miR-192-5p has gained increasing relevance in various diseases, however, its function in acute liver injury is currently unknown. We analysed miR-192-5p serum levels and hepatic miR-192-5p expression in mice after hepatic ischaemia and reperfusion (I/R) as well as in toxic liver injury. On a functional level, miRNA levels were analysed in the different hepatic cell-compartments and in the context of tumour necrosis factor (TNF)-dependent liver cell death. We detected increased serum levels of miR-192-5p after hepatic I/R- and carbon tetrachloride (CCl4)-induced liver injury. miR-192-5p levels correlated with the degree of liver damage and the presence of hepatic cell death detected by TUNEL stainings (terminal deoxynucleotidyltransferase-mediated dUTP biotin nick-end labelling stainings). Moreover, expression of miR-192-5p was increased in a hypoxia/reoxygenation (H/R) model of in vitro hepatocyte injury, supporting that the passive release of miR-192-5p represents a surrogate for hepatocyte death in liver injury. In critically ill patients, miR-192-5p levels were elevated selectively in patients with liver injury and closely correlated with the presence of hepatic injury. In contrast with up-regulated miR-192-5p in the serum, we detected a down-regulation of miR-192-5p in both injured mouse and human livers. Deregulation of miR-192-5p in livers was dependent on stimulation with TNF. Functional experiments confirmed a protective effect of down-regulation of miR-192-5p in hepatocytes, suggesting a role of miR-192-5p in limiting liver injury. Finally, we identified Zeb2, an important regulator of cell death, as a potential target gene mediating the function of miR-192-5p. Our data suggest that miR-192-5p is involved in the regulation of liver cell death during acute liver injury and might represent a potent marker of hepatic injury.

CLINICAL PERSPECTIVES

  • miRNAs are critically involved in the regulation of various pathological processes and were recently suggested as potential therapeutic targets.

  • miR-192-5p is significantly down-regulated in the context of acute liver injury and liver failure. Down-regulation of miR-192-5p was protective against cell death, via up-regulation of Zeb2 a well-known regulator of apoptosis.

  • miR-192-5p was found to be deregulated in serum of patient with acute liver injury, highlighting a potential function of this miRNA as a biomarker in this setting.

INTRODUCTION

miRNA are small non-coding RNAs that modulate the activity of gene expression by interaction with 3′-UTR of their target mRNAs [1,2]. In the last years it was demonstrated that miRNAs play a key role in the regulation of inflammation, carcinogenesis and cell death [3,4,5]. Moreover, serum miRNAs were proposed as biomarkers in the context of different diseases such as inflammation, bacterial infection [6] and fibrosis [7]

Several miRNAs were suggested as serum-based biomarkers for acute liver injury. It was demonstrated that miR-122, an abundantly expressed miRNA in the liver, is a part of 7 miRNA-signature showing higher accuracy and sensitivity as a biomarker in acetaminophen-induced liver injury than measurement of serum aspartate transaminase/alanine transaminase (AST/ALT) [8]. Moreover, miR-122 levels were elevated in models of nanoparticle and drug-induced liver injury [8,9]. Notably these results remained unchanged when instead of mice rats were analysed [10] and even reflected inter-strain differences in disease severity between different strains of mice [11] By translating these results into the human setting, it became apparent that median levels of miR-122 were significantly up-regulated in HCV-infected patients as well as in drug-induced liver injury when compared with healthy controls [12,13] Moreover, elevated miR-122 serum levels were found in patients after liver transplantation (LTX) when an acute rejection occurred [14]. Besides miR-122, several other miRNAs, including miR-29a/b/c, miR-15a, miR-130a, miR-146a, miR-194 and miR-192 were suggested as serum-based markers for acute liver injury [8,9,1215]. MicroRNA-192 (miR-192-5p) is ubiquitously expressed in liver, colon, kidney and small intestine [16]. Up-regulation of miR-192-5p has been investigated in inflammation-related cancers such as gastric cancer, pancreatic ductal adenocarcinoma and oesophageal carcinoma. Much attention has been given to miR-192-5p as a serum-based biomarker. As an example, miR-192-5p levels were highly elevated in various liver diseases [8,9,12,14,15] as well as in patients with cancer, where elevated serum levels are associated with tumour metastasis and poor survival [17]. In the present study, we analysed the expression of miR-192-5p in ischaemia/reperfusion (I/R) and toxin-induced liver injury. We show that miR-192-5p is specifically down-regulated in hepatocytes upon liver damage. Furthermore, its expression in hepatocytes was regulated by tumour necrosis factor α (TNFα) and promoted the de-repression of antiapoptotic genes in these cells. Finally, we demonstrate that serum levels of miR-192-5p are elevated in mice and patients after I/R, indicating hepatocyte cell death.

MATERIALS AND METHODS

Animal studies

Male C57BL/6 mice (6–8 weeks of age) were obtained from The Jackson Laboratory. Hepatic warm I/R in mice (n=4–6) was performed as described previously [18]. Blood and liver samples were taken at 3 and 24 h after reperfusion for analysis. For lipopolysaccharide (LPS) treatment mice were injected intra-peritoneally with 2.5 μg/g body weight for 8 h as recently described [19]. In all analysis, only sex-matched animals were compared. Acute toxic liver damage in mice was induced by intraperitoneal injection of 0.6 ml/kg body weight of carbon tetrachloride (CCl4, Merck) in corn oil according to the previously described method [3]. Mice were killed 48 h after the injection. Animals were cared for according to European, national and institutional regulations.

Histology and stainings

As described previously [20], determination of cell death was performed on livers after I/R procedure from mice using the TUNEL kit (Promega) according to manufacturer's instructions.

Isolation of hepatic immune cells and hepatocytes

Immune cells and hepatocytes were isolated from livers of mice after I/R procedure as described recently [21].

Cell culture

Hepa1–6 cells were obtained from A.T.C.C. and cultured in DMEM supplemented with 10% FBS and penicillin/streptomycin. Cells were grown at 37°C in a humidified atmosphere with 5% CO2. Cells were stimulated with 20 ng/ml TNFα, transforming growth factor-β (TGFβ), IL-1β, interferon-γ (IFNγ), interleukin-6 (IL-6) or interleukin-17a (IL-17a). Cells were treated with 3 mM hydrogen peroxide (H2O2) (Sigma–Aldrich, Germany) for 4 h.

Cell transfection

miR-192-5p mimic, antagomir, siRNA against Zeb2 and negative control were purchased from Qiagen. Hepa1–6 cells were cultured and transfected with either 50 nM miR mimic, Zeb2 siRNA or antagomiR using Lipofectamine 2000 (Invitrogen) for 72 h according to the manufacturer's instructions.

Cell viability assay

Cell viability was measured by using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Thermofischer Scientific). MTT (final concentration 0.5 mg/ml) was added to the cells and after 2 h of incubation at 37°C, DMSO was added to dissolve the crystals. Absorbance was measured at 540 nm.

Luciferase assay

Hepa1–6 cells were seeded in a 24-well plate for 24 h and transiently transfected with the luciferase constructs containing 3′-UTR of Zeb2 (61771, Addgene) along with miR mimic using Lipofectamine 2000 (Invitrogen). After 24 h of transfection, cells were harvested and lysed with 100 μl of passive lysis buffer (Promega). Lysates were analysed using dual luciferase assay kit (Promega) according to the manufacturer's instructions. The luciferase activities were measured and ratios of firefly luciferase luminescence to Renilla luciferase luminescence were calculated. IKKβE/E plasmid was a gift from Gilles Courtois (INSERM U1038, iRTSV, CEA, Grenoble, France). Cells were transfected with nuclear factor ‘kappa-light-chain-enhancer’ of activated B-cells (NF-κB)-dependent reporter plasmid (Igκ-Luc) as described in [22].

miRNA isolation from tissue

Total RNA from tissue was isolated using Trizol reagent (Invitrogen) by Direct-zol™ RNA Miniprep (Zymo Research Corporation) according to manufacturer's protocol.

miRNA isolation from serum

Blood samples were taken by a small team of trained physicians according to the protocol and miRNA extraction from serum was performed as described recently [21,23].

