MicroRNA-125b-5p mimic inhibits acute liver failure

The lack of broad-spectrum anti-acute liver failure (ALF) therapeutic agents contributes to ALF-related mortality. MicroRNAs (miRNAs) are suggested to be potent serum biomarkers for ALF, but their functional and therapeutic relevance in ALF are unclear. Here we show an unbiased approach, using two complementary miRNA screens, to identify miRNAs that can attenuate ALF. We identify miR-125b-5p as a regulator of cell death that attenuates paracetamol-induced and FAS-induced toxicity in mouse and human hepatocytes. Importantly, administration of miR-125b-5p mimic in mouse liver prevents injury and improves survival in models of ALF. Functional studies show that miR-125b-5p ameliorates ALF by directly regulating kelch-like ECH-associated protein 1, in turn elevating expression of nuclear factor-E2-related factor 2, a known regulator in ALF. Collectively, our findings establish miR-125b-5p as an important regulator of paracetamol-induced and FAS-induced cell death. Thus, miR-125b-5p mimic may serve as a broad-spectrum therapeutic attenuator of cell death during ALF.

A cute liver failure (ALF) is a rare multi-organ-failure disease that is usually caused by viral hepatitis or drug toxicity 1,2 . ALF patients often undergo orthotopic liver transplantation or otherwise die due to shortage of donor livers 3 . One of the major obstacles in treatment of ALF is the lack of both suitable mechanistic biomarkers and broad-spectrum anti-ALF agents 4 . During the last decade, microRNAs (miRNAs) have been suggested as potential biomarkers for various life-threatening diseases, including ALF [5][6][7][8][9][10] . For example, miR-122, the most abundant miRNA in the liver, has been identified as a biomarker of ALF in mice and humans 5,7,11 . Similarly, miR-125b-5p has been reported as one of several miRNAs that have elevated expression in cases of paracetamol overdose 10 . Circulating miRNAs in plasma may even be better biomarkers than classical parameters, such as alanine transaminase (ALT) and aspartate transaminase (AST), as miRNAs precede the injury, whereas ALT and AST follow the injury 9 . In contrast to their use as biomarkers, miRNAs are only now beginning to be explored as therapeutic agents 12,13 . An important determinant for the use of miRNAs as ALF biomarkers in the clinic is a clear, mechanistic understanding of their functional roles in liver tissue. Furthermore, identification of miRNAs as broad-spectrum anti-ALF agents, which are capable of attenuating cell death, would have the potential to reduce the mortality rate.

Results
Functional miRNA screens identify miR-125b-5p as a regulator of ALF. We undertook an unbiased screening approach to identify miRNAs that are capable of inhibiting cell death during ALF. We screened for 302 miRNAs from a mimic library, representing the maximum number of completely conserved miRNAs between human and mouse from miRbase version 20.0. First, we functionally screened miRNAs in primary mouse hepatocytes mimicking ALF, induced by paracetamol, also known as acetaminophen or N-acetyl-p-aminophenol (APAP), the most commonly reported liver toxicity due to drug overdose. Primary hepatocytes were transfected with 25 nM miRNA mimic before onset of APAP-induced toxicity (Fig. 1a). We evaluated all 302 conserved miRNAs and examined cell viability in response to APAP-induced toxicity (Fig. 1b). The hepatoprotective miRNAs were selected based on the following two criteria: (a) an miRNA should provide protection by more than 20% and (b) the miRNA must have high or at least a modest expression in normal human liver 10 . We found seven miRNAs fulfilling these two criteria: miR-194-5p, miR-125b-5p, miR-21-5p, let-7a-5p, miR-122-5p, miR-30c-5p and miR-193a-3p (Fig. 1c). Next, we validated all seven miRNAs in three independent experiments, which confirmed protection against APAP by all seven miRNA mimics, except miR-193a-3p ( Supplementary Fig. 1A). We then examined whether inhibition of those seven miRNAs reverses the protective effect against APAP. Inhibition of three miRNAs, let-7a-5p, miR-125b-5p and miR-122-5p, showed lower cell viability compared with control ( Supplementary Fig. 1B). Thus, gainand loss-of-function of let-7a-5p, miR-125b-5p and miR-122-5p inversely influences APAP-induced hepatocyte cell death.
