Introduction
Acute liver injury (ALI) is a pernicious clinical condition marked by rapid hepatocyte dysfunction and defects in patients with or without a history of liver disease.1 Hepatitis viruses, drugs, immunologic injury and other hepatotoxic factors cause significant hepatocyte death, ultimately inducing ALI or even acute liver failure (ALF).2 Increases in aspartate aminotransferase (AST) and alanine aminotransferase (ALT) are used to identify the likelihood of ALI.3,4 Currently, immunosuppressors, antiviral agents, bioartificial livers, and liver transplantation are available treatment options for ALI.5 For patients with ALF and acute-on-chronic liver failure, the only definitive option is liver transplantation when there is a poor prognosis.6 Of note, the prognosis is often made worse by ineffective treatment, high cost, risk of organ rejection, limited liver donors, and severe treatment-related adverse effects.5,7 Under the circumstances, it is imperative to develop novel treatment strategies to prevent ALI.
Long noncoding (lnc)RNAs are transcripts >200 nucleotides that are dysregulated in liver disease and are considered biological markers for the diagnosis, prognosis, and treatment.8,9 Abnormal expression of MALAT1 has been identified in rodent models and patients with acute kidney injury.10 Interestingly, downregulation of MALAT1 blocks hypoxia/reoxygenation-induced hepatocyte apoptosis and limits the release of lactic dehydrogenase (LDH).11 Knockdown of MALAT1 improves the outcome of lipopolysaccharide (LPS)-induced acute lung injury and suppresses apoptosis of human pulmonary microvascular endothelial cells.12 Moreover, MALAT1 recruits the histone methyltransferase EZH2 to the microRNA (miR)-22 promoter region, thus inhibiting the expression of miR-22.13 Furthermore, MALAT1 recruits EZH2 to the promoter region of ABI3BP to downregulate its expression and modulate H3K27 methylation in gallbladder cancer cells.14 Notably, regulation of methylation plays an essential role in the deterioration and management of ALI.15,16 In addition, GFER encodes augmenter of liver regeneration (ALR) a protein that specifically supports liver regeneration.17 Transient knockdown of GFER has been found to promote cell death and reduce cell proliferation in liver tissue.18 We hypothesized that GFER is a downstream gene regulated by MALAT1 in ALI. The study aim was to elucidate the role and potential mechanism of MALAT1 in LPS-induced ALI. To that end, animal and cellular models of ALI were established using LPS, and the severity of ALI, apoptosis and proliferation were assessed.
Methods
Clinical sample collection
This study included 26 patients with acute drug-induced liver injury (ADILI) who were hospitalized in the liver disease department of The Third Xiangya Hospital of Central South University and 19 healthy people from the physical examination clinic of The Third Xiangya Hospital of Central South University. The diagnostic criteria of ADILI were based on the guidelines for the diagnosis and treatment of drug-induced liver injury of the 17th National Conference on Viral Hepatitis and Hepatology of the Chinese Medical Association in 2015 and a Roussel Uclaf Causality Assessment Method (RUCAM) scale score of >8 points.19 Patients were excluded if they had viral hepatitis B or C or other types of viral hepatitis, autoimmune liver disease, alcoholic liver disease, nonalcoholic fatty liver disease, cholestatic disease, or inherited metabolic liver disease. Peripheral blood was selected for clinical research in this study because liver biopsies were not performed. Approximately 4 mL of peripheral blood was collected in the morning after overnight fasting, placed in a serum tube, centrifuged at 4,000 rpm at 4°C for 10 m, and stored at −80°C until. use. This study complied with the Declaration of Helsinki and was approved by the ethics committee of The Third Xiangya Hospital of Central South University. All patients signed an informed consent form (No. 22014). Patient information is shown in Table 1.