Quantitative real-time PCR

Quantitative real-time PCR (qPCR) was performed as described recently [23]. In brief, total RNA from both serum and tissue was used to synthesize cDNA by miScript Reverse Transcriptase Kit (Qiagen) according to the manufacturer's protocol. All qPCR reactions were performed in Applied Biosystems 7300 Sequence Detection System (Applied Biosystems) using the miScript SYBR Green PCR Kit (Qiagen). Data were generated and analysed using the SDS 2.3 and RQ manager 1.2 software packages.

Primers

The following primers were used for qPCR

Cdc6_for TGGCATCATACAAGTTTGTGTGG 
Cdc6_rev CAGGCTGGACGTTTCTAAGTTTT 
ZEB2_for AGGCTCGGA GAC AGA TGA AGA 
ZEB2_rev GCG GAC AGA CAG ACA CTT ACC 
Dnmt3a_for CTGTCAGTCTGTCAACCTCAC 
Dnmt3a_rev GTGGAAACCACCGAGAACAC 
Cxcr5_for ATGAACTACCCACTAACCCTGG 
Cxcr5_rev TGTAGGGGAATCTCCGTGCT 
IGF1_for CAC ATC ATG TCG TCT TCA CAC C 
IGF1_rev GGA AGC AAC ACT CAT CCA CAA TG 
Cdc6_for TGGCATCATACAAGTTTGTGTGG 
Cdc6_rev CAGGCTGGACGTTTCTAAGTTTT 
ZEB2_for AGGCTCGGA GAC AGA TGA AGA 
ZEB2_rev GCG GAC AGA CAG ACA CTT ACC 
Dnmt3a_for CTGTCAGTCTGTCAACCTCAC 
Dnmt3a_rev GTGGAAACCACCGAGAACAC 
Cxcr5_for ATGAACTACCCACTAACCCTGG 
Cxcr5_rev TGTAGGGGAATCTCCGTGCT 
IGF1_for CAC ATC ATG TCG TCT TCA CAC C 
IGF1_rev GGA AGC AAC ACT CAT CCA CAA TG 

Patient cohorts analysed

We analysed liver specific expression of miR-192-5p in liver samples of patients with liver failure due to different aetiologies. This cohort has been previously described [24]. miR-192-5p serum concentrations were measured in serum of 28 acute liver failure (ALF) patients. This cohort included 8 with spontaneous recovery (SR) and 20 with nonspontaneous recovery (NSR) that underwent LTX (n=11) or died (n=9). Causes of ALF included viral hepatitis, toxic liver injury, Budd–Chiari syndrome, Wilson's disease, autoimmune hepatitis and indeterminate aetiology. This patient cohort was recently described in [25]. Finally, we analysed miR-192-5p serum levels in 223 critically ill patients and 76 respective controls. This cohort has been previously described [21], and characteristics of the population are given in Supplementary Table S1.

Statistics

Statistical analyses were performed as described previously [21]. Data from humans are given as median and range considering the skewed distribution of most parameters. Differences between two groups were assessed by Mann–Whitney U-test. Box plot graphics display a statistical summary of the median, quartiles, range and extreme values. Correlations between variables have been analysed using the Spearman correlation test, and values of P< 0.05 were considered statistically significant. All statistical analyses on human samples were performed with SPSS version 12.0 (SPSS). Differences in mice are displayed as mean +/− S.E.M. Differences between two groups were assessed by t-test. Graphical presentations in mice were performed with GraphPad Prism 5 (Graph-Pad).

RESULTS

miR-192-5p is primarily expressed in the liver and is down-regulated in liver injury

To examine the relevance of miR-192-5p for liver homoeostasis, we first analysed its expression in different organs of C57BL/6 mice. miR-192-5p was most prominently expressed in the liver, whereas only low concentrations were found in all other organs (Supplementary Figure S1A). Moreover, we compared expression levels of miR-192-5p between different hepatic cell compartments. miR-192-5p expression was restricted to hepatocytes (Supplementary Figure S1B), suggesting that–similar to miR-122 [21]–it represents a miRNA relatively specific for the liver and hepatocytes.

To explore the specific role of miR-192-5p for acute liver injury, we applied the well-established model of hepatic I/R injury and measured the intrahepatic expression of miR-192-5p after induction of ischaemic liver damage. As shown in Figure 1(A), I/R led to significant down-regulation of miR-192-5p compared with SHAM-operated mice. Of note, this down-regulation could be confirmed in a model of CCl4-induced liver injury (Figure 1B) as well as in liver samples from patients with acute hepatic injury (Figure 1C). To further identify the cell compartments involved in the regulation of miR-192-5p during I/R, we next analysed the expression of miR-192-5p in distinct hepatic parenchymal and non-parenchymal cell compartments. miR-192-5p was down-regulated in hepatocytes whereas its expression was not changed in other cell types (Figure 1D), confirming the hepatocyte specificity of the regulation of miR-192-5p during acute liver injury.

Expression of miR-192-5p correlates with the degree of liver injury and cell death in acute liver injury

Figure 1
Expression of miR-192-5p correlates with the degree of liver injury and cell death in acute liver injury

(A) miR-192-5p expression levels are down-regulated in liver tissue of I/R-operated mice compared with mice after sham surgery. (B) miR-192-5p expression levels are down-regulated in liver tissue after treatment with CCl4 compared with control mice. (C) miR-192-5p expression levels are down-regulated in liver tissue of patients with acute liver injury. (D) miR-192-5p expression in CD45+ immune cells (left panel) or hepatocytes (right panel) isolated from murine livers that underwent the I/R-procedure is down-regulated in hepatocytes compared with sham group. (E) Expression of miR-192-5p was analysed by qPCR in mice after treatment for 8 h with 2.5 μg/g body weight LPS or control substance. (F) Hepa1–6 cells were stimulated for 8 h with 20 ng/ml TNF and expression of miR-192-5p was determined by qPCR. (G) Hepa1–6 cells were transfected with 1.5 μg of empty plasmid (Ctrl) or IKKβE/E plasmid along with 200 ng of Igκ-Luc. Luciferase assay was performed and expression of miR-192-5p was determined by qPCR. The bar graphs show mean +/− S.E.M., r=Spearman rank correlation, *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Figure 1
Expression of miR-192-5p correlates with the degree of liver injury and cell death in acute liver injury