Glutathione (GSH) depletion is one of the hallmarks of APAP-induced hepatotoxicity, which can be detected by measuring the ratio of hepatocyte GSH and glutathione disulfide (GSSG). Hence, to test whether the short-listed seven miRNAs inhibit APAP-induced toxicity specifically, we measured GSH/GSSG ratio in primary hepatocytes transfected with either miRNA mimics or inhibitors. Although we found that mimics of miR-125b-5p, miR-194-5p miR-21-5p and miR-122-5p restored GSH levels significantly (Fig. 1d), inhibitors of only miR-125b-5p and miR-122-5p led to significant reduction in GSH levels ( Fig. 1e). Together, miRNA screening and subsequent validation revealed that miR-125b-5p and miR-122-5p protect against APAP-induced hepatocyte toxicity.
Apoptosis is a substantial contributor to ALF, especially in ALF associated with viral hepatitis and acute Wilson's disease 14,15 . Therefore, we complemented our APAP-induced ALF miRNA screening with an apoptosis-driven ALF miRNA screening, using a model of FAS/CD95 receptor-induced apoptosis. To induce apoptosis, hepatocytes transfected with miRNA mimics were cultured in the presence of FAS-agonist antibody (anti-CD95 and clone CD95) (Fig. 1f). FAS antibody causes massive apoptosis in vitro and in vivo, leading to ALF. Likewise, we screened the library of 302 conserved miRNAs in primary mouse hepatocytes, to identify miRNAs that inhibit apoptosis-induced ALF (Fig. 1g). Our screen identified five miRNAs, miR-130a-3p, miR-125b-5p, miR-29c-3p, miR-16-5p and miR-23b-3p, whose mimics suppressed FAS-induced apoptosis in primary hepatocytes (Fig. 1h). The criteria for selection of these five miRNAs were the same as mentioned for the miRNA screening in APAPinduced ALF model. We then validated these five miRNAs in three independent experiments using miRNA mimics and inhibitors. Our cell viability assay showed hepatoprotection by all miRNAs, except miR-16-5p (Supplementary Fig. 2A,B). However, terminal deoxynucleotidyl transferase-mediated dUTP nick-end labelling (TUNEL) assay confirmed hepatoprotection by miR-125b-5p and miR-29c-3p, indicating that only these two miRNAs exhibit the ability to suppress FAS-induced apoptosis (Fig. 1i,j). Based on our complementary miRNA screenings in two different models of ALF, we identified miR-125b-5p as a common miRNA that protected primary hepatocytes in vitro against both APAP-induced ALF and FAS-induced ALF. We then tested the dose-dependent effect of miR-125b-5p and found that mimic at the concentration of 25 nM or higher provides significant protection against APAP-and FAS-induced ALF in vitro ( Supplementary Fig. 3). In addition, we analysed whether APAP influences miR-125b-5p expression and found that miR-125b-5p expression reduces in a dose-and time-dependent manner on ALF induction in vitro ( Supplementary Fig. 4). Thus, on the basis of our screens and in vitro experiments, we selected miR-125b-5p for further studies.
Administration of miR-125b-5p ameliorates ALF in vivo. To examine the therapeutic relevance of miR-125b-5p in ALF, we investigated whether miR-125b-5p protects hepatocytes against APAP-induced ALF in male BALB/c mice (Fig. 2a). At first, we determined an optimal dose by injecting BALB/c mice with various doses of APAP ranging from 150 to 800 mg kg À 1 . We found that administration of 350 mg kg À 1 APAP or a higher dose leads to 100% lethality in mice. Therefore, we used 350 mg kg À 1 APAP as a lethal dose in BALB/c mice for further experiments ( Supplementary Fig. 5). To overexpress miR-125b-5p in the mouse liver, we cloned pri-miR-125b-5p under the transcriptional control of the hepatocyte-specific promoter transthyretin (Ttr) in an adeno-associated virus (AAV) plasmid and subsequently prepared high-titre AAV serotype 8 encoding miR-125b-5p (henceforth referred to as AAV-Ttr-miR-125b-5p). The successful overexpression of miR-125b-5p was confirmed in BALB/c mice administered with 1 Â 10 10 AAV-Ttr-miR-125b-5p virions via the tail vein (Fig. 2b). We then tested efficacy of miR-125b-5p in an in vivo ALF mouse model using BALB/c mice injected with 350 mg kg À 1 , a lethal dose of APAP, intraperitoneally. We observed significantly higher survival in mice injected with AAV-Ttr-miR-125b-5p than in mice injected with a control AAV (Fig. 2c). Surviving mice in AAV-Ttr-miR-125b-5p group were monitored and kept alive for 6 months after the induction of APAP toxicity. Hence, the survival study indicates that miR-125b-5p overexpression in mouse liver renders resistance against APAP-induced ALF and thus improves survival.