Table 1GAPDH, MALAT1, and GFER primers
| Name of primer | Sequences |
|---|
| GAPDH-F | AATGGGCAGCCGTTAGGAAA |
| GAPDH-R | GCGCCCAATACGACCAAATC |
| MALAT1-F | GCTCTGTGGTGTGGGATTGA |
| MALAT1-R | GTGGCAAAATGGCGGACTTT |
| GFER-F | AAGCCTGACTTCGACTGCTC |
| GFER-R | CAAACCTAAGAGGGGCAGGG |
Hepatocyte culture and treatment
HL7702 human hepatocyte cells were provided by Cell Bank of Chinese Academy of Science and maintained in Dulbecco’s modified Eagle’s medium (Gibco, Grand Island, NY, USA) containing 10% fetal bovine serum plus 1% streptomycin-penicillin and cultured in a 37°C incubator with 5% CO2. Hepatocytes were induced by 1 µg/mL LPS (Sigma-Aldrich, St Louis, MO, USA) for 16 h in a 5% CO2 atmosphere at 37°C to induce ALI. In some cases, hepatocytes were cultured with an adenosine monophosphate-activated protein kinase (AMPK) inhibitor (10 µM, Compound C; Sigma-Aldrich) for 1 h before LPS induction.
HL7702 cells were seeded in culture plates and transfected with MALAT1 overexpression vector (oe-MALAT1), MALAT1 knockdown vector (sh-MALAT1, 20 µL, viral titer 5×108 TU/mL), EZH2 overexpression vector (oe-EZH2, 20 µL, viral titer 5×108 TU/mL), EZH2 knockdown vector (sh-EZH2, 20 µL, viral titer of 5×108 TU/mL), GFER overexpression vector (oe-GFER, 20 µL, viral titer of 3×108 TU/mL), GFER knockdown short hairpin vector (sh-GFER, 20 µL, viral titer of 5×108 TU/mL) or the corresponding negative controls (oe-NC and sh-NC) (20 µL, 3×108 TU/mL). Lentivirus vectors used for gene overexpression (LV5-GFP) or knockdown (pSIH1-H1-copGFP) were provided by GenePharma (Shanghai, China). Each experiment was conducted in triplicate. LPS induction was performed in HL7702 cells 24 h after transfection.
Animals
The animal procedures in this study were approved by the Ethics Committee of The Third Xiangya Hospital of Central South University (No. 22014) and followed the guidelines of the National Institutes of Health. Forty-two Sprague-Dawley rats 7–8 weeks of age, and 200–250 g (Shanghai Laboratory, Animal Research Center, Chinese Academy of Science) were fed under pathogen-free conditions and kept in a 12 h light/dark cycle and 60–65% humidity. The rats were allowed free access to food and water.
The rat ALI model was induced by intraperitoneal injection of 10 mg/kg LPS (Sigma-Aldrich), with normal saline treatment as the control. The model was maintained for 6 h before sh-MALAT1 (5×108 TU/mL, 300 µL/rat) or oe-GFER (3×108 TU/mL, 300 µL/rat) was introduced into rats through intravenous injection in tails for MALAT1 knockdown or GFER overexpression, with sh-NC (3×108 TU/mL, 300 µL/rat) or oe-NC (3×108 TU/mL, 300 µL/rat) as the negative control. The 42 rats were randomly divided into seven groups of six rats each, including normal controls, and saline (administration of an equal volume of normal saline), LPS (intraperitoneal injection of 10 mg/kg LPS for 6 h modelling), LPS + sh-NC group (intravenous injection of sh-NC in the tail vein 42 h before LPS induction, followed by intraperitoneal injection of 10 mg/kg LPS for 6 h modelling), LPS + sh-MALAT1 (injection of sh-MALAT1 in the tail vein 42 h before LPS induction, followed by intraperitoneal injection of 10 mg/kg LPS for a 6-h modelling), LPS + oe-NC group (injection of oe-NC in the tail vein 42 h before LPS induction, followed by intraperitoneal injection of 10 mg/kg LPS for a 6-h modelling), and LPS + oe-GFER (injection of oe-GFER in the tail vein 42 h before LPS induction, followed by intraperitoneal injection of 10 mg/kg LPS for a 6-h modelling). All animals were sacrificed to collect serum and liver tissue 48 h after modelling for 6 h and injection of lentivirus vectors.
Quantitative real-time polymerase chain reaction (qRT-PCR)
RNA extraction of cells or tissues was carried out using TRIzol (Invitrogen, Carlsbad, CA, USA) followed by detection of RNA concentration and purity. Qualified RNA samples were adjusted to an appropriate concentration and then reverse-transcribed using random primers and a reverse transcription kit (TaKaRa, Tokyo, Japan) following the manufacturer’s instructions. Gene expression was quantified by fluorescence qRT-PCR (LightCycler 480; Roche, Indianapolis, IN, USA), and was carried out following the manufacturer’s instructions (SYBR Green Mix; Roche Diagnostics). In brief, cDNA templates were predenatured at 95°C for 5 m, denatured at 95°C for 10 s, annealed at 60°C for 10 s and finally extended at 72°C for 20 s, followed by 40 cycles of cycling. Each qPCR assay was performed with three replicates. The relative expression of GFER and MALAT1 was determined by the 2−ΔΔCt method with GAPDH as the reference gene. The primer sequences for GAPDH, MALAT1, and GFER are shown in Table 2.
