key: cord-0005704-xwgin0yb
authors: Wu, Hui-Lin; Huang, Li-Rung; Huang, Chuan-Chuan; Lai, Hsiao-Lei; Liu, Chun-Jen; Huang, Yu-Tzu; Hsu, Yun-Wei; Lu, Cheng-Yi; Chen, Ding-Shinn; Chen, Pei-Jer
title: RNA interference-mediated control of hepatitis B virus and emergence of resistant mutant
date: 2005-07-27
journal: Gastroenterology
DOI: 10.1053/j.gastro.2004.12.007
sha: 43f3abb33bb3cbef01096f58378b774bbcbe3b41
doc_id: 5704
cord_uid: xwgin0yb

Background & Aims Present therapy for chronic hepatitis B attains control only in limited proportions. Small interfering RNA (siRNA) offers a new tool with potential therapeutic applications for hepatitis B virus (HBV). Given the importance of sequence identity in the effectiveness of siRNA and the heterogeneity of HBV sequences among different isolates, a short hairpin RNA (shRNA)-expressing plasmid, pSuper/HBVS1, was developed to target a region conserved among major HBV genotypes and assess its effectiveness control of HBV. Methods HBV replication-competent plasmid was cotransfected with pSuper/HBVS1 to HuH-7 cells or to mice. The levels of viral proteins, RNA, and DNA were examined in transfected cells and animals. The effects of pSuper/HBVS1 on clinical isolates with genotypes B and C were also determined. Results pSuper/HBVS1 significantly decreased levels of viral proteins, RNA, and DNA for HBV genotype A in cell culture and in mice. Comparable suppressive effects were observed on clinical isolates of genotypes B and C. A clone with a silent mutation in the target region was identified from a patient with genotype C. This mutant revealed diminished sensitivity to pSuper/HBVS1 and could be selected out in the presence of pSuper/HBVS1 in cell culture. Conclusions These findings indicated that shRNA could suppress HBV expression and replication for genotypes A, B, and C, promising an advance in treatment of HBV. However, the emergence of resistant mutants in HBV quasispecies should be considered.

W ith an estimated 300 million chronic carriers, hepatitis B virus (HBV) infection remains one of the most prevalent chronic viral infections of humans. 1 Chronic infections cause serious consequences, including cirrhosis and hepatocellular carcinoma, 2 responsible for at least 1 million deaths annually worldwide. 1 Thus, control of HBV infection by prevention or treatment is imperative.

A preventive vaccine is available to block HBV infection successfully; however, therapeutic intervention is the only option for chronic HBV carriers to delay or prevent the progression to life-threatening, end-stage liver diseases. Immune modulators (ie, interferon ␣ [IFN-␣]) and nucleoside analogues (ie, lamivudine) are 2 approved mainstream treatments for chronic hepatitis B. However, both therapies achieve only limited response, and none of them can eradicate the virus effectively. 3 Other strategies, including antisense RNA and DNA constructs, hammerhead ribozyme, and dominant negative HBV core proteins, are still under development. 4, 5 Therefore, the quest for potent antiviral medications to treat chronic hepatitis B continues.

Recently, it was found that double-stranded RNA (dsRNA) can trigger a sequence-specific, gene-silencing process called RNA interference (RNAi). 6 RNAi is a very specific and potent mechanism to silence gene expression through processing of double-stranded RNA into 21-26 nucleotide (nt), short interfering RNA (siRNA). In plants, it serves as a host-defense mechanism against viruses and transposable elements (reviewed in Waterhouse et al 7 ) . After the discovery that RNAi pathways also exist in mammalian cells and that the use of siRNA can avoid cell death induced by dsRNA longer than 30 nucleotides, 8 siRNA has emerged as a novel therapeutic approach in the fight against human viral infections. This study therefore examined whether siRNA can in-hibit HBV replication and expression and exhibit the potential to combat chronic hepatitis B.

During the course of this work, a growing number of studies have shown that siRNA can be effective against several pathogenic viruses, including poliovirus, HIV, HCV replicons, influenza virus, SARS coronavirus, and others. 9, 10 Several studies employing siRNA or short hairpin RNA (shRNA) to inhibit HBV [11] [12] [13] [14] [15] [16] [17] [18] also revealed that RNA interference indeed has the potential to treat viral infections, including HBV. However, variation of sequences among different viral isolates in clinical settings and emergence of resistant mutants constitute potential problems hindering the efficacy of siRNA. This work identified a target site for siRNA that is conserved among HBV genotypes A to G. A plasmid expressing shRNA targeted this site significantly inhibited the steps in HBV replication that occur in cultured cells and in mice. The problem of resistant mutants was also explored in this study.