(A) miR-192-5p expression levels are down-regulated in liver tissue of I/R-operated mice compared with mice after sham surgery. (B) miR-192-5p expression levels are down-regulated in liver tissue after treatment with CCl4 compared with control mice. (C) miR-192-5p expression levels are down-regulated in liver tissue of patients with acute liver injury. (D) miR-192-5p expression in CD45+ immune cells (left panel) or hepatocytes (right panel) isolated from murine livers that underwent the I/R-procedure is down-regulated in hepatocytes compared with sham group. (E) Expression of miR-192-5p was analysed by qPCR in mice after treatment for 8 h with 2.5 μg/g body weight LPS or control substance. (F) Hepa1–6 cells were stimulated for 8 h with 20 ng/ml TNF and expression of miR-192-5p was determined by qPCR. (G) Hepa1–6 cells were transfected with 1.5 μg of empty plasmid (Ctrl) or IKKβE/E plasmid along with 200 ng of Igκ-Luc. Luciferase assay was performed and expression of miR-192-5p was determined by qPCR. The bar graphs show mean +/− S.E.M., r=Spearman rank correlation, *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Based on these data, we aimed to better define the molecular mechanism controlling the regulation of miR-192-5p. Hepatic I/R is associated with increased expression and release of different cytokines such as IL-1, IL-6, IL-17a, IFNγ, TNF and TGFβ [26]. In line with previous reports [27], we found increased miR-192-5p expression after stimulation with TGFβ, whereas stimulation with IL-6, IL-17a or IFNγ had no influence on miR-192-5p expression (Supplementary Figure S2). Binding of TNF to its receptor activates different signalling cascades controlling cell cycle or cell death pathways, where the most prominent is associated with NF-κB activation [28]. As a surrogate for the activation of the NF-κB pathway, we found a deregulation of various NF-κB target genes in I/R (Supplementary Figure S3). Therefore, we hypothesized that miR-192-5p might be regulated by TNF/NF-κB-dependent signalling cascades in hepatocytes. To validate this hypothesis, we first treated mice with LPS, a potent inducer of endogenous TNF from immune cells [29], and examined alterations in miR-192-5p expression. As shown in Figure 1(E), miR-192-5p was down-regulated in liver tissue after LPS stimulation, reflecting the regulation as found after I/R. Moreover, stimulation of hepatoma cells with TNF and IL-1β led to miR-192-5p down-regulation in vitro (Figure 1F and Supplementary Figure S2). To examine whether miR-192-5p represents a direct target of NF-κB, we transfected Hepa1–6 cells with a dominant active variant of IKKβ ([30], Figure 1G, left panel). In line with the hypothesis that miR-192 expression is dependent on NF-κB, miR-192 expression was lower (Figure 1G, right panel) when NF-κB was constitutively active.

Down-regulation of miR-192-5p protects from liver cell death by up-regulating Zeb2

I/R dependent liver injury has been shown to trigger cell death by increasing reactive oxygen species (ROS) [31]. To test the functional relevance of altered miR-192-5p expression in ROS-related cell death, we treated Hepa1-6 cells with H2O2 and measured miR-192-5p expression. These experiments revealed a trend towards a lower expression of miR-192-5p in H2O2-treated cells when compared with untreated cells (Figure 2A). On a functional level, the influence of miR-192-5p on cell viability was tested by overexpressing or knocking down miR-192-5p in Hepa1–6 hepatoma cells (Figures 2A and 2B). Overexpression or inhibition of miR-192-5p alone did not alter cell viability (Figure 2C). However, cells overexpressing miR-192-5p showed impaired viability upon H2O2 treatment as evidenced by MTT cell viability assays and microscopic cell evaluation (Figure 2D and Supplementary Figure S4B), whereas knocking down miR-192-5p expression was associated with an increased cell survival (Figure 4D, right panel), suggesting that the down-regulation of miR-192-5p in hepatocytes after acute liver injury might represent a protective mechanism against hepatocyte cell death.

Down-regulation of miR-192-5p is protective against liver cell damage

Figure 2
Down-regulation of miR-192-5p is protective against liver cell damage

(A) Hepa1–6 cells were treated for 24 h with 3 mM of H2O2 or left untreated and expression of miR-192-5p was analysed by PCR. (B) Hepa1–6 cells were transfected with 50 nM of synthetic miR-192-5p/antagomir or control vector and expression of miR-192-5p was analysed by qPCR 72 h after transfection. (C) Hepa1–6 cells were transfected with 50 nM of synthetic miR-192-5p/antagomir or control vector and cell viability was accessed 72 h after cell transfection. (D) Hepa1–6 cells were transfected with 50 nM of synthetic miR-192-5p/antagomir or control vector and 72 h after cell transfection cells were treated for 24 h with 3 mM of H2O2. Cell viability was accessed by applying an MTT assay as indicated by light microscopy. *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Figure 2
Down-regulation of miR-192-5p is protective against liver cell damage

(A) Hepa1–6 cells were treated for 24 h with 3 mM of H2O2 or left untreated and expression of miR-192-5p was analysed by PCR. (B) Hepa1–6 cells were transfected with 50 nM of synthetic miR-192-5p/antagomir or control vector and expression of miR-192-5p was analysed by qPCR 72 h after transfection. (C) Hepa1–6 cells were transfected with 50 nM of synthetic miR-192-5p/antagomir or control vector and cell viability was accessed 72 h after cell transfection. (D) Hepa1–6 cells were transfected with 50 nM of synthetic miR-192-5p/antagomir or control vector and 72 h after cell transfection cells were treated for 24 h with 3 mM of H2O2. Cell viability was accessed by applying an MTT assay as indicated by light microscopy. *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

To investigate the molecular mechanism mediating this protective effect, we used the in silico target prediction programme TargetScan, revealing 59 potential target genes of miR-192-5p (Supplementary Table S1). Based on our data as well as previously published data on the role of these genes in the regulation of cell death [3236], we measured expression of Cdc6, Cxcr5, Dnmt3a, Igf1 and Zeb2 in the different models of hepatic injury (Figures 3A–3C and Supplementary Figure S5). Only Zeb2 demonstrated a concordant down-regulation in all models (Figures 3A–3C) and was therefore chosen for further analysis. To establish a direct link between miR-192-5p and Zeb2 expression, we first performed luciferase based reporter assays demonstrating that miR-192-5p binds to the 3′-UTR of Zeb2 (Figure 3D). Next we transfected Hepa1–6 cells with miR-192-5p and measured Zeb2 gene expression. Of note, miR-192-5p overexpression led to a down-regulation of Zeb2 in a dose dependent manner, suggesting a direct regulation of this gene by miR-192-5p (Figure 3E and Supplementary Figure S6). On a functional level, the influence of Zeb2 on cell viability was tested by knocking down Zeb2 in Hepa1–6 hepatoma cells (Figure 3F). After inhibition of Zeb2, Hepa1–6 cells showed impaired viability upon H2O2 treatment as evidenced by MTT cell viability assays. (Figure 3G) Moreover, the effect of miR-192 inhibition (see also Figure 2D) could be reverted by co-transfecting Zeb2-siRNA (Figure 3H), suggesting that the effect of miR-192 on cell survival during acute liver injury is mediated by Zeb2.