Table 2Patient characteristics
| Normal (n=19) | ADILI (n=26) |
|---|
| Age (years) | 52.56±1.85 | 56.46±1.42 |
| Sex (male/female) | 11/15 | 8/11 |
| ALT (U/L) | 27.45±3.74 | 410.53±114.02*** |
| AST (U/L) | 27.12±3.51 | 316.74±84.85*** |
| γ-GT (U/L) | 24.22±4.56 | 285.84±76.43*** |
| ALP (U/L) | 85.16±12.35 | 153.21±21.85* |
| TBil (µmol/L) | 13.51±2.13 | 35.51±2.13* |
| INR | 0.95±0.11 | 1.15±0.23 |
| RUCAM score | 0.67±0.96 | 9.27±1.84* |
Western blot assays
Cells or tissues were lysed in RIPA lysis buffer and centrifuged for protein extraction. Protein concentration was detected with a bicinchoninic acid assay kit (Beyotime Biotechnology, Shanghai, China) to ensure equal loading volume of protein. The corresponding volume of protein was homogenized with loading buffer (Beyotime) and then denatured for 3 m in a boiling water bath. Proteins were separated on 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) gels following the kit (Beyotime) manufacturer’s instructions. Briefly, the protein was electrophoresed at 80 V, and then the voltage was switched to 120 V for 1–2 h. After protein separation, membrane transfer was performed in an ice bath at 300 mA for 60 m. Then, the membranes were rinsed in a washing solution for 1–2 m, followed by blocking at room temperature for 60 m or at 4°C overnight. Incubation with primary antibodies against rabbit anti-human GAPDH (1:1,000; Cell Signaling, Boston, MA, USA), GFER (1:500; Santa Cruz Biotechnology, Dallas, TX, USA), p-mTOR (Ser2448 1:1,000; Cell Signaling Technology), mTOR (1:1,000; Cell Signaling Technology), p-AMPK (Thr172 1:1,000; Cell Signaling Technology), AMPK (1:1,000; Cell Signaling Technology), H3K27me3 (1:1,000; Abcam, Cambridge, UK), or EZH2 (1:500; Abcam) performed at room temperature on a shaking table for 1 h. After incubation, the membranes were washed three times for 10 m each and then incubated with horseradish peroxidase-labeled goat anti-rabbit IgG (1:5,000; Beijing ComWin Biotech Co., Ltd., Beijing, China) for 1 h at room temperature. Before color development, the membranes were washed three times for 10 m each. A chemiluminescence imaging system (Bio-Rad, Hercules, CA, USA) was used to visualize the membranes.
Assay of AST, ALT and LDH
The culture supernatant of HL7702 cells or rat serum was collected for liver function testing. AST, ALT, and LDH were assayed with kits following the manufacturer’s (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany) instructions.
Cell counting kit (CCK)-8 assay
Twenty-four hours after transfection, 100 µL of the cell suspension was seeded into a 96-well plate, with three replicates for each sample. Cells were cultured in an incubator for 24, 48, or 72 h, 10 µL of CCK-8 reagent (Dojindo, Tokyo, Japan) for 3 h, and absorbance was determined at 450 nm.
Ethynyl-deoxyuridine (EdU) staining
Cells (1×105 cells/well) were transferred to 96-well plates and cultured for 2 h with 100 µL EdU stain (5 µM; Sigma-Aldrich, Merck KGaA). The cells were then immobilized in 50 µL fixation buffer for 30 m. After removing the buffer, the cells were incubated with 50 µL glycine (2 mg/mL) for 5 m and washed with 100 µL phosphate buffered saline (PBS). After culture with 100 µL of permeabilization buffer and washing in 100 µL PBS, the cells were incubated with 100 µL of 1× Apollo solution in the dark for 30 m. Finally, the cells were cultured with 100 µL diamidino-phenylindole at room temperature for 5 m away from light and washed in 100 µL PBS, followed by observation by fluorescence microscopy (Olympus, Tokyo, Japan).
Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay
Collected cells were fixed in 4% paraformaldehyde for 30 m and then in 70% cold alcohol for 15 m. The cells were then incubated with PBS containing 0.3% Triton X-100 at room temperature for 5 m and incubated with TUNEL solution (Beyotime) for 60 m at 37°C. After washing in PBS three times, the cells were blocked with an antifade reagent and observed by fluorescence microscopy. The cell nuclei were stained with diamidino-phenylindole and apoptosis (%)=(TUNEL-positive cells/total cells)×100.
Rats were sacrificed to collect liver tissues, which were fixed in 10% neutral buffer formalin (Beijing Solarbio Science & Technology Co., Ltd., Beijing, China) for 24 h, dehydrated, and embedded in paraffin. After being sliced into 4 µm serial sections, the tissue was dewaxed with xylene and dehydrated in an alcohol gradient. A TUNEL detection kit (ZK-8005; ZSGB-Bio, Beijing, China) was used to determine the apoptosis rate in rat liver tissues. Five random fields in each section were evaluated by light microscopy (Olympus Optical Co. Ltd., Tokyo, Japan). Apoptotic cells were brown or brownish in color and with apoptotic cell morphology. Apoptosis was reported as the apoptosis index, and apoptosis (%)=(TUNEL-positive cells/total cells)×100.
Measurement of malondialdehyde (MDA), glutathione (GSH), superoxide dismutase (SOD) levels
Assay kits were used to determine MDA, SOD (Abcam), and GSH (Sigma-Aldrich) levels in cultured cells and liver tissues following manufacturer’s instructions.
RNA immunoprecipitation (RIP)
Magna RIP RNA-binding protein immunoprecipitation kits (Millipore Corp, Billerica, MA, USA) was used for the RIP assay. Briefly, HL7702 cells were lysed in 100 µL lysis buffer containing protease and RNase inhibitors, and then the protein lysate was incubated with rabbit anti-human EZH2 antibody (ab186006 1:500; Abcam) at 4°C for 30 m or anti-IgG antibody (ab109489, 1:100; Abcam) as the control. Subsequently, 10–50 µL of protein A/G beads were added and incubated with the cells at 4°C overnight. After incubation, the protein A/G-bead-antibody complexes were washed 3–4 times in 1 mL lysis buffer, and RNA was extracted and purified using the RNA extraction method. qRT-PCR was carried out with a MALAT1-specific primer to identify the interaction between EZH2 and MALAT1.
Chromatin immunoprecipitation (ChIP)
The ChIP assay was performed with SimpleChIP Plus sonication chromatin IP kits following the manufacturer’s (Cell Signaling Technology) instructions. HL7702 cells were fixed in 1% formalin to crosslink DNA and proteins. Then, the cells were lysed in lysis buffer and nuclear lysis buffer and ultrasonicated to generate 200–300 bp chromatin segments. The cell lysate was immunoprecipitated with protein A-beads conjugated with the corresponding antibodies, including anti-EZH2 antibody (ab228697; Abcam) and anti-histone 3 antibody (trimethyl-K27, ab6002; Abcam). Anti-IgG antibody (ab171870; Abcam) was added as a negative control. Protein-DNA crosslinking was reversed the RNA was purified, and enrichment of the DNA segment was detected by qRT-PCR.
Hematoxylin and eosin (H&E) staining
Rat liver tissues were collected, fixed in 10% neutral buffered formalin (Beijing Solarbio Science & Technology Co., Ltd.) for 24 h, dehydrated, embedded in paraffin, cut into 4 µm serial sections, and stained with H&E (Beijing Solarbio Science & Technology Co., Ltd.). Tissue histology was evaluated by optical microscopy (Olympus Optical, Tokyo, Japan).
Immunohistochemistry
Collected liver tissues were fixed in 4% paraformaldehyde for 48 h, embedded in paraffin, and sectioned at 4 µm. The sections were baked for 20 m, dewaxed in xylene, and washed once in distilled water. After washing three times in PBS, the sections were placed in 3% H2O2 for 10 m and subjected to antigen retrieval. After washing with PBS three times, the sections were blocked in goat serum at room temperature for 20 m. Excess serum was discarded, and the primary antibody against Ki-67 (ab16667, 1:200; Abcam) was added to the sections for incubation (4°C, overnight). Afterwards, the sections were incubated with secondary antibody at room temperature for 1 h and washed three times in PBS. Color development was sustained for 1–3 m using diaminobenzidine solution, and H&E was then used for 3 m for nuclear staining. The sections were then dehydrated, permeabilized, and cover slipped for observation. The percentage of positive cells in five randomly selected fields was reported (positive cells (%)=(positive cells/total cells)×100.
