Inhibition of hepatitis B virus surface antigen expression by small hairpin RNA in vitro

Abstract

AIM: To explore the anti-hepatitis B virus effect of RNA interference (RNAi) using small hairpin RNA (shRNA) expression vector.

METHODS: Hepatitis B virus surface antigen green fluorescent protein (HBs-GFP) fusion vector and shRNA expression vectors were constructed and cotransfected transiently into HepG2 cells. mRNAs extracted from HepG2 cells were detected by real-time PCR. Fluorescence of HBs-GFP protein was detected by fluorescence-activated cell sorting (FACS). The effective shRNA expression vector was transfected into HepG2.2.15 cells. HBsAg and HBeAg in HepG2.2.15 cells were analyzed by radioimmunoassay (RIA) method.

RESULTS: FACS revealed that shRNA targeting at HBsAg reduced the GFP signal by 56% compared to the control. Real-time PCR showed that HBs-GFP mRNA extracted from HepG2 cells cotransfected with pAVU6+27 and HBs-GFP expression plasmids decreased by 90% compared to the empty vector control. The expressions of HBsAg and HBeAg were also inhibited by 43% and 64%, respectively.

CONCLUSION: RNAi using shRNA expression vector can inhibit the expression of HBsAg, providing a fresh approach to screening the efficient small interfering RNAs (siRNAs).

Key Words: Hepatitis B Surface Antigens; Small hairpin RNA; RNA interference; Gene expression

INTRODUCTION

Chronic hepatitis B virus (HBV) infection is recognized worldwide as a leading cause of cirrhosis and hepatocellular carcinoma. Its prevalence approaches 10% in such hyperendemic areas as China. Current therapies, including immune modulators such as interferon alfa, or nucleoside analogs such as lamivudine or adefovir have provided some cures, but could not clear HBV entirely due to persistence of viral replication[1-4].

RNA interference is a recently discovered antiviral mechanism in plants and invertebrates that is induced by double-stranded RNA (dsRNA) and leads to sequence-specific gene silencing at the post-transcriptional level[5-7]. In mammalian cells short interfering RNAs (siRNA) can also significantly and specifically suppress (knockdown) gene expression[8-13]. This process is mediated by 21-23 nucleotides, called siRNA. Recently researchers have found that shRNA can be processed into siRNA in vivo and in vitro[14-17]. A vector system called pAVU6+27, which directs the synthesis of shRNAs was applied to RNAi and could persistently suppress gene expression in mammalian cells[18].

To explore the anti-HBV therapeutic potential of RNAi by shRNA expression vector, we designed two pAVU6+27 vectors, which targeted at two distinct 21nt sequences in the HBV surface gene and a target-gene vector, which expressed the GFP and HBV surface fusion protein. We analyzed the levels of protein and RNA. It shows the model can be used for screening target site and sequence-specific shRNA expression vector can inhibit HBV surface gene expression.

MATERIALS AND METHODS

Reagents

Endoenzymes and Taq DNA polymerases were purchased from Promega Company. Pyrobest DNA polymerase was purchased from Takara Co. Ltd. Purification kit was obtained from Qiagene Company. T4 DNA ligase, Trizol reagent and Lipofectamine 2000 were purchased from Invitrogene Company. Reverse transcriptase (M-MLV) and T4 polynucleotide kinase were purchased form MBI Fermentas Company. High-grade neogenetic bovine serum was purchased from Hangzhou Sijiqing Biological Engineering Material Co. Ltd. All PCR primers were synthesized by Shanghai Boya Biological Company. SYBR gold nucleic acid gel stain was purchased from Molecular Probe Company. HBsAg and HBeAg solid phase radioimmunoassay (RIA) kits were obtained from 3V Diagnostic Tech. Co. Ltd.

Plasmids, bacterial strains and cell lines

Plasmid pcDNA3.1+S was kindly presented by Dr. Jun Chen (Gene Therapy Research Center, 302 Hospital of PLA, Beijing, China). Plasmid pAVU6+27 was a gift of Dr. Paul D. Good (Engelke Laboratory, Department of Biological Chemistry). pEGFPN-1 was purchased from Clontech Company. E.coli strain DH5a was maintained in our laboratory. HepG2, a human hepatoma cell line, and HepG2.2.15 cell line, which was derived from HepG2 cells that were transfected with a plasmid containing HBV DNA, were maintained in our laboratory.