Antiapototic Zeb2 represents a target gene of miR-192-5p in acute liver injury

Figure 3
Antiapototic Zeb2 represents a target gene of miR-192-5p in acute liver injury

(A) Expression of Zeb2 was analysed by qPCR in mice after induction of acute liver injury by I/R. (B) Expression of Zeb2 was analysed by qPCR in livers of mice after injection of 0.6 ml/kg body weight intraperitoneally CCl4 and compared with control mice. (C) Expression of Zeb2 was analysed by qPCR in Hepa1–6 cells treated for 24 h with 3 nM H2O2. (D) Hepa1–6 cells were transfected with 500 ng Zeb2-3′-UTR-psiCheck2™ luciferase construct along with 50 nM of either miR-negative or miR-192 mimic and luciferase activity was measured after 8 or 24 h transfection. (E) Hepa1–6 cells were transfected with 50 nM synthetic miR-192-5p and expression of Zeb2 was determined by qPCR. (F) Hepa1–6 cells were trans-fected with 50 nM of specific siRNA against Zeb2 or control vector and 72 h after cell transfection cells, Zeb2 expression was measured by qPCR. (G) Hepa1–6 cells were transfected with 50 nM of specific siRNA against Zeb2 or control vector and 72 h after cell transfection cells were treated for 24 h with 3 mM of H2O2. Cell viability was accessed by applying an MTT assay. (H) Hepa1–6 cells were transfected as indicated and treated for 24 h with 3 mM H2O. Cell viability was accessed by applying an MTT assay. *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Figure 3
Antiapototic Zeb2 represents a target gene of miR-192-5p in acute liver injury

(A) Expression of Zeb2 was analysed by qPCR in mice after induction of acute liver injury by I/R. (B) Expression of Zeb2 was analysed by qPCR in livers of mice after injection of 0.6 ml/kg body weight intraperitoneally CCl4 and compared with control mice. (C) Expression of Zeb2 was analysed by qPCR in Hepa1–6 cells treated for 24 h with 3 nM H2O2. (D) Hepa1–6 cells were transfected with 500 ng Zeb2-3′-UTR-psiCheck2™ luciferase construct along with 50 nM of either miR-negative or miR-192 mimic and luciferase activity was measured after 8 or 24 h transfection. (E) Hepa1–6 cells were transfected with 50 nM synthetic miR-192-5p and expression of Zeb2 was determined by qPCR. (F) Hepa1–6 cells were trans-fected with 50 nM of specific siRNA against Zeb2 or control vector and 72 h after cell transfection cells, Zeb2 expression was measured by qPCR. (G) Hepa1–6 cells were transfected with 50 nM of specific siRNA against Zeb2 or control vector and 72 h after cell transfection cells were treated for 24 h with 3 mM of H2O2. Cell viability was accessed by applying an MTT assay. (H) Hepa1–6 cells were transfected as indicated and treated for 24 h with 3 mM H2O. Cell viability was accessed by applying an MTT assay. *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Serum levels of miR-192-5p correlate with liver injury

Alterations of intracellular miRNA concentrations are frequently reflected by corresponding changes in serum levels [8]. Hence, we measured serum miR-192-5p levels in acute liver injury after I/R performed in mice [21]. I/R led to a time-dependent increase in miR-192-5p serum levels (Figure 4A). Of note, miR-192-5p strongly correlated with markers of liver cell death such as AST/ALT as well as with TUNEL-positive cells (Figure 4A and Supplementary Figure S3A) indicating a direct link between miR-192-5p elevation and the degree of liver injury after I/R. We and others have recently demonstrated that serum levels of the liver specific miRNA miR-122 represent a potent marker of liver injury and hepatic cell death [21,37]. Interestingly, in our present analysis, miR-192-5p concentrations strongly correlated with miR-122 (Supplementary Figure S7A). Importantly, the hepatic I/R results could be reproduced in the toxic CCl4, where serum miR-192-5p levels also correlated with the extent of liver injury (Supplementary Figure S7B). To confirm that hepatocytes are the main cells responsible for elevated miR-192-5p serum levels, we performed an in vitro hypoxia and reoxygenation experiment (H/R) with primary hepatocytes and analysed miR-192-5p concentrations in the supernatants. We detected significantly higher miR-192-5p levels in supernatants after H/R induced cells compared with untreated controls (Figure 4B). Interestingly, miR-192-5p concentrations significantly correlated with AST, ALT and miR-122 levels in the supernatant (Figure 4B and Supplementary Figure S7C), suggesting that hepatocytes are the major source of elevated miR-192-5p in serum after I/R injury in vivo.

Elevated miR-192-5p serum levels indicate liver injury

Figure 4
Elevated miR-192-5p serum levels indicate liver injury

(A) miR-192-5p serum concentrations are increased in mice after ischaemia compared with sham-operated mice that correlated to serum transaminases in I/R-operated mice after 24 h reperfusion. (B) miR-192-5p levels are significantly higher in supernatant of cells after H/R compared with control cells and correlate to transaminases. The bar graph shows mean +/− S.E.M., r=Spearman rank correlation. (C) miR-192-5p serum concentrations were analysed in a cohort of patients with ALF (n=28) and compared with healthy controls (n=30). miR-192-5p serum levels are depicted with respect to patients’ fate [SR (n=8), recovery after LTX (n=11), death (n=9)]. (D) Serum levels of miR-192-5p were determined in critically ill patients at admission to the medical ICU (n=223) and compared with healthy controls (n=76), revealing higher levels in patients compared with controls (left panel). miR-192-5p serum levels are elevated only in patients with liver injury according to elevated serum transaminases and miR-192-5p levels correlate with serum transaminases (middle panels). miR-192-5p serum levels do not differ with respect to the presence of liver cirrhosis in critically ill patients (right panel). *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Figure 4
Elevated miR-192-5p serum levels indicate liver injury

(A) miR-192-5p serum concentrations are increased in mice after ischaemia compared with sham-operated mice that correlated to serum transaminases in I/R-operated mice after 24 h reperfusion. (B) miR-192-5p levels are significantly higher in supernatant of cells after H/R compared with control cells and correlate to transaminases. The bar graph shows mean +/− S.E.M., r=Spearman rank correlation. (C) miR-192-5p serum concentrations were analysed in a cohort of patients with ALF (n=28) and compared with healthy controls (n=30). miR-192-5p serum levels are depicted with respect to patients’ fate [SR (n=8), recovery after LTX (n=11), death (n=9)]. (D) Serum levels of miR-192-5p were determined in critically ill patients at admission to the medical ICU (n=223) and compared with healthy controls (n=76), revealing higher levels in patients compared with controls (left panel). miR-192-5p serum levels are elevated only in patients with liver injury according to elevated serum transaminases and miR-192-5p levels correlate with serum transaminases (middle panels). miR-192-5p serum levels do not differ with respect to the presence of liver cirrhosis in critically ill patients (right panel). *P<0.05, **P<0.01, ***P<0.001. ns, not significant.

Recently, we have described that elevated serum levels of miR-122 represent an independent marker for acute liver diseases [21]. Thus, we hypothesized that alterations in miR-192-5p concentrations might also be indicative of injury and cell death in acute liver pathologies. Therefore, we analysed miR-192-5p serum levels in 28 patients with ALF and respective controls (cohort described in Supplementary Table S2). We found significantly elevated levels of miR-192-5p in the ALF-patients group (Figure 4C, left panel). Within this cohort of patients with different aetiologies, 8 patients spontaneously recovered from liver failure, 11 recovered after LTX and 9 died in the course of the disease. Unexpectedly, we found no correlation between miR-192-5p concentration and patients’ outcome (Figure 4C, right panel). To further identify factors regulating miR-192-5p concentrations in human diseases, we measured miR-192-5p serum levels in 76 healthy controls and a large and well defined cohort of 223 critically ill patients upon admission to the intensive care unit (ICU; Supplementary Table S3). This analysis revealed a significant up-regulation of miR-192-5p concentrations in critically ill patients (Figure 4D). However, miR-192-5p concentrations were not related to disease severity according to the Acute Physiology and Chronic Health Evaluation II (APACHE-II) scores (Supplementary Figure S8A), the presence of septic disease (Supplementary Figure S8B and Supplementary Table S4) or other factors such as pre-existing type 2 diabetes, obesity and patients’ age and sex (not shown). Furthermore, miR-192-5p serum levels were not indicative for patient's ICU or long-term prognosis in our cohort of patients (Supplementary Figure S8C).

