Intracellular and extracellular biomarkers of drug-induced liver injury

Drug-induced liver injury (DILI) is a common form of adverse drug reaction (ADR) seen within the clinic. Sensitive and specific circulating biomarkers would aid in the prediction of DILI early in its course. However, the current biomarkers of DILI, such as alanine transaminase (ALT) suffer from a lack of specificity and sensitivity. Because of this, we have examined both intracellular and extracellular biomarkers of DILI in order to validate and identify novel biomarkers of DILI. Protein tyrosine nitration (PTN), an intracellular marker of oxidative stress, has been shown to be present in paracetamol-DILI, and it is thought that the causative factor for its occurrence is mitochondrial damage. In order to test this, we used Furosemide (FS), a hepatotoxin not thought to cause mitochondrial injury or glutathione (GSH) depletion with the hypothesis that this compound would not generate PTN. First, we tested the role of GSH in protecting from PTN through depleting GSH in a mouse hepatoma cell line, followed by incubation with peroxynitrite. We found that GSH depletion was required in order to elicit PTN. Following this, we compared the ability of toxic doses of FS and paracetamol to cause PTN in mice. Interestingly, we found both compounds lead to PTN, suggesting that there is a pathway independent of GSH-depletion and mitochondrial injury which may lead to PTN. MicroRNA-122 (miR-122) a potential novel extracellular biomarker of DILI, has been demonstrated to be elevated in the circulation earlier in the course of injury than current DILI biomarkers. We examined the potential for miR-122 to be released actively in exosomes during paracetamol-DILI in rats and multiple forms of DILI in humans. Our findings suggested that in both human and rodent’s, miR-122 is released in a similar profile throughout the course of DILI in exosomes, and in an exosome-free (protein-rich fraction) form. We also examined whether miR-122 is selectively released in exosomes during hepatocellular, mixed and cholestatic DILI in humans which had been prescribed a number of hepatotoxic compounds. Our study suggested that, similar to our rodent model, there is no specific pattern of exosomal release of miR-122 in any of these forms of injury. We then looked at new and relatively unexplored aspects of microRNAs, in order to evaluate how they may be used to look at damage to different zones of the liver, and damage to cells other than hepatocytes. Hepatocytes are heterogeneous, with their phenotype depending on their localisation along the porto- central axis, which has resulted in certain drugs causing zone-specific hepatotoxicity. None of the current biomarkers is able to identify zone-specific injury. We examined zonal profiles of microRNA expression within the liver of rats under basal conditions and following paracetamol. Our analysis demonstrated that 45 miRNAs are significantly differentially expressed between zone I and zone III, with three species being expressed in only one zone. Of these differentially expressed miRNAs we found that 9 species were involved in regulating 109 members Wnt/β-Catenin pathway, the molecular driver of liver zonation. We also examined changes in zonal miRNA expression following a toxic dose of paracetamol in rats. Our study was able to demonstrate that paracetamol was able to cause significant changes in the profiles of 42 and 43 miRNAs in zone I and zone III respectively. Importantly, miRNAs were both up and down regulated in both zones, suggesting that not only a loss of miRNAs is occurring during liver injury in each zone. Biliary epithelial cells (BEC) can be damaged by a plethora of different compounds, leading to vanishing bile duct syndrome, or bile duct hyperplasia. Current biomarkers for the diagnosis of BEC- injury lack in specificity, and are prone to false-positives. We developed a method to isolate BEC from the mouse liver and performed a global miRNA profile comparison of hepatocytes and BEC. As in previous studies we found miR-122-5p to be the most enriched miRNA in hepatocytes and miR-1224- 5p in BEC. On comparison of the profiles we found that 83 miRNAs were detectable in BEC but not in Hepatocytes, however further work will be required to validate any of these as markers of BEC injury.