Statistical analysis
GraphPad 7.0 software was used for the statistical analysis, and data were reported as means ± SD. The significance of between-group differences was determined by t-tests. Multiple comparisons was carried out with one-way analysis of variance followed by Tukey’s multiple comparisons test. P-values <0.05 were considered statistically significant.
Results
MALAT1 was upregulated and GFER downregulated in ALI
To explore the expression of MALAT1 and GFER in patients with ALI, we included 26 patients with ADILI and 19 healthy individuals. There were no significant differences in sex or age between the two groups. Compared with the normal group, the serum levels of ALT, AST, γ-GT, alkaline phosphatase (ALP), and TBil, and the RUCAM scores of patients in the ADILI group increased significantly (Table 2; p<0.05). The results of qRT-PCR and western blotting showed that, compared with the normal group, the expression of MALAT1 in the serum of patients in the ADILI group increased, and the expression of GFER decreased (Fig. 1A, B; p<0.05). The results indicated that MALAT1 was highly expressed but GFER was weakly expressed in the serum of patients with ALI.
MALAT1 was upregulated and GFER downregulated in LPS-induced HL7702 cells
Human hepatocytes HL7702 were treated with LPS to induce an ALI cellular model, and ALT, AST, and LDH levels were assayed in the culture supernatant. LPS induction increased ALT, AST, and LDH levels (Fig. 2A, C; p<0.05). The CCK-8 assay found that proliferation was decreased in the LPS group compared with the control group (Fig. 2D; p<0.05), which was confirmed by EdU staining (Fig. 2E; p<0.05). TUNEL staining found that the cell apoptosis rate was increased in the LPS group compared with the control group (Fig. 2F, G; p<0.05). Assays of MDA, SOD, and GSH in cells found that after LPS treatment, MDA increased significantly and SOD and GSH decreased significantly (Fig. 2H, J; p<0.01). Moreover, qRT-PCR and western blot assays found that LPS increased MALAT1 expression (Fig. 2K; p<0.001) and decreased GFER expression (Fig. 2L, M; p<0.05) in HL7702 cells compared with the control group. Overall, LPS induction enhanced MALAT1 expression and reduced GFER expression in HL7702 cells.
Downregulation of MALAT1 suppressed cell apoptosis and oxidative stress injury but induced cell proliferation through GFER
HL7702 cells were transfected with sh-MALAT1, oe-MALAT1, sh-GFER, or oe-GFER for 24 h, followed by treatment with LPS for 16 h. The transfection efficiency was validated by qRT-PCR and western blot analysis (Fig. 3A–D; p<0.05). In addition, MALAT1 overexpression inhibited GFER expression and MALAT1 knockdown enhanced GFER expression (Fig. 3B–D; p<0.05). However, overexpression or knockdown of GFER did not influence MALAT1 expression (Fig. 3A). Introduction of MALAT1 overexpression or GFER knockdown substantially increased the levels of ALT, AST and LDH, while MALAT1 knockdown or GFER overexpression had the opposite effects (Fig. 3E–G; p<0.05). Compared with the LPS + oe-MALAT1 group, the LPS + oe-MALAT1 + oe-GFER group had decreased ALT, AST and LDH levels (Fig. 3E–G; p<0.05). Moreover, MALAT1 overexpression or GFER knockdown inhibited HL7702 cell proliferation and induced cell apoptosis and oxidative stress injury, but MALAT1 knockdown or GFER overexpression increased the proliferation rate and decreased the apoptosis rate and oxidative stress injury (Fig. 3H–M; p<0.05). In the LPS + oe-MALAT1 + oe-GFER group, HL7702 cells possessed increased proliferative ability and decreased apoptosis rate and oxidative stress injury than those in the LPS + oe-MALAT1 group (Fig. 3H–M; p<0.05). The data indicate that downregulation of MALAT1 inhibited hepatocyte apoptosis and oxidative stress injury but promoted cell proliferation by regulating GFER.
