From June 2015 to July 2017, a total of 66 patients participated in the present study, including 40 patients with ALF and 26 healthy controls (HCs). The median age of the study subjects was 42 years (age range: 15–82 years) and 72.73% of patients were male. Enrolled patients were treated in accordance with the scheduled criteria (18), and the inclusion criteria were as follows: An illness duration of <26 weeks in a patient without preexisting liver disease or cirrhosis associated with any degree of mental status alteration (encephalopathy) and coagulopathy (an international normalized ratio ≥1.5). The exclusion criteria were as follows: i) Liver transplant; ii) the presence of any concomitant illness; and iii) the use of liver support devices. All enrolled participants were hospitalized and followed-up at the Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou Medical College (Hangzhou, China). Written informed consent was obtained from subjects prior to participation. The study was performed according to the Ethical Principles for Medical Research Involving Human Subjects outlined in The Helsinki Declaration in 1975, and the research was approved by the Ethics Committee of Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou Medical College (approval no. ZRY-T123-59).

A total of 21 C57BL/6 male mice (age, 6–8 weeks; weight, 18–22 g), which were randomly divided into three groups (n=7 mice/group), were obtained from the Experimental Animal Center of the Chinese Academy of Sciences. All mice were raised at a room temperature of 22±2°C and 55±5% relative humidity, with free access to food and water. Mice were acclimated to the environment for 6–7 days under a 12-h dark/light cycle.

The combination of LPS/GalN (MilliporeSigma) was dissolved and injected intraperitoneally (400 mg/kg GalN and 10 µg/kg LPS). To assess the role of miR-21 in LPS/GalN-induced ALF, mice were challenged with a tail vein injection of 20 nmol miRNA antagomir negative control (NC) or 20 nmol miR-21 antagomir (both from Shanghai GenePharma Co., Ltd.). The mice were then challenged intraperitoneally with 10 µg/kg LPS and 400 mg/kg GalN, or phosphate-buffered saline (PBS; control), after 48 h. Mice were sacrificed if they exhibited listlessness, respiratory abnormalities, loss of appetite, weight loss (maximum 20%) and disordered mobility. Otherwise, the mice were sacrificed 24 h after LPS/GalN injection. Mice were euthanized by cervical dislocation following anesthesia with an intraperitoneal injection of sodium pentobarbital (35 mg/kg). All procedures were conducted in accordance with the guide for the care and use of laboratory animals (19) and the study was approved by the Ethics Committee of Zhejiang Provincial People's Hospital, People's Hospital of Hangzhou Medical College (approval no. 20190010).

THP-1 cells were purchased from Pricella Life Science & Technology Co., Ltd. The cells (3×10⁵ cells/well) were cultured in 6-well plates in RPMI-1640 (cat. no. 11875119; Gibco; Thermo Fisher Scientific, Inc.) containing 10% heat-inactivated fetal calf serum (cat. no. 12484028; Gibco; Thermo Fisher Scientific, Inc.) and 1% penicillin-streptomycin (Hyclone; Cytiva) at 37°C in a 5% CO₂ incubator. To obtain THP-1-derived macrophages, THP-1 cells were then treated with 160 ng/ml phorbol 12-myristate 13-acetate (PMA; cat. no. P8139; MilliporeSigma) followed by 24 h of incubation at 37°C in a 5% CO₂ incubator in RPMI-1640 to obtain M0 macrophages. The macrophages were polarized to M1 macrophages by incubation with 10 pg/ml LPS (MilliporeSigma) and 20 ng/ml IFN-γ (R&D Systems, Inc.) for 18 h at 37°C in a 5% CO₂ incubator. Then, the THP-1-derived macrophages were maintained in fresh RPMI-1640.

Total miRNA was extracted and purified from 150 µl mouse human plasma, which was obtained by centrifuging blood samples at 2,000 × g at 4°C for 10 min using the miRNeasy Serum/Plasma Kit (cat. no. 217184; Qiagen GmbH) according to the manufacturer's instructions. The quality and concentration were assessed using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc.) based on the ratio of absorbance at 260 nm. The purified miRNA was immediately reverse-transcribed using the miScript II Reverse Transcription Kit (Qiagen GmbH) according to the manufacturer's instructions. The expression levels of serum miR-21 were quantified using an Applied Biosystems 7500 real-time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). qPCR was performed using miScript SYBR Green PCR Kit with primers specific for miR-21 (all from QIAGEN GmbH) according to the manufacturer's instructions. The thermocycling protocol was: 15 min at 95°C, followed by 40 cycles at 94°C for 15 sec, 55°C for 30 sec and 70°C for 30 sec. The primer sequences are listed in Table I.