Next, we analysed miR-192-5p levels in patients with or without ongoing liver damage (according to elevated transaminases). Strikingly, serum miR-192-5p concentrations were only elevated in patients with liver damage (Figure 4D, middle panels). Moreover, correlation analyses revealed that miR-192-5p serum levels were significantly associated with clinical markers of liver damage such as AST, ALT, glutamate dehydrogenase (GLDH) and miR-122 (Figure 4D, middle panels, Supplementary Figures S8D, S8E, S8F and Supplementary Table S1). In contrast no correlation with markers of liver function, markers of infection and inflammation and markers of renal function (Table 1) was found. Finally, the presence of liver cirrhosis did not further affect miR-192-5p levels in critically ill patients (Figure 4D, right panel). In summary, these data demonstrate that miR-192-5p is up-regulated in both experimental and human liver injury and might represent a hepatocyte-specific serum biomarker.

Table 1
Correlations of miR-192-5p serum concentrations at ICU admission with other laboratory markers

r, correlation coefficient; r and P-values by Spearman rank correlation.

ICU patients
ParameterrP
Liver damage 
AST 0.621 <0.001 
ALT 0.627 <0.001 
γGT 0.421 <0.001 
GLDH 0.670 <0.001 
LDH 0.318 <0.001 
Liver function 
Quick −0.284 <0.001 
INR 0.269 <0.001 
AT III −0.281 0.017 
Bili (total) 0.206 0.002 
Bili (direct) 0.331 <0.001 
Renal function 
Urea −0.309 <0.001 
Creatinine −0.280 <0.001 
GFR 0.258 0.001 
GFR_Cystatin 0.230 0.008 
Inflammation 
CRP −0.110 0.102 
PCT 0.030 0.707 
IL-6 0.163 0.087 
APRIL   
Others 
miR-122 0.869 <0.001 
ICU patients
ParameterrP
Liver damage 
AST 0.621 <0.001 
ALT 0.627 <0.001 
γGT 0.421 <0.001 
GLDH 0.670 <0.001 
LDH 0.318 <0.001 
Liver function 
Quick −0.284 <0.001 
INR 0.269 <0.001 
AT III −0.281 0.017 
Bili (total) 0.206 0.002 
Bili (direct) 0.331 <0.001 
Renal function 
Urea −0.309 <0.001 
Creatinine −0.280 <0.001 
GFR 0.258 0.001 
GFR_Cystatin 0.230 0.008 
Inflammation 
CRP −0.110 0.102 
PCT 0.030 0.707 
IL-6 0.163 0.087 
APRIL   
Others 
miR-122 0.869 <0.001 

DISCUSSION

In the current study, we demonstrate a hepatocyte-specific role of decreased miR-192-5p expression in protecting cells from acute liver injury. We propose that miR-192-5p is part of a TNF/LPS-dependent signalling pathway that mediates de-repression of antiapoptotic genes such as Zeb2 protecting hepatocytes from death during acute liver injury. Moreover, we show an elevation of miR-192-5p concentrations in sera of mice and patients with acute liver injury and demonstrate that increased serum levels of miR-192-5p correlate with the degree of liver damage.

I/R dependent liver injury as well as toxic injury are associated with massive apoptotic and necrotic cell death, finally leading to liver failure. Despite representing distinct pathophysiological concepts, apoptosis and necrosis might be commonly induced by the release of ROS such as H2O2 during hepatic I/R [31]. We demonstrate that miR-192-5p is down-regulated in murine I/R liver damage as well as upon treatment with H2O2, as an in vitro model of I/R. This down-regulation was triggered by TNF and LPS stimulation, being important NF-κB activators. Down-regulation of miR-192-5p displayed a protective effect towards H2O2-induced cell death. In line with our data, down-regulation of miR-192-5p was recently identified as a key characteristic of various cancers preventing cell death and promoting cell proliferation [38]. Along with the protective down-regulation of miR-192-5p, we demonstrate that Zeb2, a recently identified antiapoptotic target gene of miR-192-5p [38] is up-regulated in experimental liver injury as well as after exposure of Hepa1–6 cells to H2O2, providing a molecular link between cell protection and miR-192-5p down-regulation.

Zeb2 is a zinc finger DNA-binding protein which functions as a master transcriptional regulator especially during EMT in liver cancer. On a molecular level, TNF directly induces Zeb2 in cholangiocellular cancer cells and thereby promotes cell survival and disease progression [39]. We demonstrate that next to Zeb2, the down-regulation of miR-192-5p is mediated by TNF. Thus, these data suggest a new cascade down-stream of TNF consisting of miR-192-5p and Zeb2, which is involved in the control of cell survival in response to acute liver injury.

miR-based therapies recently gained considerable attention in the context of patients with chronic HCV genotype 1 infection. Administration of miRavirsen, a miR-122 inhibitor, dramatically decreased HCV replication without relevant side effects [40]. Considering the cell protective effect of miR-192-5p in acute liver injury, as well as the published evidence of its antiapoptotic effect in other models, it seems reasonable to speculate that miR-192-5p may represent an important therapeutic target to treat acute liver diseases. In contrast with targeting of single genes, miR-based therapeutic approaches influence whole gene networks and might therefore be more efficient in complex pathological scenarios such as acute liver diseases. Of note, further experiments, using genetically modified animals might be helpful to test this hypothesis in a pre-clinical setting.

In the present paper we show elevated serum levels of miR-192-5p in experimental and human liver injury. These data are in line with previous reports, demonstrating elevated miR-192-5p concentrations in patients with acute liver damage as well as in experimental models of acute liver injury [8,9,12,14,15]. Serum levels of liver specific miRNAs were found to increase earlier than transaminases in an acetaminophen-induced hepatic failure [8]. In line with this, we detected a higher correlation of miR-192-5p with hepatocyte death according to TUNEL staining (terminal deoxynucleotidyltransferase-mediated dUTP biotin nick-end labelling staining) as compared with AST/ALT serum levels in mice. Moreover, alterations of miR-192-5p levels in critically ill patients were not biased by disease conditions that might cause falsely elevated AST/ALT serum levels such as cardiac diseases or pre-existing liver cirrhosis. These results support a potential use of miR-192-5p measurements to specifically detect acute liver injury at an early time point in patients with multiple or complex diseases. In this context, miR-192-5p was found to be part of a panel of circulating miRNAs, also comprising miR-122, that correlate with ALT/AST serum levels in a model of acetaminophen-induced liver injury [8,41]. In summary, our data as well as the previous reports on miR-192 as a biomarker in the setting of acute liver diseases, suggest a ‘translational’ potential of this miRNA. However, before clinical use can be considered, several problems need to be solved: at present, there is no established consensus on standardization of sample collection, data normalization and analysis, providing a rationale for some inter-study differences in the regulation of miR-192. Moreover all available studies are based on retrospective analysis and presently no longitudinal, prospective data are available, highlighting the need for further research in this field.

We found a remarkably close correlation between miR-192-5p and miR-122 serum levels in experimental liver injury as well as in patients with acute liver injury, suggesting common mechanisms regulating the equilibrium between intra- and extra-cellular serum levels of these miRNAs. Bala et al. [42] showed an association of miR-122 and exosome-rich fraction in both alcoholic and inflammatory liver damage, suggesting an active mechanism driving the secretion of miR-122 into the circulation. Similar to miR-122, we observed a significant and time-dependent down-regulation of miR-192-5p in acute liver injury. These findings suggest that the lower intracellular levels of miR-192-5p in hepatocytes might be the consequence of an increased release into the circulation rather than a down-regulation of miRNA expression per se. These findings highlight the inverse relationship between tissue and circulating miR-192-5p and may reflect an active process of cellular adaptation during acute liver injury, an issue that could be addressed using miR-192-5p knockout mice.