Next, we examined whether higher survival observed in mice is indeed due to specific inhibition of APAP toxicity. To address this, a different set of miR-125b-5p-overexpressing mice and control mice was killed 6 h after APAP injection. First, we detected lower levels of alanine transaminase (ALT) and aspartate transaminase (AST) (Fig. 2d), and reduced hepatic injury (Fig. 2e) in miR-125b-5p-overexpressing mice compared with their respective controls. Importantly, decreased serum glutamate dehydrogenase (GDH) levels, elevated GSH/GSSG levels and decreased serum mitochondrial DNA (mtDNA) in miR-125b-5poverexpressing mice confirmed the specific inhibition of APAPinduced ALF (Fig. 2f-h). It is noteworthy to mention that serum GDH and mtDNA not only confirm ALT and AST values but also represent mechanistic biomarkers for mitochondrial damage as reported previously in mice and humans during APAP-induced liver injury 16 . Thus, miR-125b-5p overexpression improves survival of mice in APAP-induced ALF model. We then evaluated the protective effect of miR-125b-5p in FAS-induced ALF (Fig. 2i). First, we overexpressed miR-125b-5p in mouse liver by administering AAV-Ttr-miR-125b-5p in BALB/c mice (Fig. 2j) and subsequently injected them with a lethal dose of FAS antibody. Similar to APAP-induced ALF, we observed significantly higher survival of mice injected with AAV-Ttr-miR-125b-5p than their respective controls (Fig. 2k). Furthermore, serum ALT and AST levels, haematoxylin and eosin staining, caspase-3/7 activity assay, TUNEL assay ( Fig. 2l-p), cleaved caspase-3 staining and caspase-7 staining ( Supplementary  Fig. 6A,B) provided evidence that miR-125b-5p overexpression in mice inhibited ALF and hence improved survival.
Notably, a one-time injection of 1 Â 10 10 AAV virions did not exert any effect on the survival of mice. However, when we administered 2 Â 10 10 AAV virions once, we observed improved survival and inhibition of APAP-or FAS-induced ALF ( Supplementary Fig. 7). We then investigated whether overexpression of miR-125b-5p influences proliferation in APAP-induced ALF. Our Ki67 staining and quantification revealed the absence of contribution of increased proliferation  in APAP-induced ALF ( Supplementary Fig. 8A,B). Thus, miR-125b-5p inhibits ALF by suppressing cell death rather than directly effecting proliferation. AAV-based overexpression in ALF would remain a prophylactic approach; therefore, to test whether miR-125b-5p delivery may serve as a treatment option, we examined the effect of miR-125b-5p in ALF models when injury has already begun. To examine this, we first injected BALB/c mice with 350 mg kg À 1 APAP. One hour after APAP injection, these mice were administered with stabilized 10 mg miRIDIAN miR-125b-5p mimic (Fig. 3a). We first confirmed the overexpression of miR-125b-5p in mouse liver after administration of miR-125b-5p mimic (Fig. 3b). Our Kaplan-Meier survival curve analyses revealed that miR-125b-5p administration after APAP injection significantly improved the survival of mice (Fig. 3c). Analyses of indicators of APAP-induced ALF revealed reduced ALT and AST levels, reduced hepatic injury, lower serum GDH levels, higher GSH/GSSG ratio and reduced serum mtDNA, suggesting that improved survival was indeed due to inhibition of APAP-induced ALF ( Fig. 3d-h). Likewise, we investigated the effect of administration of stabilized 10 mg miRIDIAN miR-125b-5p mimic 1 h after onset of FAS-induced ALF (Fig. 3i). We again confirmed the overexpression of miR-125b-5p expression (Fig. 3j). Importantly, we observed significantly improved survival (Fig. 3k) due to suppressed apoptosis as shown by lower ALT and AST levels, reduced injury, lower caspase-3/7 activity and decreased TUNEL staining in mice injected with stabilized 10 mg miRIDIAN miR-125b-5p mimic ( Fig. 3l-p). Thus, these experiments suggest that delivery of miR-125b-5p after onset of ALF is capable of suppressing ALF, and  hence leads to improved survival. Therefore, miR-125b-5p mimic delivery may have a therapeutic relevance for treating ALF.