Construction and identification of HBs-GFP fusion gene plasmid

Full length cDNA fragment of HBsAg was obtained from pcDNA3.1+S, and amplified by pyrobest DNA Polymerase using sense primer, 5’- GACAAGCTTATGGAGAACATCACATCAGG -3’ and antisense primer, 5’- CGAGAATTCCAATGTATACCCAAAGAC -3’. PCR conditions were as follows: an initial denaturation at 95 °C for 3 min, followed by 35 amplification cycles, each cycle consisting of denaturation at 94 °C for 30 s, annealing at 58 °C for 30 s, and extension at 72 °C for 45 s, and a final extension at 72 °C for 7 min. The resulting products were analyzed by electrophoresis on an 1.5% agarose gel and stained with ethidium bromide. The expected size of amplified DNA fragment was 678 bp. The fragment was inserted into an eukaryotic reporter vector pEGFPN-1 by cutting with EcoRI and HindIII. The fusion gene vector, named as pHBS-GFP, was transformed into E.coli strain DH5a and identified by restricted endonuclease digestion and DNA sequence analysis.

Construction and identification of shRNA expression vectors

Target sequences for the siRNAs were determined by using the Ambion web-based criteria. The structure of sequences is shown in Figures 1A, B. The pAVU6+27 vector contained one U6 promoter cassette and the first 27 nucleotides of human U6 RNA (Figure 1C). The pAVU6+27 vector was digested with SalI and XbaI to generate compatible ends for cloning. Two pairs of primers were incubated at 95 °C for 30 min with 1×annealing buffer (10 mmol/L Tris-HCL/100 mmol/L NaCl), gradually cooled to room temperature. The annealed shDNAs were inserted into the digested pAVU6+27 vector. The shRNA expression vectors, pAVU6+4sh421 and pAVU6+4sh81 (named on the basis of its nucleotide location in the surface gene sequence and its hairpin size), were transformed into E.coli strain DH5a and identified by PCR using sense primer, 5’-CTAACTGACACACATTCCAC-3’ and antisense primer, 5’-GCAATAAACAAGTTACTAGTCC-3’. The PCR conditions were as before. The resulting products were analyzed by electrophoresis on an 1.5% agarose gel and DNA sequence analysis.

Figure 1 Structures of hairpins and U6 expression cassette. A and B: Short-hairpin structures of pAVU6+4sh421 and pAVU6+4sh81, respectively. The loop was designed with TTCG. C: Map of pAVU6+27 U6 expression cassette including U6 promoter, the first 27 nucleotides of human U6 RNA, inserted fragments and polymerase III stem terminator.

Cell culture and cotransfection

HepG2 cells were maintained in Dulbecco’s modified Eagle’s minimal essential medium (DMEM) supplemented with 10% fetal bovine serum (FBS), streptomycin (100 μg/mL) and penicillin (100 IU/mL) at 37 °C in a humidified atmosphere containing 5 mL/L CO2. HepG2.2.15 cells were grown in DMEM with 10% FBS, 300 μg/mL G418, streptomycin (100 μg/mL) and penicillin (100 IU/mL). Twenty four hours before transfection, HepG2 and HepG2.2.15 cells were seeded in 24-well plates and cultured with fresh medium without antibiotics to 50-70% and 20-30% confluency, respectively. For transient cotransfection of HepG2 cells, 0.2 μg of reporter gene plasmids (pHBS-GFP) and 0.6 μg of shRNA expression vectors at 1:3 ratio with Lipofectamine 2000 were used as described by the manufacturer. HepG2.2.15 cells were transfected by pAV-4sh421 (1.5 μg/well) with 3 μL Lipofectamine 2000. Cells were also transfected with pAVU6+27 vector as control group; the cells not transfected also served as control.

Detection of the changes of images by FACS and microscopy

To assess the effect of RNAi on HBs-GFP fusion protein, fluorescent imaging was carried out to monitor enhanced green fluorescent protein (EGFP) expression by an inverted fluorescent microscope (DM IRB, Leica). After incubated for 72 h, HepG2 cells were harvested to measure EGFP fluorescence by FACS, using a flow cytometer (Roche Company). Fluorescence was recorded in the FL1 488-500 nm range. Specific silencing of target genes was confirmed in three independent experiments.