MALAT1 recruited EZH2 to the GFER promoter region to inhibit GFER expression
GFER significantly reduced ALT and AST in a mouse ALI model, and alleviated liver injury caused by ischemia-reperfusion.20,21 The University of California Santa Cruz (UCSU) genome browser (http://genome-asia.ucsc.edu/ ) predicted the presence of H3K27me3 methylation peak in the GFER promoter region (Fig. 4A). Inhibition of MALAT1 reduces liver ischemia-reperfusion injury,11 and MALAT1 changes the progression of liver fibrosis by regulating SIRT1.22 A previous study reported that MALAT1 recruited histone methyltransferase EZH2 to the pri-miR-22 promoter region to inhibit miR-22 expression.13 In this study, MALAT1 expression was negatively associated with GFER expression in the cellular ALI model, and the regulation of HL7702 cell proliferation and apoptosis by MALAT1 was involved in GFER. We hypothesized that MALAT1 recruited EZH2 to the GFER promoter region to suppress GFER expression. As expected, the RIP assay identified the interaction between MALAT1 and EZH2 (Fig. 4B; p<0.001). A ChIP assay was performed to confirm whether EZH2 regulated GFER expression via H2K27me3 methylation. EZH2 and H3K27me3 were found to be more abundant in the GFER promoter region in the oe-MALAT1 group than in the oe-NC group (p<0.01); enrichment of EZH2 and H3K27me3 in the GFER promoter region was reduced in the sh-MALAT1 group when compared with the sh-NC group (Fig. 4C; p<0.05). HL7702 cells were transfected with oe-EZH2 or sh-EZH2, and western blots of EZH2 expression demonstrated that transfection of oe-EZH2 promoted EZH2 and H3K27me3 expression and reduced GFER expression. sh-EZH2 treatment resulted in contrary findings (Fig. 4D; p<0.05). Overall, the results show that MALAT1 inhibited GFER expression by recruiting EZH2 to the GFER promoter region and promoting H3K27me3 methylation.
MALAT1/EZH2/GFER activated the AMPK/mTOR signaling pathway
HL7702 cells were transfected with sh-MALAT1, oe-MALAT1, sh-EZH2, oe-EZH2, sh-GFER and oe-GFER for 24 h, followed by treatment with LPS for 16 h. The levels of phosphorylated proteins active in the AMPK/mTOR signaling pathway were assayed by western blotting. MALAT1/EZH2 overexpression or GFER knockdown significantly upregulated phosphorylated AMPK levels and downregulated phosphorylated mTOR levels in HL7702 cells;. MALAT1/EZH2 knockdown or GFER overexpression had the opposite effects (Fig. 5A–C; p<0.05).
After HL7702 cells were transfected with oe-MALAT1 lentiviral vector and its control (oe-NC) for 24 h, they were treated with the AMPK inhibitor Compound C (CC) for 1 h, induced by LPS for 16 h, and then collected for subsequent assays. Western blot assays demonstrated that compared with the LPS + oe-NC group, phosphorylated AMPK level was decreased and phosphorylated mTOR expression was increased in the LPS + oe-NC + CC group. HL7702 cells in the LPS + oe-MALAT1 group had the reverse responses; phosphorylated AMPK level was reduced and that of mTOR was increased in the LPS + oe-MALAT1 + CC group compared with the LPS + oe-MALAT1 group (Fig. 5D–F; p<0.01).
In the LPS + oe-NC + CC group, the levels of ALT, AST and LDH were decreased (Fig. 5G–I; p<0.05), HL7702 cell proliferation was increased (Fig. 5J, K; p<0.05), and the apoptosis rate and oxidative stress injury were inhibited (Fig. 5L–O; p<0.05) compared with those in the LPS + oe-NC group. HL7702 cells in the LPS + oe-MALAT1 group had increased ALT, AST, and LDH levels (Fig. 5G–I; p<0.05) concurrent with decreased cell proliferation (Fig. 5J, K; p<0.05) and enhanced apoptosis rate and oxidative stress injury (Fig. 5L–O; p<0.05), compared with those in the LPS + oe-NC group. In the LPS + oe-MALAT1 + CC group, the levels of ALT, AST and LDH in HL7702 cell supernatant were suppressed (Fig. 5G–I; p<0.05) along with increased proliferation (Fig. 5J, K; p<0.05) and reduced apoptosis rate and oxidative stress injury (Fig. 5L–O; p<0.05). The data indicate that MALAT1/EZH2/GFER activated the AMPK/mTOR signaling pathway.