In summary, we here demonstrate for the first time that miR-192-5p is involved in regulating liver cell death during acute liver injury and might represent a potent marker of hepatic injury and liver cell death. These data help to improve diagnostic assessments in patients with acute liver injury and should provoke further research to validate our results in larger and prospective studies in critically ill patients.

AUTHOR CONTRIBUTION

Alexander Koch, Anne Schneider, Christoph Roderburg, Christian Trautwein, Fabian Benz, Frank Tacke, Florian Schüller, Heike Bantel, Jan Alder, Jeremie Gautheron, Joern Janssen, Mark Luedde, Mihael Vucur, Sven Loosen, Sanchari Roy and Tom Luedde designed the study. Anne Schneider, Fabian Benz, Christoph Roderburg, Jan Alder, Joern Janssen and Sanchari Roy performed measurements. Alexander Koch, Christoph Roderburg, Fabian Benz, Jan Alder, Mark Luedde, Frank Tacke and Sanchari Roy performed statistical analyses, analysed data. Christoph Roderburg, Mark Luedde, Jeremie Gautheron, Sanchari Roy and Tom Luedde wrote the manuscript. Alexander Koch, Frank Tacke and Heike Bantel organized patient recruitment and collected human serum samples. All authors read and approved the manuscript.

The authors would like to express their gratitude to Francisco Javier Cubero for providing us with cytokines, Michaela Roderburg-Goor, Dr Jane Beger-Lüdde and the members of the Lüdde-lab for helpful discussions.

FUNDING

This work was supported by the German Research Foundation [grant number DFG RO 4317/4-1]; the START grant from the medical faculty RWTH Aachen (to C.R.); the European Research Council within the FP 7 [grant number ERC-2007-Stg/208237-Luedde-Med3-Aachen]; the German-Research-Foundation [grant number SFB-TRR57, P06+P09]; the German Cancer Aid [Deutsche Krebshilfe grant number 110043]; the EMBO Young Investigator Program; the Ernst-Jung-Foundation Hamburg; and the Medical Faculty Aachen, from the Interdisciplinary Center of Clinical Research (‘IZKF’) Aachen [grant number 114/14].

Abbreviations

     
  • ALF

    acute liver failure

  •  
  • ALT

    alanine transaminase

  •  
  • APRIL

    a proliferation inducing ligand

  •  
  • AST

    aspartate transaminase

  •  
  • AT III

    antithrombin III

  •  
  • Bili

    bilirubin

  •  
  • CCl4

    carbon tetrachloride

  •  
  • CRP

    C-reactive protein

  •  
  • GFR

    glomerular filtration rate

  •  
  • GLDH

    glutamate dehydrogenase

  •  
  • H/R

    hypoxia and reoxygenation

  •  
  • H2O2

    hydrogen peroxide

  •  
  • I/R

    ischaemia/reperfusion

  •  
  • ICU

    intensive care unit

  •  
  • IFNγ

    interferon-γ

  •  
  • IL-1/6

    interleukin-1/6

  •  
  • INR

    international normalized ratio

  •  
  • LDH

    lactate dehydrogenase

  •  
  • LPS

    lipopolysaccharide

  •  
  • LTX

    liver transplantation

  •  
  • NF-κB

    nuclear factor ‘kappa-light-chain-enhancer’ of activated B-cells

  •  
  • PCT

    procalcitonin

  •  
  • qPCR

    quantitative real-time PCR

  •  
  • ROS

    reactive oxygen species

  •  
  • SR

    spontaneous recovery

  •  
  • TNF

    tumour necrosis factor

  •  
  • TUNEL staining

    terminal deoxynucleotidyltransferase-mediated dUTP biotin nick-end labelling staining

  •  
  • TGFβ

    transforming growth factor-β

  •  
  • γGT

    γ-glutamyl transferase

References

References
1
Kloosterman
 
W.P.
Plasterk
 
R.H.
 
The diverse functions of microRNAs in animal development and disease
Dev. Cell
2006
, vol. 
11
 (pg. 
441
-
450
)
[PubMed]
2
Ambros
 
V.
 
The functions of animal microRNAs
Nature
2004
, vol. 
431
 (pg. 
350
-
355
)
[PubMed]
3
Roderburg
 
C.
Urban
 
G.W.
Bettermann
 
K.
Vucur
 
M.
Zimmermann
 
H.
Schmidt
 
S.
Janssen
 
J.
Koppe
 
C.
Knolle
 
P.
Castoldi
 
M.
, et al 
Micro-RNA profiling reveals a role for miR-29 in human and murine liver fibrosis
Hepatology
2011
, vol. 
53
 (pg. 
209
-
218
)
[PubMed]
4
Reid
 
G.
Kirschner
 
M.B.
van Zandwijk
 
N.
 
Circulating microRNAs: association with disease and potential use as biomarkers
Crit. Rev. Oncol. Hematol.
2011
, vol. 
80
 (pg. 
193
-
208
)
[PubMed]
5
Pineau
 
P.
Volinia
 
S.
McJunkin
 
K.
Marchio
 
A.
Battiston
 
C.
Terris
 
B.
Mazzaferro
 
V.
Lowe
 
S.W.
Croce
 
C.M.
Dejean
 
A.
 
miR-221 overexpression contributes to liver tumorigenesis
Proc. Natl. Acad. Sci. U.S.A.
2010
, vol. 
107
 (pg. 
264
-
269
)
[PubMed]
6
Tacke
 
F.
Roderburg
 
C.
Benz
 
F.
Cardenas
 
D.V.
Luedde
 
M.
Hippe
 
H.J.
Frey
 
N.
Vucur
 
M.
Gautheron
 
J.
Koch
 
A.
, et al 
Levels of circulating miR-133a are elevated in sepsis and predict mortality in critically ill patients
Crit. Care Med.
2014
, vol. 
42
 (pg. 
1096
-
1104
)
[PubMed]
7
Roderburg
 
C.
Luedde
 
M.
Vargas Cardenas
 
D.
Vucur
 
M.
Mollnow
 
T.
Zimmermann
 
H.W.
Koch
 
A.
Hellerbrand
 
C.
Weiskirchen
 
R.
Frey
 
N.
, et al 
miR-133a mediates TGF-beta-dependent derepression of collagen synthesis in hepatic stellate cells during liver fibrosis
J. Hepatol.
2013
, vol. 
58
 (pg. 
736
-
742
)
[PubMed]
8
Wang
 
K.
Zhang
 
S.
Marzolf
 
B.
Troisch
 
P.
Brightman
 
A.
Hu
 
Z.
Hood
 
L.E.
Galas
 
D.J.
 
Circulating microRNAs, potential biomarkers for drug-induced liver injury
Proc. Natl. Acad. Sci. U.S.A.
2009
, vol. 
106
 (pg. 
4402
-
4407
)
[PubMed]
9
Nagano
 
T.
Higashisaka
 
K.
Kunieda
 
A.
Iwahara
 
Y.
Tanaka
 
K.
Nagano
 
K.
Abe
 
Y.
Kamada
 
H.
Tsunoda
 
S.
Nabeshi
 
H.
, et al 
Liver-specific microRNAs as biomarkers of nanomaterial-induced liver damage
Nanotechnology
2013
, vol. 
24
 pg. 
405102
 
[PubMed]
10
Church
 
R.J.
Otieno
 
M.
McDuffie
 
J.E.
Singh
 
B.
Sonee
 
M.
Hall
 
L.
Watkins
 
P.B.
Ellinger-Ziegelbauer
 
H.
Harrill
 
AH.
 
