Serum biomarkers of isoniazid-induced liver injury: Aminotransferases are insufficient,and OPN,L-FABP and HMGB1 can be promising novel biomarkers

Isoniazid (INH)-induced liver injury is a great challenge for tuberculosis treatment. Existing biomarkers cannot accurately determine the occurrence of this injury in the early stage. Therefore, developing early specific sensitive biomarkers of INH-induced liver injury is urgent. A rat model of liver injury was established with gastric infusion of INH or INH plus rifampicin (RFP). We examined seven potential novel serum biomarkers, namely, glutamate dehydrogenase (GLDH), liver-fatty acid-binding protein (L-FABP), high-mobility group box-1 (HMGB1), macrophage colony-stimulating factor receptor (MCSF1R), osteopontin (OPN), total cytokeratin 18 (K18), and caspase-cleaved cytokeratin-18 (ccK18), to evaluate their sensitivity and specificity on INH-induced liver injury. With the increase of drug dosage, combining with RFP and prolonging duration of administration, the liver injury was aggravated, showing as decreased weight of the rats, upgraded liver index and oxidative stress level, and histopathological changes of liver becoming marked. But the activity of serum aminotransferases decreased significantly. The area under the curve (AUC) of receiver-operating characteristic (ROC) curve of OPN, L-FABP, HMGB1, MCSF1R, and GLDH was 0.88, 0.87, 0.85, 0.71, and 0.70 (≥0.7), respectively, and 95% confidence interval of them did not include 0.5, with statistical significance, indicating their potential abilities to become biomarkers of INH-induced liver injury. In conclusion, we found traditional biomarkers ALT and AST were insufficient to discover the INH-induced liver injury accurately and OPN, L-FABP, and HMGB1 can be promising novel biomarkers.     

Plasma biomarkers of liver injury and inflammation demonstrate a lack of apoptosis during obstructive cholestasis in mice

Cholestasis is a pathological common component of numerous liver diseases that results in hepatotoxicity, inflammation, and cirrhosis when untreated. While the predominant hypothesis in cholestatic liver injury remains hepatocyte apoptosis due to direct toxicity of hydrophobic bile acid exposure, recent work suggests that the injury occurs through inflammatory necrosis. In order to resolve this controversy, we used novel plasma biomarkers to assess the mechanisms of cell death during early cholestatic liver injury. C57Bl/6 mice underwent bile duct ligation (BDL) for 6–72 h, or sham operation. Another group of mice were given d-galactosamine and endotoxin as a positive control for apoptosis and inflammatory necrosis. Plasma levels of full length cytokeratin-18 (FL-K18), microRNA-122 (miR-122) and high mobility group box-1 protein (HMGB1) increased progressively after BDL with peak levels observed after 48 h. These results indicate extensive cell necrosis after BDL, which is supported by the time course of plasma alanine aminotransferase activities and histology. In contrast, plasma caspase-3 activity, cleaved caspase-3 protein and caspase-cleaved cytokeratin-18 fragments (cK18) were not elevated at any time during BDL suggesting the absence of apoptosis. In contrast, all plasma biomarkers of necrosis and apoptosis were elevated 6 h after Gal/End treatment. In addition, acetylated HMGB1, a marker for macrophage and monocyte activation, was increased as early as 12 h but mainly at 48–72 h. However, progressive neutrophil accumulation in the area of necrosis started at 6 h after BDL. In conclusion, these data indicate that early cholestatic liver injury in mice is an inflammatory event, and occurs through necrosis with little evidence for apoptosis.