Reverse transcription and real-time PCR analysis

Total RNA was extracted by Trizol according to the manufacturer’s instructions. cDNA was prepared from 0.5 μg of total RNA using oligo (dT)18 as primer and M-MuLV reverse transcriptases (MBI Fermentas). Real-time PCR was performed using SYBR gold nucleic acid stain and the PCR conditions consisted of 25 amplification cycles, each at 94 °C for 30 s, at 56 °C for 30 s, and at 72 °C for 45 s. The temperature of detection point was 72 °C. Primers for HBV surface gene expression analysis were: sense primer, 5’-CTCACAATACCGCAGAGTC-3’, and antisense, 5’-TAAACTGAGCCAGGAGAA A-3’. After quantified with real-time PCR, quantification was performed using the comparative CT method. The target transcript was normalized to an endogenous reference (simultaneous GAPDH reactions). The primers of GAPDH were: sense, 5’-ACAGTCAGCCGCATCTTCTTT-3’ and antisense, 5’-GCAACAATATCCACTTTACCAGAG-3’.

Radioimmunoassay analysis

To improve the shRNA expression vector in natural HBV replication, the vector was transfected into HepG2.2.15 cells. The cultured medium was harvested on days 2, 5, 8 and 11 postransfection, in same wells. A solid phase radioimmunoassay (RIA) method was applied to detecting the binding activity of HBsAg and HBeAg according to the instructions of solid RIA kit.

RESULTS

Identification of HBs-GFP fusion gene plasmid

There were two BamH I sites in the recombinant pHBS-GFP vector and one in the pEGFPN-1 vector. pHBS-GFP was characterized by digestion of BamH I. The length of digested fragment was about 400 bp. The fragment was checked on 1.5% agarose gel electrophoresis (Figure 2). The positive clones were subjected to sequencing and the results were shown as expected.

Figure 2 Electrophoresis analysis of digestion using BamH I. Lanes 1 and 2: pHBS-GFP plasmid DNA; lane 3: DNA marker pHBS-GFP.

Identification of shRNA expression vectors

shRNA expression vectors, pAVU6+4sh421 and pAVU6+4sh81, were identified using PCR protocol, but not digestion because the length of the inserted fragment was only 53 bp. The length of products of pAVU6+4sh421 and pAVU6+4sh81 were 454 and 494 bp, respectively (Figure 3). The result of DNA sequencing showed the inserted fragment had expected known sequences.

Figure 3 Electrophoresis analysis of PCR products. Lane 1: pAVU6-4sh421 vector; lane 2: pAVU6-4sh81 vector; lane 3: pAVU6+27 vector; lane 4: DNA marker.

Inhibition of the expression of HBs-GFP fusion protein in HepG2 cells by RNAi

To determine whether siRNA specific to the HBV surface gene sequence could inhibit the expression of HBs-GFP fusion protein, we monitored the images of GFP of HepG2 cells by an inverted fluorescent microscope, and found the intensity of GFP fluorescence in HepG2 cells trended to decrease and reached the lowest degree at 72 h after transfection. Thus, we screened two shRNA vectors by FACS (Figure 4). pAVU6+4sh421 inhibited fusion protein expression significantly. The mean positive cells reduced by 56.0% compared to pAVU6+27. But pAVU6+4sh81 did not show inhibitory effects.

Figure 4 Flow cytometric detection of HBS-GFP fusion protein in HepG2 cells. HepG2 cells were transfected with pAVU6+4sh421 and pAVU6+4sh81 vectors and analyzed by flow cytometry at 72 h post-cotransfection. The values obtained from three independent experiments were expressed as mean±SD.

Inhibition of the expression of HBs-GFP gene mRNA in HepG2 cells using shRNA

Because pAVU6+4sh421 vector had a more significant inhibition on HBs-GFP fusion protein expression, we only selected this vector to investigate the effects of RNAi at RNA level. The real time PCR results are shown in Figure 5. The concentrations of HBV cDNA of pAVU6+4sh421 group and pAVU6+27 group were 104 and 105 copies/mL, respectively. The inhibition of expression of HBs-GFP fusion gene was about 90% compared to the empty vector control.

Figure 5 HBV cDNA copies detected by real time PCR analysis. Data showed shRNA-mediated inhibition of expression of HBs-GFP gene. Positive, mock, and negative groups represented as pHBS-GFP DNA, cDNA of cells only transfected with pHBS-GFP, respectively.