Beyond miR-122: identification of microRNA alterations in blood during a time course of hepatobiliary injury and biliary hyperplasia in rats
Toxicol. Sci.
2016
, vol. 
150
 (pg. 
3
-
14
)
[PubMed]
11
Tryndyak
 
V.P.
Latendresse
 
J.R.
Montgomery
 
B.
Ross
 
S.A.
Beland
 
F.A.
Rusyn
 
I.
Pogribny
 
I.P.
 
Plasma microRNAs are sensitive indicators of inter-strain differences in the severity of liver injury induced in mice by a choline- and folate-deficient diet
Toxicol. App. Pharmacol.
2012
, vol. 
262
 (pg. 
52
-
59
)
12
van der Meer
 
A.J.
Farid
 
W.R.
Sonneveld
 
M.J.
de Ruiter
 
P.E.
Boonstra
 
A.
van Vuuren
 
A.J.
Verheij
 
J.
Hansen
 
B.E.
de Knegt
 
R.J.
van der Laan
 
L.J.
Janssen
 
H.L.
 
Sensitive detection of hepatocellular injury in chronic hepatitis C patients with circulating hepatocyte-derived microRNA-122
J. Viral Hepat.
2013
, vol. 
20
 (pg. 
158
-
166
)
[PubMed]
13
Starkey Lewis
 
P.J.
Dear
 
J.
Platt
 
V.
Simpson
 
K.J.
Craig
 
D.G.
Antoine
 
D.J.
French
 
N.S.
Dhaun
 
N.
Webb
 
D.J.
Costello
 
E.M.
, et al 
Circulating microRNAs as potential markers of human drug-induced liver injury
Hepatology
2011
, vol. 
54
 (pg. 
1767
-
1776
)
[PubMed]
14
Hu
 
J.
Wang
 
Z.
Tan
 
C.J.
Liao
 
B.Y.
Zhang
 
X.
Xu
 
M.
Dai
 
Z.
Qiu
 
S.J.
Huang
 
X.W.
Sun
 
J.
, et al 
Plasma microRNA, a potential biomarker for acute rejection after liver transplantation
Transplantation
2013
, vol. 
95
 (pg. 
991
-
999
)
[PubMed]
15
Krauskopf
 
J.
Caiment
 
F.
Claessen
 
S.M.
Johnson
 
K.J.
Warner
 
R.L.
Schomaker
 
S.J.
Burt
 
D.A.
Aubrecht
 
J.
Kleinjans
 
J.C.
 
Application of high-throughput sequencing to circulating microRNAs reveals novel biomarkers for drug-induced liver injury
Toxicol. Sci.
2015
, vol. 
143
 (pg. 
268
-
276
)
[PubMed]
16
Jenkins
 
R.H.
Martin
 
J.
Phillips
 
A.O.
Bowen
 
T.
Fraser
 
D.J.
 
Transforming growth factor beta1 represses proximal tubular cell microRNA-192 expression through decreased hepatocyte nuclear factor DNA binding
Biochem. J.
2012
, vol. 
443
 (pg. 
407
-
416
)
[PubMed]
17
Silakit
 
R.
Loilome
 
W.
Yongvanit
 
P.
Chusorn
 
P.
Techasen
 
A.
Boonmars
 
T.
Khuntikeo
 
N.
Chamadol
 
N.
Pairojkul
 
C.
Namwat
 
N.
 
Circulating miR-192 in liver fluke-associated cholangiocarcinoma patients: a prospective prognostic indicator
J. Hepatobiliary Pancreat. Sci.
2014
, vol. 
21
 (pg. 
864
-
872
)
[PubMed]
18
Luedde
 
T.
Assmus
 
U.
Wustefeld
 
T.
Meyer zu Vilsendorf
 
A.
Roskams
 
T.
Schmidt-Supprian
 
M.
Rajewsky
 
K.
Brenner
 
D.A.
Manns
 
M.P.
Pasparakis
 
M.
Trautwein
 
C.
 
Deletion of IKK2 in hepatocytes does not sensitize these cells to TNF-induced apoptosis but protects from ischemia/reperfusion injury
J. Clin. Invest.
2005
, vol. 
115
 (pg. 
849
-
859
)
[PubMed]
19
Vucur
 
M.
Reisinger
 
F.
Gautheron
 
J.
Janssen
 
J.
Roderburg
 
C.
Cardenas
 
D.V.
Kreggenwinkel
 
K.
Koppe
 
C.
Hammerich
 
L.
Hakem
 
R.
, et al 
RIP3 inhibits inflammatory hepatocarcinogenesis but promotes cholestasis by controlling caspase-8- and JNK-dependent compensatory cell proliferation
Cell Rep
2013
, vol. 
4
 (pg. 
776
-
790
)
[PubMed]
20
Bettermann
 
K.
Vucur
 
M.
Haybaeck
 
J.
Koppe
 
C.
Janssen
 
J.
Heymann
 
F.
Weber
 
A.
Weiskirchen
 
R.
Liedtke
 
C.
Gassler
 
N.
, et al 
TAK1 suppresses a NEMO-dependent but NF-kappaB-independent pathway to liver cancer
Cancer Cell
2010
, vol. 
17
 (pg. 
481
-
496
)
[PubMed]
21
Roderburg
 
C.
Benz
 
F.
Vargas Cardenas
 
D.
Koch
 
A.
Janssen
 
J.
Vucur
 
M.
Gautheron
 
J.
Schneider
 
A.T.
Koppe
 
C.
Kreggenwinkel
 
K.
, et al 
Elevated miR-122 serum levels are an independent marker of liver injury in inflammatory diseases
Liver Int
2015
, vol. 
35
 (pg. 
1172
-
1184
)
[PubMed]
22
Gautheron
 
J.
Pescatore
 
A.
Fusco
 
F.
Esposito
 
E.
Yamaoka
 
S.
Agou
 
F.
Ursini
 
M.V.
Courtois
 
G.
 
Identification of a new NEMO/TRAF6 interface affected in incontinentia pigmenti pathology
Hum. Mol. Genet.
2010
, vol. 
19
 (pg. 
3138
-
3149
)
[PubMed]
23
Benz
 
F.
Roderburg
 
C.
Vargas Cardenas
 
D.
Vucur
 
M.
Gautheron
 
J.
Koch
 
A.
Zimmermann
 
H.
Janssen
 
J.
Nieuwenhuijsen
 
L.
Luedde
 
M.
, et al 
U6 is unsuitable for normalization of serum miRNA levels in patients with sepsis or liver fibrosis
Exp. Mol. Med.
2013
, vol. 
45
 pg. 
e42
 
[PubMed]
24
Roy
 
S.
Benz
 
F.
Vargas Cardenas
 
D.
Vucur
 
M.
Gautheron
 
J.
Schneider
 
A.
Hellerbrand
 
C.
Pottier
 
N.
Alder
 
J.
Tacke
 
F.
, et al 
miR-30c and miR-193 are a part of the TGF-beta-dependent regulatory network controlling extracellular matrix genes in liver fibrosis
J. Dig. Dis.
2015
, vol. 
16
 (pg. 
513
-
524
)
[PubMed]
25
John
 
K.
Hadem
 
J.
Krech
 
T.
Wahl
 
K.
Manns
 
M.P.
Dooley
 
S.
Batkai
 
S.
Thum
 
T.
Schulze-Osthoff
 
K.
Bantel
 
H.
 