MiR-125b-5p regulates Keap1 at the post-transcriptional level. The major mechanism by which miRNAs regulate cellular processes is via posttranscriptional regulation by binding to the 3 0untranslated region (3 0 -UTR) of target messenger RNAs.
Therefore, to find a mechanism for the protective effect of miR-125b-5p against ALF, we performed in-silico analyses on key regulators of ALF, to find a novel target of miR-125b-5p. We found that miR-125b-5p is predicted to target 3 0 -UTR of kelch-like ECH-associated protein1 (Keap1) (Fig. 4a). Hepatocyte-specific deletion of KEAP1 has been shown to attenuate ALF, indicating that KEAP1 is an important regulator of ALF 17 . Indeed KEAP1 protein levels decreased in miR-125b-5p-overexpressing mice, indicating that miR-125b-5p regulates KEAP1 expression ( Fig. 4b and Supplementary  Fig. 10a). In contrast, we did not find significant difference in mRNA levels of miR-125b-5p-overexpressing mice and control mice (Fig. 4c). We then investigated whether miR-125b-5p regulates KEAP1 by binding to the 3 0 -UTR of Keap1. To validate predicted binding sites of miR-125b-5p in the 3 0 -UTR of Keap1 mRNA, we cloned the 3 0 -UTR of Keap1 in a luciferase reporter vector and co-transfected miR-125b-5p and reporter vector in primary mouse hepatocytes. Luciferase reporter assay demonstrated that miR-125b-5p binds to the 3 0 -UTR and thus regulates KEAP1 directly (Fig. 4d). To further confirm the binding of miR-125b-5p with 3 0 -UTR of Keap1 mRNA, we performed the luciferase reporter assay using the mutated 3 0 -UTR of Keap1 mRNA (Fig. 4e). Unchanged levels of luciferase activity in the presence of mutated 3 0 -UTR confirmed that miR-125b-5p regulates the Keap1 expression at posttranscriptional level (Fig. 4f). Thus, we identified Keap1 as a novel target of miR-125b-5p. Next, we examined whether direct regulation of KEAP1 by miR-125b-5p affects subsequent signalling involved in progression of ALF. KEAP1 facilitates degradation of nuclear factor (erythroid-derived 2)-like2 (NRF2) by acting as an adapter for cullin3, a subunit of E3 ubiquitin ligase 18 . KEAP1-NRF2 signalling has been reported to govern key regulatory functions during ALF 17,19 . We therefore determined the expression of NRF2 by western blotting. In fact, we found elevated levels of NRF2 in miR-125b-5p-overexpressing mice ( Fig. 4g and Supplementary Fig. 10b). Furthermore, the expression of NRF2 target genes such as Ugt1a6, Gclc, Nqo1 and Gsta2 increased in mice injected with AAV-Ttr-miR-125b-5p (Fig. 4h). Thus, miR-125b-5p directly regulates Keap1, which leads to enhanced NRF2 signalling and hence inhibits APAP-induced ALF.