Downregulation of MALAT1 restored GFER expression in rats with ALI
Rats were intravenously injected with sh-MALAT1 or oe-GFER in the tail vein for 42 h, followed by treatment of LPS to induce ALI. We demonstrated that the levels of ALT, AST and LDH in rat serum were increased in the LPS group (p<0.01, vs. the saline group), while those of the LPS + sh-MALAT1 group and the LPS + oe-GFER group were lower than those in the LPS + sh-NC group and the LPS + oe-NC group (Fig. 6A–C). As shown by H&E staining, rats in the normal group and the saline group had normal histology with clear hepatic lobules and regular hepatic sinusoids. In contrast, rats in the LPS group had significant liver injury, manifested as cellular edema, hematolysis, diffuse necrosis, and inflammatory cell infiltration (Fig. 6D). However, LPS-induced injury was alleviated in the liver tissues of rats in the LPS + sh-MALAT1 and LPS + oe-GFER groups (Fig. 6D).
qRT-PCR and western blotting revealed an increase in MALAT1 expression (Fig. 6E; p<0.01) and a decrease in GFER expression (Fig. 6F, G; p<0.05) in the LPS group compared with those in the saline group. In the LPS + sh-MALAT1 group, MALAT1 expression was significantly reduced (Fig. 6E; p<0.01) and GFER expression was upregulated (Fig. 6F, G, p<0.05) in liver tissues, compared with the LPS + sh-NC group. Moreover, increased GFER expression (p<0.001) and unchanged MALAT1 expression (p > 0.05) were found in the LPS + oe-GFER group compared with the LPS + oe-NC group (Fig. 6E–G). Taken together, downregulation of MALAT1 promoted GFER expression in ALI-model rats.
Downregulation of MALAT1 or overexpression of GFER inhibited ALI in vivo
Rats were injected with sh-MALAT1 or oe-GFER in the tail vein for 42 h, followed by LPS induction for modelling, and then immunohistochemistry was performed. Mice in the LPS group had fewer Ki-67-positive liver cells than the saline group, but MALAT1 knockdown or GFER overexpression increased the percentage of Ki-67-positive cells in liver tissues (Fig. 7A; p<0.05). In addition, TUNEL staining indicated increased apoptosis rate in rat liver tissue from the LPS group (p<0.01, vs. the saline group) and decreased hepatocyte apoptosis in the LPS + sh-MALAT1 or the LPS + oe-GFER group (both p<0.05) (Fig. 7B). We also found that MDA was increased and SOD and GSH were decreased in liver tissue from the LPS group relative to the saline group. In the LPS + sh-MALAT1 group or the LPS + oe-GFER group, MDA level was decreased and SOD and GSH levels were increased (Fig. 7C–E; p<0.01).
In addition, liver tissues in the LPS group were found to have upregulated levels of phosphorylated AMPK (Fig. 7F, G; p<0.01) and downregulated levels of phosphorylated mTOR (Fig. 7F. H; p<0.05). However, MALAT1 knockdown or GFER overexpression reduced phosphorylated AMPK and increased phosphorylated mTOR levels in liver tissue (Fig. 7F–H; p<0.05). Thus, these data proved the ameliorative effect of MALAT1 downregulation or GFER overexpression on ALI in vivo.