MicroRNAs play a role in spontaneous recovery from acute liver failure
Hepatology
2014
, vol. 
60
 (pg. 
1346
-
1355
)
[PubMed]
26
Takahashi
 
Y.
Ganster
 
R.W.
Gambotto
 
A.
Shao
 
L.
Kaizu
 
T.
Wu
 
T.
Yagnik
 
G.P.
Nakao
 
A.
Tsoulfas
 
G.
Ishikawa
 
T.
, et al 
Role of NF-kappaB on liver cold ischemia-reperfusion injury
Am. J. Physiol. Gastrointest. Liver Physiol.
2002
, vol. 
283
 (pg. 
G1175
-
G1184
)
[PubMed]
27
Meng
 
X.M.
Tang
 
P.M.
Li
 
J.
Lan
 
H.Y.
 
TGF-beta/Smad signaling in renal fibrosis
Front. Physiol.
2015
, vol. 
6
 pg. 
82
 
[PubMed]
28
Teoh
 
N.
Field
 
J.
Sutton
 
J.
Farrell
 
G.
 
Dual role of tumor necrosis factor-alpha in hepatic ischemia-reperfusion injury: studies in tumor necrosis factor-alpha gene knockout mice
Hepatology
2004
, vol. 
39
 (pg. 
412
-
421
)
[PubMed]
29
Luedde
 
T.
Kaplowitz
 
N.
Schwabe
 
R.F.
 
Cell death and cell death responses in liver disease: mechanisms and clinical relevance
Gastroenterology
2014
, vol. 
147
 (pg. 
765
-
783
)
[PubMed]
30
Gautheron
 
J.
Vucur
 
M.
Reisinger
 
F.
Cardenas
 
D.V.
Roderburg
 
C.
Koppe
 
C.
Kreggenwinkel
 
K.
Schneider
 
A.T.
Bartneck
 
M.
Neumann
 
U.P.
, et al 
A positive feedback loop between RIP3 and JNK controls non-alcoholic steatohepatitis
EMBO Mol. Med.
2014
, vol. 
6
 (pg. 
1062
-
1074
)
[PubMed]
31
Zhang
 
W.
Wang
 
M.
Xie
 
H.Y.
Zhou
 
L.
Meng
 
X.Q.
Shi
 
J.
Zheng
 
S.
 
Role of reactive oxygen species in mediating hepatic ischemia-reperfusion injury and its therapeutic applications in liver transplantation
Transplant. Proc.
2007
, vol. 
39
 (pg. 
1332
-
1337
)
[PubMed]
32
Rubinfeld
 
H.
Kammer
 
A.
Cohen
 
O.
Gorshtein
 
A.
Cohen
 
Z.R.
Hadani
 
M.
Werner
 
H.
Shimon
 
I.
 
IGF1 induces cell proliferation in human pituitary tumors–functional blockade of IGF1 receptor as a novel therapeutic approach in non-functioning tumors
Mol. Cell. Endocrinol.
2014
, vol. 
390
 (pg. 
93
-
101
)
[PubMed]
33
Qiuping
 
Z.
Jie
 
X.
Youxin
 
J.
Qun
 
W.
Wei
 
J.
Chun
 
L.
Jin
 
W.
Yan
 
L.
Chunsong
 
H.
Mingzhen
 
Y.
, et al 
Selectively frequent expression of CXCR5 enhances resistance to apoptosis in CD8(+)CD34(+) T cells from patients with T-cell-lineage acute lymphocytic leukemia
Oncogene
2005
, vol. 
24
 (pg. 
573
-
584
)
[PubMed]
34
Shen
 
J.Z.
Zhang
 
Y.Y.
Fu
 
H.Y.
Wu
 
D.S.
Zhou
 
H.R.
 
Overexpression of microRNA-143 inhibits growth and induces apoptosis in human leukemia cells
Oncol. Rep.
2014
, vol. 
31
 (pg. 
2035
-
2042
)
[PubMed]
35
Dasgupta
 
P.
Rizwani
 
W.
Pillai
 
S.
Davis
 
R.
Banerjee
 
S.
Hug
 
K.
Lloyd
 
M.
Coppola
 
D.
Haura
 
E.
Chellappan
 
S.P.
 
ARRB1-mediated regulation of E2F target genes in nicotine-induced growth of lung tumors
J. Natl. Cancer Inst.
2011
, vol. 
103
 (pg. 
317
-
333
)
[PubMed]
36
Qi
 
S.
Song
 
Y.
Peng
 
Y.
Wang
 
H.
Long
 
H.
Yu
 
X.
Li
 
Z.
Fang
 
L.
Wu
 
A.
Luo
 
W.
, et al 
ZEB2 mediates multiple pathways regulating cell proliferation, migration, invasion, and apoptosis in glioma
PLoS One
2012
, vol. 
7
 pg. 
e38842
 
[PubMed]
37
Clarke
 
J.D.
Sharapova
 
T.
Lake
 
A.D.
Blomme
 
E.
Maher
 
J.
Cherrington
 
N.J.
 
Circulating microRNA 122 in the methionine and choline-deficient mouse model of non-alcoholic steatohepatitis
J. Appl. Toxicol.
2014
, vol. 
34
 (pg. 
726
-
732
)
[PubMed]
38
Kim
 
T.
Veronese
 
A.
Pichiorri
 
F.
Lee
 
T.J.
Jeon
 
Y.J.
Volinia
 
S.
Pineau
 
P.
Marchio
 
A.
Palatini
 
J.
Suh
 
S.S.
, et al 
p53 regulates epithelial-mesenchymal transition through microRNAs targeting ZEB1 and ZEB2
J. Exp. Med.
, vol. 
208
 (pg. 
875
-
883
)
39
Techasen
 
A.
Namwat
 
N.
Loilome
 
W.
Duangkumpha
 
K.
Puapairoj
 
A.
Saya
 
H.
Yongvanit
 
P
 
Tumor necrosis factor-alpha modulates epithelial mesenchymal transition mediators ZEB2 and S100A4 to promote cholangiocarcinoma progression
J. Hepatobiliary Pancreat. Sci.
, vol. 
21
 (pg. 
703
-
711
)
[PubMed]
40
Janssen
 
H.L.
Reesink
 
H.W.
Lawitz
 
E.J.
Zeuzem
 
S.
Rodriguez-Torres
 
M.
Patel
 
K.
van der Meer
 
A.J.
Patick
 
A.K.
Chen
 
A.
Zhou
 
Y.
, et al 
Treatment of HCV infection by targeting microRNA
N. Engl. J. Med.
2013
, vol. 
368
 (pg. 
1685
-
1694
)
[PubMed]
41
Krauskopf
 
J.
Caiment
 
F.
Claessen
 
S.M.
Johnson
 
K.J.
Warner
 
R.L.
Schomaker
 
S.J.
Burt
 
D.A.
Aubrecht
 
J.
Kleinjans
 
J.C.
 
Application of high-throughput sequencing to circulating microRNAs reveals novel biomarkers for drug-induced liver injury
Toxicol. Sci.
2015
, vol. 
143
 (pg. 
268
-
276
)
[PubMed]
42
Bala
 
S.
Petrasek
 
J.
Mundkur
 
S.
Catalano
 
D.
Levin
 
I.
Ward
 
J.
Alao
 
H.
Kodys
 
K.
Szabo
 
G.
 
Circulating microRNAs in exosomes indicate hepatocyte injury and inflammation in alcoholic, drug-induced, and inflammatory liver diseases
Hepatology
2012
, vol. 
56
 (pg. 
1946
-
1957
)
[PubMed]

Author notes

1

These authors share first authorship.