The liver is responsible for maintaining proper lipid, glycidic, and iron metabolism homeostasis and detoxification of xenobiotics [4]. Since their discovery in 1993, several studies have described the role of miRNAs in liver physiological functions. Additionally, their role in xenobiotic detoxification still needs to be better understood. MiRNAs are single-stranded small non-coding RNAs, between 18 and 22 nucleotides, that regulate gene expression in various conditions [5,6,7]. The liver miRNome is represented by miR-122 for about 70% [8]. It is worth mentioning that the miR-122 sequence is highly conserved between species, and its expression increases over embryonic liver development and persists throughout adulthood. This miRNA in the hepatocyte interacts with different genes modifying their expression; for example, HNF6, HNFA4, and FOXA1 genes participate in a positive feedback loop with miR-122 to maintain the balance between proliferation and differentiation of the cells during embryonic development [9]. MiR-122 has been the first miRNA related to hepatic lipid metabolism so far. Inhibition of miR-122 expression leads to the downregulation of enzymes involved in lipid metabolism. The inhibition also leads to the upregulation of enzymes participating in β-oxidation. Consequently, a significant reduction in cholesterol associated with a downregulation of genes involved in its synthesis was studied in the murine liver [9]. However, the results in the human hepatocyte cell line (HepG2) are reversed; the silencing of miR-122 leads to an initial increase in HMGCR and SREBP-2, a key protein for cholesterol biosynthesis, and an increase in the AMPK, also significant in lipid metabolism. According to Gatfield et al., suppression of miR-122 hepatic showed a marked decrease in total serum triglyceride levels. This is because it activates MTTP, increasing the concentration of VLDL in the blood [10]. In vivo depletion of miR-122 causes iron deficiency due to increased mRNA expression of the peptide hormone hepcidin, which is the critical factor of duodenal iron absorption and release [11]. Additionally, the miR-33 is an essential regulator of lipid homeostasis targeting ABCA1, ABCG1, NPC1 [12,13], and key enzymes involved in the oxidation of fatty acids, such as CPT1α, CROT, SIRT6, and AMPKα [13]. Either the miR-23b is involved in lipid metabolism [14], or the miR-1/miR-206 axes can attenuate LXRα-induced lipogenesis by targeting the LXRα mRNA in the hepatocytes [15]. Besides that, it has been seen that miR-181a downregulation improves hepatic insulin sensitivity [16]. Recently, it was shown that the reactivation of liver miR-192 alleviates steatosis [17].

The use of miRNAs as biomarkers from serum or plasma was first proposed in a DILI experimental model by the pioneer study by Wang et al. They examined the role of those molecules in a well-established mouse model of acetaminophen-induced liver injury [72]. They demonstrated that the liver-enriched miR-122 and miR-192 were the top two miRNAs elevated in the blood in a dose- and time-exposure-dependent manner. The levels of these miRNAs preceded and then are parallel to serum ALT/AST levels. From this, several other independent groups provided additional data supporting the use of miRNAs as DILI biomarkers. In particular, the plasma levels of miR-122 were correlated with liver histopathology damage induced by D-galactosamine and alcohol [73]. This confirms, even in this case, that the plasma level fluctuations of miR-122 preceded largely the changes in transaminases blood level. Therefore, it was evident that this miR could be used for diagnosing and monitoring liver damage at early stages. Compared with mice, rats' susceptibility to acetaminophen-induced liver damage is lower [74]. Therefore, rat models were used to examine acute liver damage (cholestasis or hepatocellular damage) and chronic liver damage (fibrosis, steatohepatitis, and steatosis) [75]. Their results showed that miR-122 levels increased more quickly and intensely than aminotransferases, reflecting the extent of liver damage. Outstandingly, the study also demonstrated that the expression profiles of plasma miRNAs differed according to the type of liver damage, suggesting that the miRNAs could be specific and sensitive biomarkers for various types of liver damage. An increase in levels of several oncogenic miRNAs, such as the 17–92 cluster, miR-106a, and miR-34, was detected in rat livers following exposure to tamoxifen, a potent hepatocarcinogen [75]. Furthermore, Starckx et al. demonstrated that the levels of miR-122 in rat plasma were significantly increased following the administration of four well-characterized compounds associated with different types and mechanisms of liver toxicity to chemicals allyl alcohol and α-naphthyl isothiocyanate and drugs acetaminophen and phenobarbital [76]. The changes in plasma miR-122 were detected significantly earlier than other conventional biomarkers, exhibiting a wide dynamic range. Thus, miR-122 is expected to be an early biomarker of DILI. Additionally, it was reported that exposure to acetaminophen or carbon tetrachloride decreases miR-298 and miR-370, which are thought to regulate an oxidative stress-related gene [77]. Furthermore, lethal doses of acetaminophen lead to miR-1196, miR135a, miR-466f-3p, miR-466g, miR-877, and miR-574-5p upregulation, and a marked downregulation of miR-195, miR-375, miR342-3p, miR-29c, miR-652, and miR-148a [78,79,80,81]. It was reported by Su et al. that miR-192, miR-193, and miR-122 have the potential to aid as specific, sensitive, and noninvasive biomarkers for the diagnosis of herb-induced liver damage [82]. The hepatotoxic drug halothane in patients with low expression of miR-106b caused the upregulation of signal transducer and activator of transcription 3 (STAT3) involved in its severe liver damage [83].

Circulating miRNAs as biomarkers in human DILI showed that miR-192 and miR-122 were considerably higher in patients who suffered acetaminophen-induced liver damage than in those who did not [83]. In humans, Bala et al. discovered that serum plasma levels of miR-155 and miR-122 were mainly related to the exosome-rich fraction in inflammatory and alcoholic liver damage (hepatitis). In contrast, in acetaminophen-induced liver damage, these miRNAs were present primarily in the soluble protein-rich fraction [74]. Of note, circulating levels of the liver-enriched miR-122, but not miR-192 were associated with serum transaminase levels and a decrease to baseline much earlier than serum transaminase, suggesting that miR-122 has a shorter circulatory half-life [80]. Subsequently, it was also demonstrated that miR-19b and miR-29c were upregulated after low-dose acetaminophen treatment without ALT alteration, suggesting that the expression profile of circulating miRNAs may be changed at a very early stage when liver damage is undetectable using the conventional biomarkers [82,83]. Paraquat is the most common toxic herbicide and is widely used worldwide. By associating paraquat-exposed human subjects with healthy donors, Ding et al. found that the serum levels of miR-122 were enormously increased and correlated well with the status of liver function [83], similar to the mouse acetaminophen-induced liver damage model [80]. However, miR-192, also identified in the mouse study, was unpredictably decreased two-to-eight-fold in human hepatic samples. These studies provided convincing evidence that circulating miRNAs could be used as human DILI biomarkers. 

4. Conclusions

The estimate, diagnosis, and management of DILI are very complex issues. The current biomarkers or methods to assess DILI include biochemical markers with poor sensitivity, stability, or specificity. In pre-clinical settings, circulating levels of liver-specific miR-122, the more abundant miR in the liver, can consistently and effectively distinguish intrahepatic from extrahepatic damage with higher sensitivity and specificity. The convincing evidence that circulating miR-122 could be used as a human DILI biomarker comes from several studies in which it was evidenced that a substantial increase in miR-122 correlated well with the status of liver function.