Total RNA was extracted from THP-1-derived macrophages and from mouse liver tissues, which were collected when mice were sacrificed, using TRIzol^® reagent (Invitrogen; Thermo Fisher Scientific, Inc.). The quality and concentration of RNA were assessed using a NanoDrop 2000 spectrophotometer based on the ratio of absorbance at 260 nm. A PrimeScript ™ RT Reagent kit (Takara Bio, Inc.) was used to reverse transcribe 2 µg RNA into cDNA using the following protocol: 25°C for 5 min, 42°C for 30 min and 85°C for 5 min, followed by maintenance at 4°C for 5 min. Amplification of the cDNA was performed by qPCR using a SYBR Premix Ex Taq™ II kit (Takara Bio, Inc.). The thermocycling protocol was: 95°C for 3 min, followed by 40 cycles of denaturation at 95°C for 30 sec, annealing at 60°C for 30 sec and extension at 72°C for 1 min, and a final extension step at 72°C for 7 min. The primer sequences are listed in Table I. The relative expression levels of mRNA and miRNA were calculated using the 2^−ΔΔCq method (20). GAPDH was used as an endogenous control for mRNA expression and U6 was used as an endogenous control for miRNA expression. The specificity of the products was verified using melting curve analysis.

Total protein was extracted from mouse liver tissue and THP-1-derived macrophages using RIPA lysis buffer (Beyotime Institute of Biotechnology) and proteins were quantified using a bicinchoninic acid protein assay kit (cat. no. P0012S; Beyotime Institute of Biotechnology). An equal amount of protein (50 µg/lane) was separated by SDS on 12% gels and was then transferred to a nitrocellulose blotting membrane. The membranes were blocked with 5% non-fat milk dissolved in TBS-0.1% Tween buffer for 2 h at room temperature and were then probed with the following primary antibodies: IL-23 (1:1,000; cat. no. ab190356; Abcam), KLF6 (1:1,000; cat. no. DF13114; Affinity Biosciences), LC3A/B (1:1,000; cat.no. ab62721; Abcam), p62 (1:1,000; cat. no. ab155686; Abcam), ATG7 (1:1,000; cat. no. ab133528; Abcam), Beclin1 (1:1,000; cat. no. ab207612; Abcam) and GAPDH (1:2,500; cat. no. ab9485; Abcam). The membranes were then incubated with HRP-conjugated anti-mouse or anti-rabbit secondary antibodies (1:2,000; cat. nos. ab6789 and ab6721; Abcam) for 2 h at room temperature. Signals were visualized using enhanced chemiluminescence reagent (Thermo Fisher Scientific, Inc.) according to the manufacturer's instructions. Densitometric analysis was performed using ImageJ (Version 1.49; National Institutes of Health).

Liver sections collected from mice after sacrifice were fixed with 10% neutral-buffered formalin for 36 h at room temperature, embedded in paraffin and cut into 5-µm slices. After deparaffinization and rehydration, the slices were stained with 1% hematoxylin for 10 min and 0.5% eosin for 3 min at room temperature. Histological changes were evaluated using a light microscope by two experienced pathologists.

Plasma was obtained by centrifuging mouse blood samples at 2,000 × g for 10 min at 4°C and was then stored at −80°C. Alanine transaminase (ALT; cat. no. C009-2; Nanjing Jiancheng Bioengineering Institute), aspartate transaminase (AST; cat. no. C010-2; Nanjing Jiancheng Bioengineering Institute) and IL-23 (cat. no. BMS6017TEN; Invitrogen; Thermo Fisher Scientific, Inc.) kits were used to assess the levels of ALT, AST and IL-23, according to the manufacturers' protocols. The signal was measured at 450 nm excitation using a SpectraMax^® i3 multifunctional microplate reader (Molecular Devices, LLC). After the standard curve was established, the plasma levels were calculated based on the absorbance values of each sample.