A single miRNA often regulates multiple targets simultaneously. Therefore, we asked the question up to what extent does miR-125b-5p exert an anti-ALF effect via KEAP1. To answer this question, we co-transfected primary mouse hepatocytes with miR-125b-5p inhibitor and Keap1 small interfering RNA (siRNA). We first confirmed the efficacy of Keap1 siRNA by determining KEAP1 protein levels after transfection of hepatocytes with Keap1 siRNA alone or in combination with miR-125b-5p inhibitors (Fig. 5a,b and Supplementary Fig. 10c). Importantly, we observed significant inhibition of APAP-induced injury as shown by elevated GSH/GSSG ratio, increased cell viability and reduced TUNEL staining in hepatocytes transfected with siRNA alone compared with respective controls (Fig. 5c-f). Notably, hepatocytes co-transfected with siRNA and miR-125b-5p inhibitors had significantly increased APAPinduced injury compared with hepatocytes that were transfected with Keap1 siRNA alone (Fig. 5c-f). We further investigated whether suppression of APAP-induced injury via miR-125b-5p affects NRF2 levels and NRF2-responsive genes. Indeed, western blot and quantitative reverse transcriptase-PCR results confirmed that NRF2 protein levels and mRNA levels of Ugt1a6, Gclc, Gsta2 and Nqo1, respectively, were elevated on transfection of Keap1 siRNA alone or together with miR-125b-5p inhibitor (Fig. 5g-i and Supplementary Fig. 10d). Likewise, we examined the contribution of KEAP1 in suppression of FAS-induced apoptosis by miR-125b-5p. To this end, we transfected hepatocytes with Keap1 siRNA alone or in the presence of the miR-125b-5p inhibitor before the induction of apoptosis. Similar to APAP-induced toxicity, we observed inhibition of FAS-induced apoptosis as shown by reduced caspase-3/7 activity, increased cell viability and reduced TUNEL staining in hepatocytes transfected with Keap1 siRNA alone (Fig. 5j-m). Importantly, hepatocytes co-transfected with the Keap1 siRNA and miR-125b-5p inhibitor had significantly higher FAS-induced injury compared with hepatocytes transfected with Keap1 siRNA alone (Fig. 5j-m). It is important to mention that the inhibition of toxicity by Keap1 siRNA remained less pronounced compared with the miR-125b-5p mimic (Fig. 1). Importantly, hepatocytes co-transfected with miR-125b-5p inhibitor and Keap1 siRNA showed significantly lower toxicity than hepatocytes transfected with miR-125b-5p inhibitor alone ( Fig. 5c-m). Hence, KEAP1 contributes significantly to the anti-ALF effects of miR-125b-5p.
The circulating level of miR-125b-5p has been reported to be elevated in patients with APAP overdose 10 . We therefore measured the levels of miR-125b-5p in sera, isolated hepatocytes and liver of mice injected with 350 mg kg À 1 APAP. We confirmed the elevated levels of miR-125b-5p in mice sera ( Supplementary Fig. 9). In addition, we observed decreased miR-125b-5p levels in hepatocytes and in liver tissue of APAP-or FAS-injected mice (Supplementary Fig. 9). Thus, miR-125b-5p levels follow an inverse correlation between serum and hepatocytes or the liver.
MiR-125b-5p-KEAP1-NRF2 signaling in ALF patients. Finally, we addressed whether miR-125b-5p expression is deregulated during ALF in human patients. Upregulation of miR-125b-5p in sera of ALF patients has been recently reported 10 . We also confirmed elevated levels of miR-125b-5p in sera of ALF patients (Fig. 7a). Furthermore, we analysed miR-125b-5p expression in liver biopsies obtained from ALF patients. We found reduced  levels of miR-125b-5p in biopsies compared with control livers (Fig. 7b). Consistent with our mouse data, we observed increased protein levels of KEAP1 and reduced protein levels of NRF2 in ALF patients' liver biopsies (Fig. 7c,d). In addition, NRF2responsive genes such as UGT1A6 and GCLC were expressed at lower levels, thus indicating reduced NRF2 signalling in liver biopsies from ALF patients compared with respective controls (Fig. 7e). Taken together, the loss of hepatic miR-125b-5p expression during ALF, increased levels of serum miR-125b-5p in ALF patients and the alleviation of ALF on miR-125b-5p supplementation in mouse liver indicate that miR-125b-5p is a key regulator of ALF.

Discussion
The development of novel therapeutic agents to attenuate ALF is needed especially in cases of indeterminate, idiosyncratic and other ALF when N-acetylcysteine is ineffective. One of the prerequisites for such an agent is to suppress necrosis and apoptosis, two major cell death modes involved in the progression of ALF. Based on two complementary miRNA screenings in primary hepatocytes, we identified miR-125b-5p as a novel regulator of ALF and uncovered its hepatoprotective function in ALF. Our results demonstrate that miR-125b-5p overexpression attenuates APAP-induced necrosis and FAS-induced apoptosis in vitro, as well as ALF, in vivo.
Our findings that miR-125b-5p functions as an attenuator of ALF were somewhat unexpected, especially in light of previously reported tumour suppressor functions of miR-125b-5p in liver cancer 20,21 . Therefore, our results highlight the fact that an miRNA, such as miR-125b-5p in the liver, may have diverse functions depending on whether liver damage is acute or chronic.
