Abstract LncRNAs are involved in the pathophysiologic processes of multiple diseases, but little is known about their functions in hepatic ischemia/reperfusion injury (HIRI). As a novel lncRNA, the pathogenetic significance of hepatic nuclear factor 4 alpha, opposite strand (Hnf4αos) in hepatic I/R injury remains unclear. Here, differentially expressed Hnf4αos and Hnf4α antisense RNA 1 (Hnf4α-as1) were identified in liver tissues from mouse ischemia/reperfusion models and patients who underwent liver resection surgery. Hnf4αos deficiency in Hnf4αos-KO mice led to improved liver function, alleviated the inflammatory response and reduced cell death. Mechanistically, we found a regulatory role of Hnf4αos-KO in ROS metabolism through PGC1α upregulation. Hnf4αos also promoted the stability of Hnf4α mRNA through an RNA/RNA duplex, leading to the transcriptional activation of miR-23a and miR-23a depletion was required for PGC1α function in hepatoprotective effects on HIRI. Together, our findings reveal that Hnf4αos elevation in HIRI leads to severe liver damage via Hnf4αos/Hnf4α/miR-23a axis-mediated PGC1α inhibition. Keywords: Hnf4αos, Liver, Ischemia/reperfusion, PGC1α, Reactive oxygen species Graphical abstract [67]Image 1 [68]Open in a new tab Highlights * • High expression of Hnf4αos was associated with a severe damage in I/R mouse model. * • Hnf4αos exacerbates I/R injury by reducing ROS clearance. * • Hnf4αos promotes Hnf4α mRNA stability through a RNA-RNA duplex. * • Hnf4αos involved in I/R process via Hnf4αos/Hnf4α/miR-23a-mediated PGC1α inhibition. Abbreviations lncRNAs long noncoding RNAs Hnf4αos hepatic nuclear factor 4 alpha, opposite strand Hnf4α hepatocyte nuclear factor 4 alpha PGC1α PPARγ coactivator-1α ROS reactive oxygen species; I/R, ischemia/reperfusion A/R anoxia/reoxygenation ALT alanine aminotransferase AST aspartate aminotransferase ELISA enzyme-linked immunosorbent assay TUNEL terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling MDA malondialdehyde 4-HNE 4-hydroxynonenal SOD superoxide dismutase CAT catalase GPX glutathione peroxidase LDH lactate dehydrogenase 1. Introduction Hepatic ischemia/reperfusion injury (HIRI) is a common pathological process that occurs in several clinical scenarios, such as complex liver resection, liver transplantation, and hemorrhagic shock. During this process, the initial ischemic injury causes direct hepatocyte damage, and subsequent blood flow reflux further aggravates liver dysfunction and injury due to the propagation of reactive oxygen species (ROS), macrophage activation and inflammatory cytokines, which trigger cell death [[69]1,[70]2]. However, the underlying molecular mechanisms of ischemia/reperfusion (I/R) injury remain largely unknown. Long noncoding RNAs (lncRNAs) are defined as single-stranded RNA molecules spanning more than 200 nucleotides that are involved in multilevel gene expression regulation, including epigenetic modification, and transcriptional and posttranscriptional progression [[71]3]. According to the proximity to protein coding genes in the genome, lncRNAs are generally placed into five categories: sense, antisense, bidirectional, intronic, and intergenic lncRNAs [[72]4]. Currently, several studies have highlighted the significant roles of lncRNAs in the pathogenesis of liver disease. For instance, lncRNA HULC is upregulated in hepatocellular carcinoma and enhances hepatocarcinogenesis by promoting the phosphorylation of YB-1 via the ERK pathway [[73]5]; lncRNA ANRIL alleviates liver fibrosis and hepatic stellate cell (HSC) activation via the AMPK pathway [[74]6]; and lncRNA CCAT1 promotes nonalcoholic fatty liver disease (NAFLD) by increasing LXRα transcription [[75]7]. Nevertheless, in the case of hepatic I/R injury, little is known about lncRNAs in hepatic I/R injury. Thus, a deeper understanding of the molecular mechanisms underlying the pathogenic process of hepatic I/R is required to uncover potential lncRNA-targets for developing promising therapeutic strategies. Furthermore, we have identified a novel lncRNA hepatic nuclear factor 4 alpha, opposite strand (Hnf4αos), a natural antisense transcript (NAT) of hepatocyte nuclear factor 4 alpha (Hnf4α), which was aberrantly upregulated in mouse I/R models. Although Hnf4αos has been reported, little information is available for regarding its molecular function [[76]8,[77]9]. PPARγ coactivator 1 alpha (PGC1α) is well known as a metabolic regulator in the physiological process of oxidative phosphorylation (OXPHOS), the tricarboxylic acid (TCA) cycle and ROS metabolism [[78][10], [79][11], [80][12]]. Intriguingly, our