes. Flutamide and its major metabolite, 2-hydroxyflutamide, caused an inhibition of taurocholate efflux in human hepatocytes with an IC50=75 microM and 110 microM, respectively. Treatment of hepatocytes with flutamide or 2-hydroxyflutamide for 24 h resulted in time- and concentration-dependent toxicity as assessed by inhibition of protein synthesis. Toxicity was greater after 1 h than after 24 h of treatment. Recovery in inhibition of protein synthesis by 24 h was attributed to the decreased presence of flutamide due to its metabolism. Flutamide was metabolized by hepatocytes to several metabolites, and formation of reactive intermediates of flutamide, as evidenced by the presence of glutathione-related adducts, was observed. Inhibition of flutamide metabolism by 1-aminobenzotriazole (ABT) resulted in enhancement of flutamide toxicity, which was associated with sustained levels of nonmetabolized drug. ABT also prevented the formation of reactive intermediates of flutamide. There was an additive toxicity when cells were treated with a combination of flutamide and 2-hydroxyflutamide. Simultaneous treatment with flutamide and acetaminophen (APAP) resulted in additive to synergistic toxic effects. Flutamide and APAP were found to have significant effects on each other's metabolism. Flutamide inhibited glucuronidation and sulfation of APAP, resulting in greater amounts of APAP available for bioactivation. APAP inhibited the hydroxylation of flutamide, and subsequent sulfation and acetylation of 4-nitro-3-(trifluoromethyl) aniline, a metabolite of flutamide. In summary, we suggest that inhibition of bile acid efflux by flutamide and its 2-hydroxy metabolite may play a role in flutamide-induced liver injury. Both flutamide and 2-hydroxyflutamide are responsible for cytotoxicity if not metabolized. The data also suggest a possible drug-drug interaction between flutamide and APAP, resulting in inhibition of flutamide metabolism and increased APAP bioactivation and toxicity.