Inhibition of HBsAg and HBeAg expression in HepG2.2.15 cells by RNAi

The concentrations of HBsAg and HBeAg in the supernatant of HepG2.2.15 cells transfected with pAVU6+4sh421 were significantly lower than those in controls, with 43% inhibition of HBsAg and 64% inhibition of HBeAg during 3 d, and between d 5 and d 8 post-transfection (Figures 6 and 7; Tables 1 and 2). This inhibitory effect could continue for at least 9 d.

Table 1 HBsAg concentration.

HBsAg	D2	D5	D8	D11
pAVU6-4sh421	5.05±0.06	8.17±1.26	7.6±0.07	6.71±0.23
pAVU6+27	5.25±1.08	9.78±1.26	13.31±3.39	10.80±0.48

Table 2 HBeAg concentration.

HBeAg	D2	D5	D8	D11
pAVU6-4sh421	4.07±1.11	4.85±1.19	6.63±2.47	10.43±3.48
pAVU6+27	4.16±0.48	7.21±2.53	18.07±7.01	17.61±4.86

Figure 6 HBsAg concentration in the supernatant of shRNA-treated HepG2. 2.15 cells.

Figure 7 HBeAg concentration in the supernatant of shRNA-treated HepG2. 2.15 cells (S/N ratio, signal-to-noise ratio).

DISCUSSION

RNA interference (RNAi) is a key mechanism of posttranscriptional gene silencing (PTGS) and has been identified as the mediator of PTGS in plant, nematode, drosophila and animals, including humans[19-24]. It has been shown that in mammalian cells short interfering RNAs (21-23 nucleotides) processed from dsRNA could significantly and specifically inhibit the expression of homologous mRNAs. Recently, short hairpin RNAs expressed from the U6 promotor was reported to more effectively inhibit gene expression as well as chemical-synthesis than in vitro-synthesized siRNA duplexes[18,25]. In this study, we demonstrated that shRNAs could inhibit the expression of HBs-GFP fusion protein in HepG2 cells and result in reduction of the secreted HBsAg and HBeAg in HepG2.2.15 cells. These results suggest that RNAi induced by shRNA can be used as a tool in future experiments to study the ability of HBV genes to regulate its replication.

The HBs-GFP RNA expression was inhibited by 90% compared to the empty vector control and the HBs-GFP protein was reduced by 56% within 72 h after shRNA vector cotransfection, suggesting that the protein has a relatively long half-life. We found the effect of inhibition was dose-independent (data not shown). But higher concentrations of Lipofectamine 2000 seemed to be toxic to HepG2 cells. The effect of shRNA on HBs-GFP expression was slow and usually became visible 2 d after cotransfection. We also assumed that longer incubation of HepG2 cells was required for better suppression of transfected HBs-GFP; however, we failed to keep HepG2 cells healthy more than 72 h after cotransfection.

The HBV genome is a partially double-stranded 3.2 kb DNA molecule and is the template transcribed to generate the four viral RNAs[26]. The 3.5, 2.4, 2.1 and 0.9 kb transcripts encode the core protein/HBeAg, polymerase, HBsAg and X protein, respectively. The 3.5 kb pregenomic RNA not only serves for translation of the core protein/polymerase but also acts as the replicative intermediate[26,27]. In HepG2.2.15 cells, which can secrete natural HBsAg, HBeAg and Dane particles[28] and are not sensitive to transfection reagents, HBsAg and HBeAg expressions were inhibited within 3 d and the effect reached the maximum during 5-8 d and lasted at least for 9 d. It indicates that shRNA influences the HBV replication. The inhibition of HBeAg was more effective than HBsAg, suggesting that there are more copies of 2.1 kb mRNAs generating HBsAg than HBe mRNA in HepG2.2.15 cells.

To date, it is not clear why not all target sites are effective for RNAi. The efficacy of siRNA influenced by secondary structure has been reported[29,30]. Using the fusion protein, including report gene and target gene, we screened one site, which was simultaneously proved effective in HepG2.2.15. The approach can be used to determine the efficacy of siRNA and is likely to have wide application. In particular, it may facilitate the studies of gene function in transfectable cell lines.

Although chronic HBV infection is a major health problem worldwide, there is no complete effective antiviral treatment. siRNA technology may provide a possible therapeutic strategy against chronic HBV infection.

ACKNOWLEDGEMENTS

We thank Dr. Jun Chen for providing pcDNA3.1+S plasmid and Dr. Paul D. Good (Engelke Lab) for supplying pAVU6+27 plasmid. We also thank Dr. Hai-Hong Zhu for helpful discussions and Rong-Huang Liu and Ming-Wei Li for their technological support.