Discussion
ALI is a severe, acute disease with a high mortality. It is characterized by acute hepatocyte necrosis, and there have not been any treatment breakthroughs in the last few decades.23,24 Recently, various lncRNAs, including DINO, NEAT1, and XIST, and their downstream genes have been investigated to explain the progression and improve the effectiveness of ALI treatment.16,25,26 LPS, which was first found in the outer membrane of gram-negative bacteria, has been widely used in vivo and in vitro to mimic the pathology of ALI.27–29 In this study, we successfully established cellular and animal models of ALI using LPS, and MALAT1 was found to inhibit cell proliferation and promote apoptosis in LPS-induced rats and hepatocytes. MALAT1 has been found to limit proliferation and induce apoptosis of human renal tubular epithelial cells in the presence of LPS,30 and MALAT1 knockdown was found to block LPS-induced acute lung injury by inhibiting apoptosis and improving cell viability.31 The apoptosis-promoting effect of MALAT1 has been previously reported in hepatocytes.31 However, Li et al.32 reported that MALAT1 overexpression was required for accelerating hepatocyte proliferation and liver regeneration. Our in vivo and in vitro experiments showed that MALAT1 was upregulated after LPS exposure and identified as an ALI-promoting gene. GFER, also known as ALR, protects against chemical- or toxin-related ALI,33,34 ischemia reperfusion-induced liver and kidney injury.35,36 Deletion of ALR accelerates steatohepatitis and hepatocellular carcinoma.37 Therefore, we examined the expression of GFER in LPS-induced hepatocytes, which showed a decrease in GFER expression. Overexpression of GFER alleviated hepatocyte apoptosis, excessive hepatocyte proliferation, and improved liver function after LPS insult. Intriguingly, MALAT1 was negatively associated with, and significantly regulated, GFER expression at the cellular and animal levels. In addition, the ALI-promoting effect of MALAT1 was offset by GFER upregulation, indicated by decreases in serum AST/ALT, hepatocyte apoptosis, and enhanced proliferation of hepatocytes. The results indicated that MALAT1 regulated GFER expression to aggravate ALI. We then focused on elucidating the mechanism of MALAT1 regulation of GFER in ALI.
MALAT1 has oncogenic activity through EZH2-mediated suppression of miR-217 expression in lung carcinogenesis.38 In the presence of MALAT1, silencing of EZH2 has been shown to reduce apoptosis and enhance cardiac function.13 Increased MALAT1 recruits EZH2 to its downstream gene to promote H3K27me3 expression for specific transcription inhibition.39 EZH2 can modify the enrichment of its catalyzed H3K27me3 and contribute to the pathogenesis of liver failure by triggering the release of tumor necrosis factor (TNF) and other pro-inflammatory cytokines.40 Inhibition of EZH2 blunts H3K27me3 and restrains the activities of serum ALT and AST.41 We predicted an H3K27me3 methylation peak in the GFER peak, indicating that MALAT1 might regulate GFER expression through methylation. Our functional experiments identified an interaction between MALAT1 and EZH2, and further demonstrated that MALAT1 overexpression significantly enhanced the enrichment of EZH2 and H3K27me3 in the GFER promoter region. The study firstly demonstrated that MALAT1 suppressed GFER expression by recruiting EZH2 to the GFER promoter region, enhancing H3K27me3 methylation, and aggravating ALI.
AMPK regulates energy homeostasis and metabolism, and mTOR is an enzyme downstream of AMPK.42 Activation of the AMPK/mTOR signaling pathway has been shown to be involved in liver diseases, including nonalcoholic fatty liver disease and ALI.43–45 More important, Pu et al.46 found that deletion of ALR induced increased AMPK phosphorylation and decreased mTORC1 phosphorylation, and increased both autophagy flux and apoptosis. In this study, phosphorylated AMPK was increased, and phosphorylated mTOR was reduced in ALI model rats. At the same time, MALAT1 inhibition or GFER overexpression decreased the expression of p-AMPK and promoted p-mTOR expression. Inhibiting the phosphorylation of AMPK by Compound C inhibited ALT, AST, and LDH, and hepatocyte apoptosis, and improved hepatocyte proliferation. Knockdown of MALAT1 or overexpression of GFER had a similar effect to that of Compound C in ALI. The AMPK/mTOR pathway is a typical regulator of autophagy.47 Both endogenous and exogenous interleukin-37 protect against ischemia reperfusion-induced hepatic injury by restraining excessive autophagy and apoptosis by regulating the AMPK/mTOR signaling pathway.48 Therefore, much attention should be paid in future studies to whether MALAT1/EZH2/GFER participates in ALI by activating the AMPK/mTOR pathway to affect hepatocyte autophagy.
This study revealed that MALAT1 upregulation was associated with decreased proliferation, enhanced apoptosis, and aggravation of liver injury, which sheds light on the prevention and treatment of ALI. Notably, we showed that MALAT1 inhibited GFER by recruiting EZH2 to the GFER promoter region and enhancing H3K27me3 methylation, thus resulting in deterioration of ALI. Activation of the AMPK/mTOR signaling pathway was also linked to the ALI-promoting effect of MALAT1. The study thus characterized a possible regulatory mechanism of MALAT1 exacerbating ALI, which contributes to understanding ALI progression and provides a novel approach for ALI treatment. Further studies are required to explain the regulation and methylation of the molecules by MALAT1 in ALI to boost the application of their therapeutic benefits in clinical practice.