Julin Yang

α‑naphthylisothiocyanate (ANIT) is used to induce intrahepatic cholestasis and it is frequently used for investigations into the disease mechanism. The lithocholic acid (LCA) cholestatic model has also been extensively used in various studies; however, to the best of our knowledge, a comparative study determining the hepatotoxic mechanisms induced by these two models has not been previously conducted. In the present study, ICR mice were treated with ANIT or LCA to induce cholestatic liver injury. Biochemical analysis was used to determine the serum. Alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP) and total bile acid (TBA) levels, and histopathological assessment was used to examine the liver tissue. Metabolomic analysis was used for the serum biomarker identification. Reverse transcription‑quantitative PCR analysis and western blotting were used to analyze the inflammation biomarkers. The serum metabolome of the ANIT group clustered away from of the LCA group, which was demonstrated by the different modifications of the BA components. ALP level was found to be preferentially increased in the ANIT group from 24 to 48 h. Total BA levels was only increased in the ANIT group at 24 h. In contrast, AST and ALT activity levels were preferentially increased in the LCA group. The bile ducts in the hepatic tissues of the ANIT group were observed to be severely dilated, whereas the presence of edematous hepatocytes around the necrotic lesions and neutrophil infiltration were identified in the LCA group. The expression levels of cholesterol 7α‑hydroxylase and sterol 12α‑hydroxylase genes were significantly downregulated in the ANIT group compared with the LCA group, where a stronger adaptation of BA metabolism was supported by major differences in the concentration of the BA components. Despite the aforementioned etiological differences in the cholestasis induced by each treatment, the activation of the JNK/STAT3 signaling pathway was similar between the two cholestatic models. In conclusion, these data suggested that the liver injury induced by ANIT may be cholestatic, while the liver injury caused in the LCA model may be hepatocellular. Moreover, the downstream cholestatic liver injury in both models was indicated to be mediated by the JNK/STAT3 signaling pathway.

α-Naphthylisothiocyanate (ANIT) is an experimental agent used to induce intrahepatic cholestasis. The Ppara-null mouse line is widely employed to explore the physiological and pathological roles of PPARα. However, little is known about how PPARα influences the hepatotoxicity of ANIT. In the present study, wild-type and Ppara-null mice were orally treated with ANIT to induce cholestasis. The serum metabolome of wild-type mice segregated from that of the Ppara-null mice, driven by changes of bile acid (BA) metabolites. Alkaline phosphatase and total BAs were elevated preferentially in Ppara-null mice, which correlated with changes in Cyp7a1, Cyp8b1, Mrp3, Cyp3a11, Cyp2b10, Ugt1a2, and Ugt1a5 genes and showed cross-talk between basal PPARα and potentially adaptive pathways. Il6, Tnfa, and target genes in the STAT3 pathway (Socs3, Fga, Fgb, and Fgg) were up-regulated in Ppara-null mice but not in wild-type mice. The JNK pathway was activated in both mouse lines, while NF-κB and STAT3 were activated only in Ppara-null mice. These data suggest protection against cholestasis by basal PPARα involves regulation of BA metabolism and inhibition of NF-κB/STAT3 signaling. Considering studies on the protective effects of both basal and activated PPARα, caution should be exercised when one attempts to draw conclusions in which the PPARα is modified by genetic manipulation, fasting, or activation in pharmacological and toxicological studies.

Gemfibrozil is a fibrate drug used widely for dyslipidemia associated with atherosclerosis. Clinically, both gemfibrozil and its phase II metabolite gemfibrozil 1‐O‐β‐glucuronide (gem‐glu) are involved in drug–drug interaction (DDI). But the DDI risk caused by gem‐glu between human and mice has not been compared. In this study, six volunteers were recruited and took a therapeutic dose of gemfibrozil for 3 days for examination of the gemfibrozil and gem‐glu level in human. Male mice were fed a gemfibrozil diet (0.75%) for 7 days, following which a cocktail‐based inhibitory DDI experiment was performed. Plasma samples and liver tissues from mice were collected for determination of gemfibrozil, gem‐glu concentration and cytochrome p450 enzyme (P450) induction analysis. In human, the molar ratio of gem‐glu/gemfibrozil was 15% and 10% at the trough concentration and the concentration at 1.5 h after the 6th dose. In contrast, this molar ratio at steady state in mice was 91%, demonstrating a 6‐ to 9‐fold difference compared with that in human. Interestingly, a net induction of P450 activity andin vivoinductive DDI potential in mice was revealed. The P450 activity was not inhibited although the gem‐glu concentration was high. These data suggested species difference of relative gem‐glu exposure between human and mice, as well as a net inductive DDI potential of gemfibrozil in mouse model.

The antiepileptic drug carbamazepine (CBZ) is a typical inducer of cytochrome P450 (CYP) 3A and 2C in the clinic. It is considered a strong constitutive androstane receptor activator, however both CBZ and its main metabolite CBZ 10, 11-epoxide have been reported to be pregnane X receptor (PXR) activators whose maximal efficacy and potency are comparable with the human PXR ligand rifampicin. It is unknown whether or not PXR plays a substantially important role in in vivo induction of CYP by CBZ administration. In this study, wild type and Pxr-/- mice were administered with CBZ for 5 days. Serum and liver samples were collected and subjected to hepatotoxicity assessment and CYP induction analysis. CYP2b, 2c and 3a were induced similarly in terms of transcription level, enzyme activity and protein abundance in both wild type and Pxr-/- mice. Inductive profile of CYPs in mice by CBZ administration accorded with those reported in rats, but differed from clinically reported data. These data suggest that in vivo induction of CYP in mice by multiple administration of CBZ is independent of PXR. Knowledge of the featured CYP induction profile in mice helps us understand species related CYP induction profiles among rodents and humans resulting from administration of CBZ.

Fibrates are a group of peroxisome proliferator-activated receptor α agonists used in the treatment of dyslipidemia; however, they have been reported to cause species-related hepatocarcinogenesis and clinical myotoxicity. Gemfibrozil is one of the most commonly used fibrates, and it shows the highest risk for myotoxicity among the fibrates. The inhibitory drug-drug interaction mechanism associated with gemfibrozil has been explored recently, and the induction of human P450 3A4 and 2C8 has been reported. In this study, in vivo induction of rat P450 by gemfibrozil was studied in Sprague-Dawley rats. After the rats were dosed with gemfibrozil by oral gavage, microsomes were prepared. The metabolic activities of P450 3A1/2,2C6, and 2D2 were assayed using probe substrates, and the systemic concentration of gemfibrozil during its administration was determined. P450 3A1/2 and 2C6 activities were induced 32-77% in the rats by gemfibrozil when the exposure concentration was in the clinical range. These data indicate that the inducibility of homologous P450 isoforms by gemfibrozil is similar in Sprague-Dawley rats and in humans. Inductive drug-drug interactions and inhibitory actions are involved in the co-administration of gemfibrozil with other drugs, which suggests the relevance for a fibrate-toxicology investigation.