previous studies have demonstrated that PGC1α is an important regulator of ROS metabolism that reduces cell death, ameliorates the sterile inflammatory response and alleviates oxidative stress-induced liver damage during hepatic I/R insult [[81]13]. Moreover, several lines of evidence, including data from bioinformatic analysis and determination of oxidative stress levels, suggest a close link between the lncRNA Hnf4αos and PGC1α. Thus, we further investigated the effects of Hnf4αos on I/R progression and the underlying mechanisms between Hnf4αos and PGC1α. 2. Material and methods 2.1. Human liver samples Human liver samples were obtained from subjects who underwent partial hepatectomy due to hepatic hemangioma. All procedures involving human samples were approved by the Ethics Committee of the First Affiliated Hospital of Harbin Medical University and patient informed consent was obtained. We listed the detailed clinical information of the hemangioma patients in [82]Supplementary Table S3. 2.2. Animals Male C57BL/6 mice, hepatocyte-specific Hnf4αos knockout (Hnf4αos-KO) mice and wild-type (WT) mice (8 weeks old) were housed in specific pathogen-free (SPF) conditions and raised following institutional guidelines for animal care. Hnf4αos-KO mice were obtained by CRISPR/Cas9 methods as described previously [[83]14]. Hnf4αos-KO mice were generated by crossing Hnf4αos-floxed mice with Albumin-Cre mice (Jackson Laboratory. Bar Harbor, ME, USA) on the C57BL background. The donor vector containing the fourth exon of the Hnf4αos gene was floxed by two loxP sites. All animal experiments were performed in accordance with the standard protocols of the Committee on the Use of Live Animals in Teaching and Research of Harbin Medical University, Harbin, China. 2.3. Mouse hepatic I/R injury model The procedures for partial hepatic ischemia have been described previously [[84]15]. Mice were housed in a specific pathogen-free and temperature-controlled environment with a 12-h light/dark cycle. Briefly, the mice were anesthetized with pentobarbital sodium (50 mg/kg), and a midline laparotomy was performed. An atraumatic clip was placed across the left lateral and median lobes of the liver (∼70%). After 75 min of partial hepatic ischemia, the clip was removed for initial reperfusion. Sham control mice underwent the same operation without vascular clamping. 2.4. Cell A/R treatment model Cellular anoxic conditions were established and maintained in a modular incubator chamber (Biospherix, Lacona, NY, USA) by continuous gas flow with a 1% O[2], 5% CO[2] and 94% N[2] gas mixture. After incubation under hypoxia for 6 h, the cells were incubated under normoxic conditions with 95% air and 5% CO[2] for the indicated times (0, 3, 6, 12, 24h). The medium and cells were collected for further analysis. 2.5. Cell culture and treatment Mouse hepatocytes were isolated by a modified in situ collagenase perfusion technique as previously described [[85]15]. Hepatocyte purity and viability typically exceeded 99 and 95%, respectively. Primary hepatocytes and L02 cell lines (Type Culture Collection of the Chinese Academy of Science) were cultured in DMEM supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin in a 5% CO[2]/water-saturated incubator at 37 °C. 2.6. Immunofluorescence assay Paraffin-embedded tissue sections were used for immunofluorescence as described previously [[86]16]. The liver sections were incubated with primary antibody against Ly6G (Cell Signaling Technology) (1:500) (31469), and the slides were incubated with corresponding fluorescence-labeled secondary antibody (ThermoFisher) (1; 1000) (A32744) for further staining. 2.7. ROS detection Cellular reactive oxygen species (ROS) levels were estimated as previously described [[87]17]. For intracellular ROS levels, cells were incubated in medium containing 10 μM dihydroethidium (DHE) (Invitrogen, USA) for 30 min at 37 °C in the dark. The medium was switched to fresh medium before fluorescence detection. The relative ROS levels, which are proportional to the fluorescence intensity, were quantified using Image-Pro Plus software. 2.8. Luciferase reporter assay We predicted potential Hnf4α binding sites on the PGC1α and miR-23a promoters using the JASPAR database, and the PGC1α 3’-untranslated region (UTR) contains conserved miR-23a binding sites as reported previously [[88]18]. We then cloned the candidate binding sites in an SV40 driven luciferase reporter plasmid. Briefly, luciferase activity was assessed using a luciferase assay kit (Promega, Madison, WI, USA). HEK-293T cells containing specific plasmids and 1 ng pRL-TK Renilla luciferase plasmid were seeded into 24-well plates. After 48 h, we used the dual luciferase reporter assay system (Promega) to measure luciferase activity according to the manufacturer's instructions. 