The miR-21 mimics (sense: 5′-UAGCUUAUCAGACUGAUGUUGA-3′; antisense: 5′-AACAUCAGUCUGAUAAGCUAUU-3′), miR-21 inhibitor (5′-UCAACAUCAGUCUGAUAAGCUA-3′), miR-21 mimics negative control (NC) (sense: 5′-UUCUCCGAACGUGUCACGUTT-3′; antisense: 5′-ACGUGACACGUUCGGAGAATT-3′) and miR-21 inhibitor NC (5′-CAGUACUUUUGUGUAGUACAA-3′) were purchased from Guangzhou RiboBio Co., Ltd. THP-1-derived macrophages (3×10⁵ cells/well) were inoculated in 6-well plates and the transfection experiment was started when the cell confluence reached 70%. During transfection, 12 µl Lipofectamine^® 2000 (Invitrogen; Thermo Fisher Scientific, Inc.) was added to 150 µl serum-free opti-MEM (cat. no. 31985070; Gibco; Thermo Fisher Scientific, Inc.) and mixed with 10 µl miR-21 mimics/inhibitors, then maintained at room temperature for 20 min. The final concentration of miR-21 mimics and miR-21 mimics NC was 50 nM, and the final concentration of the miR-21 inhibitor and miR-21 inhibitor NC was 100 nM. After 20 min at room temperature, the mixed solution was added to the 6-well plates to complete the transfection and incubated for 6 h at 37°C in a 5% CO₂ incubator. Then, the medium was replaced with RPMI-1640 containing 10% heat-inactivated fetal calf serum and 1% penicillin-streptomycin. The cells were then incubated for 48 h at 37°C in a 5% CO₂ incubator.

In addition, THP-1-derived macrophages (3×10⁵ cells/well) were cultured with 5 µM rapamycin (cat. no. 53123-88-9; MedChemExpress), which is an autophagy inducer. After incubation for 1 h at 37°C in a 5% CO₂ incubator, THP-1-derived macrophages were transfected with miR-21 mimics or miR-21 mimics NC using Lipofectamine 2000 for 6 h at 37°C in a 5% CO₂ incubator. Then, the medium was replaced with RPMI-1640 containing 10% heat-inactivated fetal calf serum and 1% penicillin-streptomycin and incubated for 48 h at 37°C in a 5% CO₂ incubator.

Furthermore, 1 µg KLF6 overexpression plasmid (cat. no. HG12365-ACG; Sino Biological, Inc.), with the pCMV3 backbone, 1 µg pCMV3-C-GFPSpark control vector (cat. no. CV026; Sino Biological, Inc.), 50 nM miR-21 mimics NC and 50 nM miR-21 mimics were transfected into THP-1-derived macrophages using Lipofectamine 2000 for 6 h at 37°C in a 5% CO₂ incubator. Then, the medium was replaced with RPMI-1640 containing 10% heat-inactivated fetal calf serum and 1% penicillin-streptomycin and incubated for 48 h at 37°C in a 5% CO₂ incubator.

Target genes of miR-21 were evaluated using TargetScan 8.0 software (https://www.targetscan.org/vert_80/).

For the luciferase assay, the KLF6-expression construct was generated using a full-length of clone KLF6, which was inserted into psiCHECK™-2 Vector (cat. no. C8021; Promega Corporation). The miR-21-5p binding site found in the 3′-UTR of the KLF6 mRNA using TargetScan 8.0 software, and its mutated version, were cloned into a modified pGL3-control vector (cat. no. E1741; Promega Corporation). Subsequently, 293T cells (cat. no. CL-0005; Procell Life Science & Technology Co., Ltd.) were cotransfected with reporter constructs containing wild-type (WT) or mutant (MUT) KLF6 3′-UTR and miR-21-5p mimics (50 nM) or miR-21 mimics NC (50 nM) using Lipofectamine 2000. Then, 293T cells were harvested and lysed 48 h after transfection. Luciferase reporter activities were evaluated using the Dual-Glo Luciferase Assay System (Promega Corporation). Firefly luciferase activity was standardized to Renilla luciferase activity.