Plasmid pHBV48 consists of a greater than genomelength HBV fragment (subtype adw, genotype A) that can initiate HBV replication after being transfected into hepatoma cell lines. 19 To clone the full-length HBV genome from clinical samples, a method described by Gunther et al was used. 20 After amplication, PCR products were purified and cloned into the vector yT&A (Yeastern Biotech Co., Shijr, Taiwan).

shRNA were cloned downstream of the human H1 promoter in the vector pSuper as described. 21 The target sites for siRNA were chosen on the basis of their conservation among the major HBV genotypes. The selected sequences were submitted to a BLAST search to avoid targeting to the human genome. We used the primers 5=-GATCCCCCTGGATCCTGCGCGGGACGTTCAAG-AGACGTCCCGCGCAGGATCCAGTTTTTGGAAA-3= and 5=-AGCTTTTCCAAAACTGGATCCTGCGCGGGAC-GTCTCTTGAACGTCCCGCGCAGGATCCAGGGG-3= for pSuper/HBVX1; 5=-GATCCCCGGTATGTTGCCCGT-TTGTCTTCAAGAGAGACAAACGGGCAACATACCTT-TTTGGAAA-3= and 5=-AGCTTTTCCAAAAAGGTATG-TTGCCCGTTTGTCTCTCTTGAAGACAAACGGGCAA-CATACCGGG-3= for pSuper/HBVS1. Relative positions of shRNA targets on HBV are shown in Figure 1 . For injection into mice, endotoxin-free plasmid DNA was prepared by the EndoFree Plasmid Kit (Qiagene, Hilden, Germany).

HuH-7 cells were grown at 37°C under 5% CO 2 in Dulbecco's modified Eagle medium, supplemented with 10% FCS (Biological Industries, Kibbutz Beit Haemek, Israel) and 2% L-glutamine. The cells were plated at a density of 9 ϫ 10 5 cells per 60-mm dish or 4.5 ϫ 10 5 cells per well in 6-well plates 24 hours prior to transfection. Transfecion of cells was performed with lipofectamine 2000 (Invitrogen Life Technologies, CA) following the user guidelines. To assay the replication of HBV genome from clinical isolates, we cleaved the PCR product or HBV monomer cloned in pyT&A with SapI prior to transfection, as described previously. 22 After transfection, the supernatant was collected for the evaluation of hepatitis B surface antigen (HBsAg) and hepatitis B e antigen (HBeAg), and the cells were processed for the detection of viral DNA and RNA by Southern and Northern blotting, respectively.

Possible adverse effects of pSuper/HBV S1 on cells were evaluated by the morphology, growth rate, non-HBV protein expression, and cytotoxicity of transfected cells. Cytotoxicity was determined by MTS assay with CellTiter96 Aqueous 1 solution proliferation assay following the instructions of the manufacturer (Promega, Madison, WI). Expression level of one cellular protein, ␣-fetoprotein, was determined by tumor markers AFP kit of AXSYM systems (Abbott, GmbH Diagnostica, Wiesbaden, Germany).

For the in vivo experiments, we used 6-to 8-week-old C57BL/6 mice. Five g pHBV48, 2.5 g pSEAP2-Control (BD Biosciences, Palo Alto, CA), and 20 g pSuper or pSuperderived plasmid DNA were injected into the tail vein of mice in a volume of phosphate-buffered saline (PBS) equivalent to 8% of the mouse body weight. 23, 24 The total volume was delivered within 5 seconds. The secreted alkaline phosphatase (SEAP) level in the mouse serum was measured by BD Great EscAPe SEAP chemiluminescence detection kit (BD Biosciences, Palo Alto, CA) and used to normalize transfection efficiency. The mice sera were assayed for HBsAg, HBeAg, or HBV DNA content at 1, 4, 7, and 11 days after injection. The livers of mice were preserved in optimal cutting temperature (OCT) for immunohistochemical analysis.