2.9. Ribonuclease protection assay (RPA) A ribonuclease protection assay (RPA) and quantitative RT-PCR were performed to detect the RNA-RNA duplex. Total RNA from primary hepatocytes was isolated as described previously [[89]19]. The RNA samples were treated with DNAse Ⅰ (Sigma, 12.5 units/ml) and RNase A (QIAgen, 200 ng/ml) to remove residual DNA and single-stranded RNAs. Finally, the solutions were incubated for 40 min at 37 °C for further qRT-PCR. 2.10. Electrophoretic mobility shift assay (EMSA) An electrophoretic mobility shift assay (EMSA) was performed as described previously [[90]12]. The oligonucleotides used in EMSA were as follows: Hnf4α/miR-23a wt, 5’-GATCAGCTGGCCCCTGAAAACCTTGTTTAAC-3’ and 3’-CTAGTCGACCGGGGACTTTTGGAACAAATTG-5’. Hnf4α/miR-23a mut, 5’-GATCAGCTCCCCCCTAAAAAACTTGTTTAAC-3’ and 3’-CTAGTCGAGGGGGGATTTTTTGAACAAATTG-5’. 2.11. Statistical analysis All data are expressed as the mean ± SD. Significant differences between groups were determined by ANOVA, with Bonferroni correction for continuous variables and multiple groups. Two-tailed Student's t-test was used for comparison of a normally distributed continuous variable between 2 groups. The level of significance was set at a p value less than 0.05 for all analyses. Further details of the experimental materials and procedures are described in the Supplementary Files. 3. Results 3.1. LncRNA Hnf4αos is elevated during hepatic I/R injury Several lncRNAs were differentially expressed in the GEO data-set ([91]GSE15891) with exposure to chronic anoxia and our heatmap demonstrated the marked differentially expressed lncRNAs related to oxidative stress, inflammatory response and apoptosis pathways ([92]Fig. 1A). For examining the relationships of lncRNAs and traget genes, the top-ranked lncRNAs and mRNAs correlated oxidative stress/inflammatory response/apoptosis resident on different chromosomes ([93]Fig. 1B). Among the top-ranked differentially expressed lncRNAs, only Hnf4αos was enriched in adult mouse liver tissue ([94]Supplementary Table S1, 2). Thus, Hnf4αos was selected for further investigation during hepatic I/R injury. To explore the role of lncRNA Hnf4αos in HIRI, we first detected the expression levels of Hnf4αos in murine hepatic I/R and hepatocyte A/R models, and Hnf4αos was found to be increased after reperfusion. The human-derived lncRNA, Hnf4α-as1, was also found to be differentially expressed in clinical liver samples from patients who underwent partial hepatectomy ([95]Fig. 1C–D, [96]Supplementary Fig. S1). Furthermore, cellular fractionation of hepatocytes followed by qRT-PCR implied that Hnf4αos was predominantly expressed in the nuclei of hepatocytes rather than other compartments, compared with U6 (localized in the nucleus) and 18S (localized in the cytoplasm) expression ([97]Fig. 1E). Moreover, a fluorescence in situ hybridization (FISH) assay was performed to detect the locations of and changes in Hnf4αos in mouse hepatocytes after A/R treatment. The results showed that the fluorescence intensity of Hnf4αos was markedly enriched in hepatocyte nuclei and significantly elevated in the A/R group compared with the normoxic group ([98]Fig. 1F). Therefore, we identified Hnf4αos as a novel therapeutic target in the pathogenic process of hepatic I/R injury. Fig. 1. [99]Fig. 1 [100]Open in a new tab LncRNA Hnf4αos is elevated during hepatic I/R injury. (A) Heatmaps generated using the RNA expression of members detected by the DEG analysis. The expression of RNAs was visualized in color saturation; the expression level of genes was indicated by the colors (3 mice in the normoxia group and 5 mice in hypoxia group). (B) Genomic distance between lncRNAs and correlated with the oxidative stress, inflammatory response and apoptosis genes in KEGG. (The outer ring shows the distribution of the chromosomes of the mouse; The internal lines indicate that the top lncRNA-mRNA pairs) (C)Hnf4αos expression was assessed by qRT-PCR in mouse liver I/R models. (D)Hnf4αos expression was assessed by qRT-PCR in primary hepatocytes after A/R treatment. (E) Levels of cytoplasmic and nuclear Hnf4αos in primary hepatocytes. (F) The cellular locations and expression changes of Hnf4αos were analyzed by RNA-FISH. The scale bar represents 50 μm. n.s. P > 0.05, *P < 0.05, **P < 0.01, ***P < 0.001. (For interpretation of the references to