MiR-125b-5p ameliorates APAP-induced ALF via posttranscriptional regulation of Keap1 mRNA. As a result, NRF2 protein levels are elevated and it translocates to the nucleus to regulate transcription of key genes. Indeed, our results revealed that miR-125b-5p overexpression elevates not only NRF2 levels but also the expression of NRF2-responsive genes such as Ugt1a6, Gclc, Nqo1 and Gsta2. Therefore, miR-125b-5p regulates GSH levels and hence APAP-induced ALF by modulation of KEAP1-NRF2 signalling. Likewise, apoptosis, which is another mode of cell death, was also inhibited in vitro and in vivo by antiapoptotic function of miR-125b-5p via KEAP1-NRF2 signalling.
Notably, NRF2 has also been reported to inhibit FAS-induced ALF in mice 22 . On one hand, reduced levels of miR-125b-5p causes NRF2 downregulation via Keap1 signalling but, on the other hand, it is plausible that NRF2 may also contribute to the further changes in miR-125b-5p levels via a feedback loop as reported previously 23,24 . It is important to mention that other miR-125b-5p targets in addition to KEAP1 may also contribute to the observed attenuation of ALF, especially in light of our results, demonstrating only partial rescue of the observed effect on blockade of Keap1 expression using siRNA. Thus, miR-125b-5p acts as an anti-ALF therapeutic agent as demonstrated by our study in two different mouse models of ALF.
In addition to our current findings of protective role of miR-125b-5p during ALF, miR-125b-5p has been previously reported as one of the circulating miRNAs upregulated during the early phase of ALF, at the time when classical markers such as ALT remain low 10,25 . We were able to confirm the elevated levels of miR-125b-5p as reported in APAP-overdose patients 10 . The precise mechanism of upregulation of circulating miRNAs in human serum during ALF is currently ambiguous 25 . There are many possibilities that may lead to miR-125b-5p release into serum. MiR-125b-5p may be released from hepatocytes via exosomes or microvesicles or via apoptotic bodies or by random release on cell lysis. It is reasonable to speculate that release via exosomes or microvesicles is the most probable way that can explain miR-125b-5p elevation in serum in the initial stages of ALF 10 . In light of our data demonstrating decrease in hepatic miR-125b-5p levels in biopsies of ALF patients and amelioration of ALF on its overexpression, miR-125b-5p may prove to be a potent ALF regulator and useful mechanistic serum biomarker for the diagnosis and prognosis of ALF.
In summary, based on miRNA screenings, we identified miR-125b-5p as a novel regulator of ALF. Our in vivo studies established miR-125b-5p as an anti-ALF miRNA that possesses the capability to inhibit ALF progression. Hence, miR-125b-5p supplementation may serve as therapeutic agent to attenuate ALF.

Methods
Mice. All mouse experiments were granted permission and performed according to the guidelines of the Hannover Medical School, Germany. Eight-to 10-week-old male BALB/c mice were purchased from Charles River Laboratories (Germany).
miRNA mimic library screening. Mouse primary hepatocytes were seeded in 96-well cell culture plates pre-coated with collagen and transfected with mouse miRNA mimic library (Thermo Scientific, Germany) at a final concentration of 25 nM.
In vitro and in vivo APAP-and FAS-induced hepatocyte damage. For APAP-induced hepatocyte damage, hepatocytes were treated with 3 mg ml À 1  Figure 7 | Determination of miR-125b-5p levels in ALF patients. (a) Quantitative reverse transcriptase-PCR (qRT-PCR) analyses revealed elevated miR-125b-5p levels in sera of spontaneously recovered compared with non-spontaneously recovered ALF patients. *Po0.05, two-tailed Student's t-test. (b) Decreased endogenous level of miR-125b-5p in liver biopsies obtained from ALF patients, compared with respective controls. The aetiology of ALF patients (n ¼ 8) was unknown, except for one patient who had mushroom intoxication. *Po0.05, two-tailed Student's t-test. (c) Immunofluorescence staining for KEAP1 and NRF2 in liver sections from ALF patients and respective controls. Scale bars, 100 mm. (d) ImageJ-based quantification of staining shown in c. **Po0.01, two-tailed Student's t-test. (e) qRT-PCR analysis of UGT1A6 and GCLC in ALF patients and respective controls. **Po0.01, two-tailed Student's t-test. Data are presented as mean ± s.e.m.
Data availability. The authors declare that data supporting the findings of this study are available within the article and its Supplementary Information files.