All mouse experiments were carried out according to the guidelines established by the Institutional Animal Care and Use Committee at the National Taiwan University College of Medicine.

Levels of HBsAg and HBeAg in the media of the transfected cells, and in the sera of the treated mice, were determined using the AXSYM systems kit (Abbott, GmbH Diagnostica). HBV core protein (HBc) was visualized by immunohistochemical staining of tissues embedded in OCT by rabbit anti-HBc antibodies and Envision ϩ System, HRP (DAB) (DAKO Corp., Carpinteria, CA). The liver sections were also stained with hematoxylin.

Total RNA was extracted from transfected cells with RNA-Bee reagent (TEL-TEST Inc., Freindswood) according to the manufacturer's instructions. Twenty g RNA were separated by electrophoresis on a 1% agarose gel and transferred to a nylon membrane (Roche Diagnostics GmbH, Mannheim, Germany). The blot was probed with a DIG-labeled DNA fragment corresponding to nucleotides 1372-1833 of the HBV X sequence. The probe was generated with a PCR DIG Probe Synthesis kit (Roche Diagnostics GmbH, Mannheim, Germany).

Purification of HBV DNA from intracellular core particles was performed according to the protocol described previously, 25 with minor modifications. Briefly, the cells were lysed in 300 L isotonic lysis buffer (10 mmol/L Tris-HCl, pH7.5, 1 mmol/L EDTA, 150 mmol/L NaCl, 0.5% Nonidet P-40) per 60-mm dish. The nuclei were removed from the lysates by centrifugation for 10 minutes at 14,000 rpm. The cytoplasmic HBV DNA was purified by digestion of proteins with 200 g/mL of proteinase K in the presence of 0.5% of sodium dodecyl sulfate at 37°C for overnight, phenol-chloroform extraction, and ethanol precipitation. Purification of HBV DNA from extracellular viral particles was performed following the protocol described in the previous study. 20 One third of the DNA purified was fractionated in a 1% agarose gel, blotted to a nylon membrane, and hybridized with a DIG-labeled DNA fragment covering the entire surface gene.

Genotypic Analysis of the siRNA Target Region of HBV Viral DNA was isolated from cell-free culture supernatant with the QIAamp viral RNA kit (Qiagene) according to the manufacturer's instructions. Five L purified viral DNA was used as template for PCR reaction with sense primer HBVϩ2836 (5=-GGAACAAGAGCTACAGCATG-3=) and antisense primer HBV-695 (5=-AACAAATGGCACTAGTA-AAC-3=). PCR products were purified with Gel/PCR DNA fragments extraction kit (Geneaid, Taoyuan, Taiwan) and sequenced in an automatic sequencer ABI Model 3700 with primer HBVϩ56 (5=-CCTGCTGGTGGCTCCAGTTC-3=). For analyzing the sequence of individual clones, PCR products were cloned to the vector pyT&A (Yeastern Biotech Co.).

To examine the ability of siRNA to inhibit HBV gene expression, 2 target sequences located on the X (siRNAX1) and surface (siRNAS1) genes of HBV were selected. Because all HBV transcripts overlapped in their C-terminals, multiple viral RNA could be inhibited by a single siRNA ( Figure 1 ). Both siRNA used can target pregenomic RNA that serves as the template for HBV genomic replication and the mRNA for the core and polymerase proteins. These 2 siRNA also target transcripts for HBV surface antigens, the major viral proteins of the envelope. Furthermore, siRNAX1 targets the transcript encoding the X protein. This study employed an H1 promoter-based vector (pSuper) to express the siRNA of interest in the form of shRNA.

To evaluate the effects of HBV siRNA on HBV gene expression, we measured HBsAg and HBeAg levels in the culture media of cells cotransfected with a replication-competent HBV construct (pHBV48) and HBV shRNA-expression plasmids. The present study found that the pSuper/HBVS1 could efficiently suppress the expression of HBsAg and HBeAg for up to 11 days ( Figure 2 ) in culture. The kinetics and efficiency of suppression differs for HBsAg and HBeAg. Maximal suppression for HBeAg occurred around day 2 posttransfection, and the suppression effect decreased thereafter. Suppression of pSuper/ HBVS1 on HBsAg expression attained maximum level around days 4 to 6, and maintained a similar level thereafter. Treatment with pSuper/HBVS1 caused approximately 98.2% reduction of HBsAg on day 6 and 62.6% of HBeAg on day 2, respectively, compared with the empty-vector control ( Figure 2 ). Conversely, pSuper/HBVX1 showed no significant suppressive effect on HBV protein expression (data not shown).

