Differential expression of circular RNAs in hepatic tissue in a model of liver fibrosis and functional analysis of their target genes.

AIM: To measure the expression profile of circular RNA (circRNA) in hepatic tissues in a liver fibrosis model and to explore their function using molecular biology and bioinformatic techniques. METHODS: The classic CCl4 mouse liver fibrosis model was established alongside a normal control group. The circRNA expression profile of hepatic tissue from the two groups was compared using a high-throughput circRNA microarray. The differentially expressed circRNAs were identified, and real-time quantitative polymerase chain reaction (RT-qPCR) was used to verify a subset of the differentially expressed circRNAs (target genes). At the same time, the mouse oxidative stress injury, macrophage inflammation, and hepatic stellate cell activation models were established, and the expression of target circRNA in the above cells was measured by RT-qPCR. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) were used to predict the biological functions of target genes. Finally, one of the circRNAs was selected and its cellular function was verified using siRNA. RESULTS: A total of 10 389 circRNAs were analyzed by microarray. Compared with the normal group, there were 69 circRNAs that were differentially expressed in the liver fibrosis model group (>2-fold differential expression, P < 0.05), of which 14 were upregulated and 55 were downregulated. Five circRNAs and their differential expression were verified by RT-qPCR, and the findings were consistent with the microarray results. Of these, three circRNAs were differentially expressed (P < 0.05) in the JS1 model, one circRNA was differentially expressed (P < 0.05) in the AML12 model, and four circRNAs were differentially expressed (P < 0.05) in the RAW264.7 model. The GO analysis showed that the differentially expressed circRNAs might be involved in cell autophagy, composition of extracellular matrix components, synthesis and metabolism of retinoic acid, retinol dehydrogenase activity, ubiquitin-like protein ligase activity, histone methylation, and other biological functions. The KEGG analysis showed that the target genes of the differentially expressed circRNAs might be involved in transforming growth factor-ß1/smads, Hippo, Rap1, vascular endothelial growth factor, and other signaling pathways. Lipofection experiments showed that the expression of α-smooth muscle actin (α-SMA) in JS1 cells increased significantly after the expression of mmu_circ_34116 was decreased. CONCLUSION: The circRNA expression profile in liver fibrosis tissue shows significant changes. Partially differentially expressed circRNA could be involved in hepatic fibrosis related to hepatic oxidative stress injury, macrophage inflammation, and stellate cell activation. For instance, mmu_circ_34116 can significantly inhibit the activation of hepatic stellate cells.

Preliminary screening and functional analysis of circular RNAs associated with hepatic stellate cell activation.

OBJECTIVE: To screen for circular RNAs (circRNAs) that are associated with the activation of hepatic stellate cell (HSC) by monitoring changes in liver circRNA expression in a model of liver fibrosis. METHODS: The classic mouse model of CCl4-induced liver fibrosis was established and validated by histopathological examination. JS1 cells were activated by TGF-ß1 to model HSC activation in vitro. Differentially expressed circRNAs in the fibrotic liver tissues and JS1 cells were determined using circRNA microarray, and some of those circRNAs were verified by RT-qPCR. The target genes of the above circRNAs were then predicted by bioinformatics analysis and summarized into a "circRNA-miRNA-mRNA" network diagram. Constructed plasmid mmu_circ_34116 siRNA was transfected to JS1 cells by Lipo2000, then we detected the expression changes of α-SMA. RESULTS: A total of 10,389 circRNAs were identified by microarray screening, and 69 differentially expressed circRNAs were detected in the fibrotic liver tissues with >2-fold difference in expression level relative to normal liver tissues (P < 0.05); 14 circRNAs were up-regulated and 55 were down-regulated. Five differentially expressed circRNAs in fibrotic liver and JS1 cells were verified by RT-qPCR, while all five showed similar trends with the microarray results in the liver, only 3 circRNAs in the JS1 activation model were consistent with the microarray results while one showed no significant change and one circRNA was not detected. Bioinformatics analysis predicted that the "mmu_circ_34116/miR-22-3P/BMP7" signal axis might be involved in the activation of HSC. Transfection experiment confirmed that the expression of α-SMA is significantly elevated as a result of inhibitory expression of mmu_circ_34116. CONCLUSION: The circRNAs expression profile of liver tissue had changed in fibrosis mouse model, and some of these circRNAs may be associated with HSC activation. For instance, mmu_circ_34116 would inhibit HSC activation.

[Dynamic detection of duck hepatitis B virus cccDNA in serum of ducks with liver injury].

OBJECTIVE: To confirm whether DHBV cccDNA could be detected in serum of DHBV-infected ducks after liver injury. METHODS: Twenty four ducks with persistent DHBV infection were divided into 4 groups with the following treatments: A, D-galactosamine (D-GalN, 2.2 g/kg) and lipopolysaccharide (LPS, 100 µg/kg); B, 10 mg/kg of carbon tetrachloride (CCl4) every 12 h twice following D-GalN and LPS; C, 15 mg/kg of CCl4 every 12 h twice following D-GalN and LPS; D, normal saline as normal control (NC). At 0 h, 24 h, 36 h and 48 h post-treatment, sera were collected from each duck for determination of serum DHBV load, DHBV cccDNA and alanine aminotransferase (ALT). And ducks were eventually sacrificed to obtain liver specimens for pathological assessment of liver lesions and determination of intrahepatic total DHBV DNA and DHBV cccDNA. RESULTS: (1) No obvious pathological change was observed in the liver of ducks from NC group while the indices of liver injury were significantly different between Groups A, B and C; (2) DHBV cccDNA was undetectable in the sera of ducks from NC and A group at all time points. In contrast, DHBV cccDNA, varying from 3.17 × 10(3) copies/ml to 1.72 × 10(4) copies/ml, was detected in the sera of 2 ducks from Group B and 3 ducks from Group C at 36 h post-treatment. The occurrence of DHBV cccDNA in serum was significantly correlated with the degree of liver injury while no significant association with serum ALT level and DHBV load as well as with the level of intrahepatic total DHBV DNA and DHBV cccDNA was observed. CONCLUSION: DHBV cccDNA may be detected in the serum when the liver of duck is seriously damaged. The incidence of DHBV cccDNA occurrence in the serum is significantly associated with the severity of liver injury.