After cotransfection of THP-1-derived macrophages with miR-21 mimics or miR-21 mimics NC and 1.5 µg pCMV-GFP-LC3B (cat. no. D2815; Beyotime Institute of Biotechnology) using Lipofectamine 2000 for 6 h at 37°C in a 5% CO₂ incubator. Then, the medium was replaced with RPMI-1640 containing 10% heat-inactivated fetal calf serum and 1% penicillin-streptomycin and incubated for 24 h at 37°C in a 5% CO₂ incubator. Nuclei were stained with DAPI for 20 min at 4°C in the dark. Images of GFP-LC3 puncta were captured under a confocal fluorescence microscope (Leica Microsystems, Inc.).

All data were assessed using GraphPad 5.0 (Dotmatics) and SPSS 19.0 (IBM Corp.). Results are presented as the mean ± standard deviation. Comparisons between two groups were carried out via paired Student's t-test or unpaired Student's t-tests. Multiple comparisons were carried out via homogeneity test of variance followed by one-way ANOVA and Tukey's post hoc test. Two-tailed P≤0.05 was considered to indicate a statistically significant difference. Each in vitro experiment was repeated three times.

Serum miR-21 levels, detected by qPCR, were compared between the HC and ALF groups (Fig. 1A), and the results suggested that the serum levels of miR-21 were significantly higher in the ALF group than those in the HC group. Moreover, serum miR-21 levels were highly upregulated in 13 non-surviving patients with ALF compared with those in the 27 surviving patients with ALF (Fig. 1B). To explore the role of miR-21 in patients with ALF, serum miR-21 levels were evaluated before and after treatment in 27 surviving and 13 non-surviving patients with ALF before and after treatment (Fig. 1C and D). The results demonstrated that serum miR-21 levels were markedly decreased in the surviving group of patients with ALF after treatment. However, there was no statistically significant difference in serum miR-21 levels before and after treatment in non-surviving patients with ALF.

To further explore the role of miR-21 in the pathogenesis of ALF, the effects of miR-21 antagomir were evaluated in an LPS/GalN-induced male mouse model of ALF, which has a more stable physical condition with fewer individual differences compared with a female mouse model of ALF, similar to the liver failure model we have previously constructed (3). Briefly, following treatments, serum was isolated and liver tissues were collected. The expression levels of miR-21 in the liver tissues were evaluated by qPCR (Fig. 2A). The results demonstrated that the expression levels of miR-21 were significantly increased in the LPS/GalN-induced ALF mouse model compared with in the PBS control group. However, the injection of antagomir-21 blocked the upregulation of miR-21, demonstrating that a mouse model of LPS/GalN-induced ALF with miR-21 inhibition was successfully constructed. To explore the role of miR-21 in the LPS/GalN-induced ALF mouse model, the present study further explored liver function and pathological damage to the liver. As shown in Fig. 2B, ALT and AST levels in the LPS/GalN group were significantly increased, whereas they were decreased in the antagomir-21 ALF mouse group. Similarly, the degree of liver damage, including disordered structure of liver lobules, increased infiltration of inflammatory cells (blue arrow), hyperchromatic nuclei, hypochromatosis and appearance of necrosis (red arrow), in the LPS/GalN-induced ALF mouse model was markedly aggravated, whereas the degree of the aforementioned liver damage was relieved in the antagomir-21 LPS/GalN mouse group (Fig. 2C). These results suggested that LPS/GalN-induced ALF is attenuated by antagomir-21.

To further evaluate the potential mechanisms of miR-21, KLF6 (GenBank accession number: NM_001300.6) was considered a target of miR-21, with one putative miR-21-5p seed match site determined using the TargetScan database (Fig. 2D). Therefore, the expression levels of KLF6 were assessed, and the results demonstrated that the protein expression levels of KLF6 were markedly decreased in the LPS/GalN group compared with those in the PBS control group. By contrast, the decreased expression of KLF6 protein in the LPS/GalN-induced ALF mouse model was abrogated in the antagomir-21 mouse group (Fig. 2E).