HBV RNA levels in cells were examined by Northern blotting. Total cellular RNA derived from transfected cells was hybridized with a probe-spanning X gene to hybridize equally to all HBV transcripts. The RNA levels of 3.6 kb, 2.4 kb, and 2.1 kb were significantly suppressed by pSuper/HBVS1, compared with those in cells cotransfected with pHBV48 and pSuper vector alone ( Figure 3A ). Consistent with the results of secreted viral proteins in culture supernatant, pSuper/HBVX1 did not show a suppressive effect on HBV transcripts on the RNA level either. These results indicated that the suppressive effect of siRNA on HBV expression on the RNA level is comparable with that on the protein level. We thus used the expression levels of viral antigens as a read-out system for viral RNA expression in further studies.

To examine whether shRNA could also suppress HBV replication, we used Southern blotting to detect HBV DNA. As shown in Figure 3B , pSuper/HBVS1 can also suppress HBV replication. This was reflected by reduced HBV replication intermediates in transfected cells and HBV DNA in viral particles secreted in culture media. The decreased HBV replication and expression in pSuper/HBVS1-transfected cells was not due to nonspecific cell toxicity. The morphology of cells transfected with pSuper/HBVS1 did not differ markedly from those of pSuper-transfected or untransfected cells. MTS proliferation assay for pSuper/HBVS1 and pSuper-transfected cells showed comparable proliferation results (data not shown). Furthermore, expression levels of cellular RNA or protein were unaffected by the presence of pSuper/ HBVS1, including GAPDH RNA ( Figure 3A ), as well as the secreted ␣-fetoprotein (AFP) expressed by HuH-7 cells (data not shown). The results indicated that the decrease in HBV replication and expression in these cells is not due to a nonspecific cell toxicity problem but is due to a specific suppressive effect of pSuper/HBVS1 on HBV.

Because pSuper/HBVS1 showed substantial inhibition on HBV expression and replication in culture, this shRNA expression plasmid was further used to examine whether it could also display HBV inhibition in vivo. pHBV48 was cotransfected with pSEAP2-Control and pSuper or pSuper/HBVS1 to mouse liver by hydrodynamic injection. The level of alkaline phosphatase in serum was measured to monitor the transfection efficiency of hydrodynamic injection. As shown by studies from the present authors and others, 26 hydrodynamic injection of replication-competent HBV plasmid DNA into mouse resulted in HBV replication in the liver and secretion of viral antigens to the serum. The level of HBsAg in the mouse sera and HBc expression in the mouse liver were used as indicators to evaluate the suppressive effect of pSuper/HBVS1. In concordance with findings from cell culture experiments, the HBsAg expression in mice that received HBV replication-competent construct (pHBV48) was suppressed dramatically by pSuper/HBVS1. Serum HBsAg in mice treated with pSuper/HBVS1 was reduced by over 99% on day 4 when compared with mice without pSuper/HBV1 treatment ( Figure 4A ).

Expression of another important viral protein, the nucleocapsid (HBc), was also inhibited by pSuper/ HBVS1. HBV core protein is synthesized in infected or transfected cells from the 3.6-kb pregenomic RNA and is required for viral replication. An immunohistochemical study showed that there were approximately 10% HBc-positive hepatocytes in livers of mice receiving pHBV48 and pSuper ( Figure 4B, left panel) . In contrast, HBc-stained cells were hardly found in livers of mice receiving pHBV48 and pSuper/HBVS1 ( Figure 4B , right panel). The numbers of stained hepatocytes in sections from pSuper/HBVS1-treated mice was reduced by more than 95%.