Effect of oxymatrine on the replication cycle of hepatitis B virus in vitro.

AIM: To determine the antiviral mechanism or target of oxymatrine against hepatitis B virus (HBV). METHODS: HepG2.2.15 cells were incubated with culture medium containing 500 microg/mL of oxymatrine for 2 and 5 d. The surface antigen of HBV (HBsAg) and e antigen of HBV (HBeAg) in supernatant were determined by ELISA. HBV DNA in supernatant, and intracellular covalently closed circular DNA (cccDNA), relaxed circular DNA (rcDNA) and pregenomic RNA (pgRNA) were quantified by specific real-time polymerase chain reaction (PCR) or reverse transcription (RT)-PCR. RESULTS: Treatment with oxymatrine for 2 d and 5 d reduced the production of HBV by the cell line, as indicated by the decline of HBsAg (22.67%, t = 5.439, P = 0.0322 and 22.39%, t = 5.376, P = 0.0329, respectively), HBeAg (55.34%, t = 9.859, P = 0.0101 and 43.97%, t = 14.080, P = 0.0050) and HBV DNA (40.75%, t = 4.570, P = 0.0447 and 75.32%, t = 14.460, P = 0.0047) in the supernatant. Intracellular cccDNA was also markedly reduced by 63.98% (t = 6.152, P = 0.0254) and 80.83% (t = 10.270, P = 0.0093), and intracellular rcDNA by 34.35% (t = 4.776, P = 0.0413) and 39.24% (t = 10.050, P = 0.0097). In contrast, intracellular pgRNA increased by 6.90-fold (t = 8.941, P = 0.0123) and 3.18-fold (t = 7.432, P = 0.0176) after 500 microg/mL of oxymatrine treatment for 2 d and 5 d, respectively. CONCLUSION: Oxymatrine may inhibit the replication of HBV by interfering with the process of packaging pgRNA into the nucleocapsid, or inhibiting the activity of the viral DNA polymerase.

[The spontaneous YMDD mutation rate in chronic hepatitis B patients].

OBJECTIVE: To investigate the spontaneous YMDD mutation rate. METHODS: Serum samples collected from 196 untreated chronic HBV patients were detected by primer-specific real-time PCR. RESULTS: Among 196 patients, spontaneous YMDD variants were detected in 21 subjects (20 YVDD mutants and 1 YIDD mutant). YMDD variants account for more than 50%, 25% to 50%, 9% to 25% of total virus load in 1, 5 and 15 patients, respectively. Gender, age, HBeAg status, serum viral load, the state of disease and duration of infection were not associated with spontaneous YMDD mutation. Genotype B had higher spontaneous YMDD rate than genotype C (20.00% vs 7.38%, x(2) = 6.28, P < 0.05). CONCLUSION: Spontaneous YMDD variants exist in chronic hepatitis B patients, Genotype B is associated with higher spontaneous YMDD rate.

[Comparative study on clearance dynamics of hepatitis B virion and cccDNA in blood].

OBJECTIVE: To confirm whether it is feasible to detect HBV cccDNA based on an analysis of its stability and clearance dynamics in the blood of ducks. METHODS: Twelve 1-week-old ducklings were randomized into group A and group B, and each duckling was intravenously injected with 1 milliliter of serum containing 1.2 x 10(9) copies of HBV virion (group A) or plasmid pBS HBV 3.6 II (group B). At 0.25, 2, 4, 6, 8, 12, 24 and 32 h post-injection, serum HBV DNA level in each duckling was quantified by Real-time fluorescent PCR. RESULTS: Both of HBV virion and plasmid pBS HBV 3.6 II could be detected from 0.25 h to 24 h after intravenous administration, and declined approximately by 1 log(10) every 2 hours within 8 h and was more slowly cleared at 8 h, and became undetectable at 32 h. Both of them are characteristic of one-phase exponential decay, as indicated by the relative index of the one-phase exponential curve ranging from 0.9593 to 0.9976. The mean half-life of HBV virion and plasmid (cccDNA) was 4.536 +/- 0.769 h (range: 3.5 - 5.4 h) and 4.5 +/- 0.8 h (range: 3.9 - 6.1 h) respectively. No significant difference was observed in the half-life between HBV virion and plasmid/cccDNA (t = 0.0523, P = 0.9593). CONCLUSIONS: HBV cccDNA is similar to HBV virion in its characteristics of clearance dynamics in blood. Thus it is feasible to detect serum HBV cccDNA. The clearance half-life of Hepatitis B virus in serum is not as long as previously believed at 1.0 - 1.2 d.