Furthermore, the levels of IL-23 were detected in the antagomir-21 mouse model. The results demonstrated that serum IL-23 was significantly suppressed in the antagomir-21 ALF mouse group compared with that in the miRNA antagomir NC ALF mouse group, as detected by ELISA (Fig. 2F). As shown in Fig. 2G, the protein expression levels of IL-23 in the liver tissues exhibited the same trend, as determined by western blotting. These results indicated that miR-21 may promote the expression of IL-23.

Moreover, the levels of autophagy markers were explored in the antagomir-21 mouse model. The results demonstrated that the LC3II/I ratio was upregulated and the protein expression levels of p62 were reduced in the antagomir-21 ALF mouse group compared with those in the miRNA NC ALF mouse group (Fig. 2H). These results demonstrated that miR-21 could inhibit autophagy.

To explore the underlying mechanisms of miR-21, the miR-21 inhibitor, miRNA inhibitor NC, miR-21 mimics and miR-21 mimics NC were used to interfere with the expression levels of miR-21 in THP-1-derived macrophages, which was confirmed by qPCR. As shown in Fig. 3A, the transfection efficiency of the miR-21 inhibitor and miR-21 mimics were confirmed by qPCR. The relative expression levels of miR-21 were significantly upregulated in the miR-21 mimics group compared with those in the miR-21 mimics NC group, and the relative expression levels of miR-21 were significantly decreased in the miR-21 inhibitor group compared with those in the miRNA inhibitor NC group. Therefore, the miR-21 mimics and miR-21 inhibitor were used for subsequent overexpression and knockdown experiments.

Our previous study suggested that IL-23 was highly increased in patients with ALF (3). To explore whether a relationship exists between miR-21 and IL-23 in THP-1-derived macrophages, the expression levels of IL-23 were explored in the miR-21 overexpression and knockdown groups. The results demonstrated that IL-23 expression was significantly increased in the miR-21 mimics group compared with that in the miR-21 mimics NC group, whereas IL-23 expression was significantly suppressed in the miR-21 inhibitor group compared with that in the miRNA inhibitor NC group, as detected by western blotting (Fig. 3B). As shown in Fig. 3C, the levels of IL-23 in the cell supernatants exhibited the same trend, as determined by ELISA. These results indicated that miR-21 promoted the expression of IL-23.

KLF6 was identified as a target gene of miR-21, with one putative miR-21-5p seed match site (KLF6 3′-UTR 724–730), using the TargetScan database. To explore whether miR-21-5p directly targets the KLF6 mRNA 3′-UTR (724–730 bp), reporter constructs (pGL3-KLF6) with putative WT or MUT 3′-UTR downstream of the firefly luciferase were cloned. 293T cells were cotransfected with pGL3-KLF6 and miR-21-5p mimics or miR-21 mimics NC. The results demonstrated that transfection with the miR-21-5p mimics resulted in a significant downregulation of the luciferase activity of the reporter plasmid containing the WT 3′-UTR; however, the downregulation of luciferase activity was fully abolished after point mutation in the miR-21-5p binding site in the 3′-UTR of KLF6 (3′-UTR MUT) (Fig. 3D). These results indicated a direct interaction between miR-21-5p and the 3′-UTR of KLF6 mRNA.

To further confirm the bioinformatics-based predictions, THP-1-derived macrophages were transfected with miR-21 inhibitor or miR-21 mimics, and the mRNA and protein expression levels of KLF6 were evaluated by qPCR and western blotting, respectively. The relative expression levels of KLF6 were downregulated in the miR-21 mimics group compared with those in the miR-21 mimics NC group. However, introduction of the miR-21-5p inhibitor resulted in upregulation of the mRNA expression levels of KLF6 compared with those in the miRNA inhibitor NC group (Fig. 3E). As shown in Fig. 3F, KLF6 protein expression exhibited the same trend, as determined by western blotting. These results indicated that miR-21 inhibited the expression of KLF6.