Significant diversities occur in sequences of HBV isolates. Based on sequence divergence, HBV can be classified into at least 8 genotypes. 27 To warrant a widespread application of the potential siRNA therapy, the target sequence chosen for pSuper/HBVS1 was conserved among all major HBV genotypes A-G. pSuper/HBVS1 has been shown to suppress the expression and replication of a laboratory clone of genotype A, pHBV48. Further testing concerned whether or not pSuper/HBVS1 could also inhibit the expression and replication of HBV with different genotypes from clinical isolates. Because genotypes B and C are the most prevalent genotypes in Asia, including Taiwan, this study focused on these 2 genotypes. The effect of pSuper/S1 was examined on a heterogeneous mixture of amplified HBV genomes from clinical samples without cloning. Full-length HBV genome of genotype B or C was amplified from the patient's serum by PCR. The PCR products were digested with SapI and transfected to HuH7 cells with pSuper/ HBVS1 or pSuper. A previous study has demonstrated that linear monomeric HBV genomes with SapI ends can initiate a full replication cycle, leading to viral antigen expression. 20 Our results showed that pSuper/HBVS1 could also suppress expression of HBsAg and HBeAg of HBV genotypes B and C from clinical isolates to a similar extent as on a laboratory clone of genotype A ( Figure 5 ).

Full-length HBV genomes from clinical isolates were also cloned into a vector, and the effect of pSuper/ HBVS1 on their expressions was assayed. The results resembled those on PCR products (data not shown). Nevertheless, 1 of 10 clones from a patient with genotype C proved more resistant to pSuper/HBVS1, al- though the total PCR product from the same patient showed no obvious resistance to pSuper/HBVS1 treatment. After sequencing, this particular clone (pHBV-CI-m) was found to exhibit a silent mutation in the target sequence of pSuper/HBVS1 ( Figure 6A ). This mutation did not change the amino acid sequence of either the polymerase gene or the surface gene. Neither directly sequenced PCR product nor the other 9 clones from the same patients displayed this nucleotide change within the target sequence of pSuper/HBVS1, indicating that this mutant clone represents only a minor population. The next concern was whether or not this mutant clone would be selected out in the presence of pSuper/ HBVS1. Mimicking the heterogeneous population of HBV in patients, pHBV-CI-m was mixed with pHBV-CI-wt at a ratio of 1:9 and transfected to HuH-7 cells in the presence of pSuper/HBVS1 or pSuper. The HBV genome was purified in the culture supernatants at days 2 and 5 posttransfection. An HBV fragment covering the target sequence of pSuper/HBVS1 was amplified and the PCR products cloned. In the presence of pSuper/HBVS1, the proportion of HBV with mutant sequence was significantly increased at day 5 posttransfection, as reflected by the results from sequencing PCR products directly and at least 10 individual clones for each experiment. In the electropherogram of PCR product sequencing, it was evident that the dominant signal had changed from "T" to "C" for the sample treated with pSuper/HBVS1 at day 5 posttransfection but not at day 2 posttransfection ( Figure 6B ). Results of sequencing individual clones also revealed the increase of the mutant clone (summarized in Table 1 ). At day 2 posttransfection, ratios of mutant clone were quite similar for samples from cells cotransfected with pSuper or with pSuper/HBVS1 (20%). At day 5 posttransfection, the ratio rose significantly to 43% in the sample treated with pSuper/HBVS1, and the ratio remained at 25% for the sample treated with vector pSuper alone (Table 1 ). In another independent experiment, the ratios of mutant clones from cells transfected with pSuper/HBVS1 and pSuper were 30% and 10%, respectively, at day 5 posttransfection. All these results strongly implied that siRNA could exert a selection pressure on preexisting resistant mutant(s).

With the advantages of being highly specific, potent, and safe for gene silencing, the potential of using siRNA as an antiviral agent has been actively explored in numerous studies. In this work, RNA polymerase III promoter-based vector was used to express shRNA against HBV and clearly demonstrated the strength of this vector in inhibiting HBV expression and replication in both cell culture and in mouse model systems. Furthermore, the shRNA-expressing plasmid used in this study cannot only suppress the expression of laboratory prototype HBV strain but also can suppress expression of HBV from clinical isolates with different genotypes.

During the preparation of this work, several studies documented the anti-HBV activity of RNA silencing by either synthetic siRNA or shRNA-expression plasmids in vitro [11] [12] [13] [14] [15] [16] [17] [18] and in vivo. 12, 13, 18 These results resembled the present investigative outcomes while supporting siRNA as a promising approach of controlling HBV. What makes pSuper/HBVS1 distinctive is that its target site is the only one conserved for 7 HBV major genotypes (A to G) among all siRNA target sequences published so far. Only genotype H contains a nucleotide difference within the pSuper/HBVS1 target sequence. Nucleotide 472 (the same nucleotide in the previously identified pSuper/HBVS1 resistant clone) is G in genotype H rather than T in other genotypes. Nevertheless, genotype H is not common worldwide. Given the high heterogeneity of HBV sequences and the sensitivity of siRNA to the changes of sequences, 28 it is advantageous to design potent siRNA targeting highly conserved regions among various HBV isolates. However, it is very challenging to develop an siRNA targeting highly conserved regions and also being very effective for HBV. In fact, pSuper/ HBVS1 is the only one found so far. Conservation of an siRNA target sequence can generalize the utility of siRNA to different HBV genotypes and strains, as illustrated by the suppressive effects of pSuper/HBVS1 on the expression of HBV genotypes A, B, and C in this study. Although the inhibitory effects of pSuper/HBVS1 on other genotypes were not tested because of lack of clinical samples, pSuper/HBVS1 very likely can suppress the expression of other genotypes with the conserved target sequence.

Although this and other studies have proved the feasibility of using siRNA to control HBV expression and replication in vivo and in vitro, emergence of resistant mutants may be problematic. Previous studies on using siRNA to control replication of human immunodeficiency virus or poliovirus in culture have noted the emergence of resistant mutants. 29 -32 These resistant mutants either emerged de novo with molecularly cloned viral genomes 29, 30 or were likely present in the initial virus population and selected out in the presence of siRNA. 32 In this study, an HBV variant with a silent mutation within the target site of pSuper/HBVS1 was found from one HBV chronic carrier. To our knowledge, this is the first experimental proof that resistant mutant indeed preexists in the quasispecies of HBV patients. Data were also presented supporting the notion that this mutant can be selected out in the presence of siRNA. This finding, together with others, underlines the importance of sequence identity of siRNA to its target sequence to be effective. Combinations of multiple siRNA targeting separate regions of the genome can alleviate the problem of resistant mutants. Use of siRNA targeting functionally indispensable and conserved regions can further minimize this problem, reducing the numbers of siRNA used and therefore the cost. For pSuper/HBVS1, the mutation identified for pHBV-CI-m at nucleotide 472 caused silent mutation for both surface and polymerase genes within this conserved target region. In genotype H, the difference at nucleotide 472 left unchanged the amino acid sequence of HBV surface proteins, causing Leu to Val conservative change in the polymerase reading frame. These findings implied that probably only very few mutations can be tolerated by HBV within pSuper/HBVS1 target sequence. Therefore, preventing the emergence of resistant clones can be simply achieved by including shRNA against the mutant sequence. This problem may also be overcome by combining with other antiviral strategies or siRNA targeting host factors that are dispensable for cell but essential for virus. Because all HBV transcripts overlap in their C terminals, a single shRNA targeting multiple critical transcripts simultaneously could be designed to maximize the inhibitory potency. In this regard, pSuper/HBVS1 targets all HBV major transcripts except X mRNA. Interestingly, this investigation found that pSuper/ HBVS1 suppressed the expression of HBsAg more efficiently than that of HBeAg, although it targeted the same sequence on the transcripts for these 2 proteins. In contrast, the S-target siRNA used by Klein et al resulted in greater suppression of HBeAg than HBsAg. 18 Song et al also noticed that the same siRNA could have varying suppressive effects against human immunodeficiency virus in different cell types. 33 Currently, it is uncertain why siRNA exerts differential suppressive effects on the same target sequence on different RNA molecules or in different cell types. One possible speculation is that the target sequence in different RNA molecules may have different secondary structures or are bound with different proteins and thus make their accessibility to the same siRNA different.

In natural infections, HBV uses covalently closed circular DNA (cccDNA) as the template for transcription. Therefore, to treat chronic HBV infection, the siRNA needs to either sustain in the infected hepatocytes to continuously degrade HBV RNA or to inactivate cccDNA from transcription. Thus far, siRNA-induced gene silencing in mammalian cells seems to be transient because of the likely lack of the RNA silencing-amplification mechanism present in worms and plants. 34 Recently developed lentivirus, retrovirus, and adenovirusassociated virus (AAV)-based shRNA expression vectors may provide a promising approach for long-term expression in mammals, although safety issues would be a concern. The effect of siRNA/shRNA on cccDNA of HBV would not be easy to address in the experimental systems used in the current study. Neither HCC cells transfected with HBV DNA nor mice hydrodynamically injected with HBV DNA have been generally proficient in providing cccDNA templates for pregenomic RNA synthesis. 26, 35 Recent advances in inducing DNA methylation by siRNA in human cells 36, 37 have shed light on using siRNA to inactivate HBV cccDNA from transcription. Further studies will address the effects of siRNA on HBV cccDNA.