It has been reported that 5-methoxytryptophan effectively alleviates liver cirrhosis by regulating FOXO3a/miR-21/ATG5 signaling pathway-mediated autophagy (21); therefore, it was hypothesized that miR-21 promoted the expression of IL-23 by impairing autophagy in patients with ALF. To experimentally verify this hypothesis, THP-1-derived macrophages were transfected with miR-21 mimics or miR-21 mimics NC. As expected, LC3I to LC3II conversion was highly upregulated and p62 protein expression was decreased in the miR-21 mimics NC group compared with those in the miR-21 mimics group, as determined by western blotting (Fig. 4A), which demonstrated that miR-21 inhibited autophagy. Consistently, overexpression of miR-21 highly suppressed GFP-LC3 puncta formation (Fig. 4B). Moreover, THP-1-derived macrophages were cultured with 5 µM rapamycin, which is an autophagy inducer. After 1 h, THP-1-derived macrophages were transfected with the miR-21 mimics or miR-21 mimics NC. The results demonstrated that the miRNA mimics NC + rapamycin group exhibited increased LC3I to LC3II conversion, decreased p62 protein expression (Fig. 4C) and decreased IL-23 protein expression (Fig. 4D) compared with in the miRNA mimics NC group. In the miR-21 mimics group, LC3I to LC3II conversion was decreased, p62 protein expression was increased (Fig. 4C) and IL-23 protein expression was increased (Fig. 4D) compared with in the miRNA mimics NC group. Compared with in the miR-21 mimics group, the miR-21 mimics + rapamycin group exhibited increased LC3I to LC3II conversion, decreased p62 protein expression (Fig. 4C) and decreased IL-23 protein expression (Fig. 4D). Compared with in the miRNA mimics NC + rapamycin group, the miR-21 mimics + rapamycin group exhibited decreased LC3I to LC3II conversion, increased p62 protein expression (Fig. 4C) and increased IL-23 protein expression (Fig. 4D). Similarly, the cell supernatant levels of IL-23 exhibited the same trend, as determined by ELISA (Fig. 4E). In conclusion, these results demonstrated that miR-21 promoted the expression of IL-23 via inhibiting autophagy.

KLF6 promotes autophagy, and KLF6 blocks miR-21-induced increase in IL-23 levels. To explore the underlying mechanisms of KLF6, a KLF6 overexpression plasmid was used to induce the overexpression of KLF6, which was confirmed by western blotting. The KLF6 overexpression plasmid significantly increased the expression levels of KLF6 in THP-1-derived macrophages compared with the vector group (Fig. 5A), indicating that the KLF6 overexpression plasmid was successfully constructed. To explore whether the KLF6 overexpression plasmid inhibited the downregulation of KLF6 induced by miR-21, THP-1-derived macrophages were transfected with 50 nM miR-21 mimics or 50 nM miR-21 mimics NC, and 1 µg KLF6 overexpression plasmid or 1 µg KLF6 control plasmid. The results demonstrated that miR-21 mimics could significantly inhibit KLF6 protein expression compared with miRNA mimics NC group, and KLF6 overexpression could significantly inhibit the downregulation of KLF6 expression induced by miR-21 mimics compared with the miR-21 mimics group (Fig. 5B).

The present study also evaluated whether KLF6 blocked the miR-21-induced decrease in autophagy levels. Compared with in the miRNA mimics NC group, overexpression of KLF6 could significantly upregulate the LC3II/I ratio, and the protein expression levels of Beclin1 and ATG7, and reduce the protein expression levels of p62 (Fig. 5C). Further study suggested that overexpression of KLF6 could upregulate miR-21-induced LC3II/I ratio, and the protein expression levels of Beclin1 and ATG7, and downregulate the protein expression levels of p62, which suggested that overexpression of KLF6 could block miR-21-induced downregulation of autophagy (Fig. 5C).

Moreover, the present study explored whether KLF6 blocked the miR-21-induced increase in IL-23 expression. The results suggested that overexpression of KLF6 could significantly reduce miR-21 mimics-induced IL-23 expression in THP-1-derived macrophages compared with miR-21 mimics group (Fig. 5D). The levels of IL-23 exhibited the same trend in the supernatant, as determined by ELISA (Fig. 5E). These results indicated that KLF6 blocked miR-21 induced-increases in IL-23 expression. In conclusion, miR-21 promoted the expression of IL-23 by inhibiting KLF6, which regulates autophagy, thus promoting inflammatory storms, and subsequent severe hepatocellular necrosis and ALF (Fig. 6).