This and other studies have clearly established that siRNA/shRNA could serve as a powerful potential therapeutic tool for treating hepatitis B infection as well as other diseases, although some technical obstacles must be overcome. The quick advances of siRNA technology, including optimization of target sequences, improvement of delivery methods, and expression vectors, would definitely render siRNA more clinically feasible and useful in the future. Furthermore, more knowledge of the working mechanisms of the siRNA-mediated effect on DNA genome and systemic silencing in plants and other organisms (reviewed in Agrawal 38 ) would facilitate the development of strategies to induce these activities in mammalian cells. By then, siRNA will have become an even more potent and ideal therapeutic agent for HBV as well as other viral infections.

Strategies to prevent and control hepatitis B and C virus infections: a global perspective

Hepatocellular carcinoma and hepatitis B virus. A prospective study of 22,707 men in Taiwan

Advances in therapy for chronic hepatitis B

Nucleic acid-based antiviral and gene therapy of chronic hepatitis B infection

Gene therapy for chronic viral hepatitis: ribozymes, antisense oligonucleotides, and dominant negative mutants

Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans

Gene silencing as an adaptive defence against viruses

Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells

Human viruses under attack by small inhibitory RNA

Silencing SARS-CoV Spike protein expression in cultured cells by RNA interference

Inhibition of hepatitis B virus expression and replication by RNA interference

Inhibition of hepatitis B virus in mice by RNA interference

Small interfering RNA inhibits hepatitis B virus replication in mice

Selective inhibition of hepatitis B virus replication by RNA interference

Short interfering RNA-directed inhibition of hepatitis B virus replication

Inhibition of hepatitis B virus replication by stably expressed shRNA

Inhibition of HBV replication by siRNA in a stable HBV-producing cell line

Inhibition of hepatitis B virus replication in vivo by nucleoside analogues and siRNA

Temporal aspects of major viral transcript expression in Hep G2 cells transfected with cloned hepatitis B virus DNA: with emphasis on the X transcript

A novel method for efficient amplification of whole hepatitis B virus genomes permits rapid functional analysis and reveals deletion mutants in immunosuppressed patients

A system for stable expression of short interfering RNAs in mammalian cells

A prospective study characterizing full-length hepatitis B virus genomes during acute exacerbation

Hydrodynamics-based transfection in animals by systemic administration of plasmid DNA

High levels of foreign gene expression in hepatocytes after tail vein injections of naked plasmid DNA

Production of hepatitis B virus covalently closed circular DNA in transfected cells is independent of surface antigen synthesis

Hydrodynamic injection of viral DNA: a mouse model of acute hepatitis B virus infection

Genetic variability in hepatitis B viruses

Tolerance for mutations and chemical modifications in an siRNA

Human immunodeficiency virus type 1 escapes from RNA interference-mediated inhibition

Human immunodeficiency virus type 1 escape from RNA interference

Nucleic acid-based immune system: the antiviral potential of mammalian RNA silencing

Short interfering RNA confers intracellular antiviral immunity in human cells

Sustained small interfering RNA-mediated human immunodeficiency virus type 1 inhibition in primary macrophages

The small envelope protein is required for secretion of a naturally occurring hepatitis B virus mutant with pre-S1 deleted

Induction of DNA methylation and gene silencing by short interfering RNAs in human cells

Small interfering RNA-induced transcriptional gene silencing in human cells

RNA interference: biology, mechanism, and applications

The authors thank Dr. Reuven Agami for his generous gift of pSuper; Chih-Chiang Wang, Chung-Hsun Yu, Shiou-Lin Lin, Jia-Bao Pan, Chin-Yi Shai, Meng-Chun Shih, and Yung-Tai Chen for technical assistance; and Dr. Lih-Hwa Hwang for helpful discussion.