©2009 International Medical Press 1359-6535 (print) 2040-2058 (online) 797

Antiviral Therapy 2009 14:797–808 (doi: 10.3851/IMP1294)

Background: Toll-like receptors (TLRs) are a key component of the innate immune system and TLR2 has been shown to be involved in the immunopathogenesis of hepatitis B virus (HBV) infection in vivo. We investigated the role of TLR2 stimulation of virus-infected hepatocyte cell lines as a potential antiviral mechanism in vitro.Methods: The hepatoblastoma cell line HepG2 was transduced with recombinant HBV baculoviruses and the hepatoma cell line Huh-7 was transiently transfected with complimentary DNA clones of HBV. HBV viral replication was quantified after stimulation with interleukin (IL)-1β and Pam-2-Cys, a synthetic TLR2 ligand, by measuring intra-cellular core-associated single-stranded HBV DNA using Southern blot hybridization, as well as viral nucleocapsid formation using a non-denaturing immunoblot method.

Results: Stimulation of both cell lines in vitro with IL-1β and Pam-2-Cys, both known to induce expression of the pro inflammatory cytokines tumour necrosis factor-α and IL-8 via a nuclear factor-κB dependent pathway, resulted in the inhibition of HBV DNA replication in the transduced HepG2 cells by up to 90% and nucleocapsid formation in the transiently transfected Huh-7 cells by up to 30%, when compared with mock-treated cells.Conclusions: Hepatoma cell lines expressed functional IL-1 receptor and TLR2 receptors, which when stimulated led to a signalling cascade that inhibited HBV replica-tion. These data support an active role for hepatocytes in inhibiting HBV replication and provide a rationale for the development of TLR agonists as potentially novel antiviral agents.

Chronic infection with hepatitis B virus (HBV) is a major cause of liver disease, particularly in the Asia-Pacific region, contributing to cirrhosis, hepatic decom-pensation and hepatocellular carcinoma (HCC) [1,2]. Liver damage is a consequence of the immune system’s attempted clearance of HBV [3,4]. Conventional thera-pies for chronic HBV (CHB) include direct antiviral agents (for example, lamivudine [3TC]) and immune modulating agents (for example, interferon [IFN]), which reduce HBV replication in order to control dis-ease progression [1]. However, IFN-based treatments are poorly tolerated and antiviral agent efficacy is limited by acquired drug resistance [1]. These shortcomings high-light the need for improved therapeutic approaches.

The innate immune response is the first line of host defence targeting exogenous pathogen infec-tion through the recognition of pathogen-associated molecular patterns. The Toll-like receptors (TLRs) are the primary pathogen pattern recognition receptors involved in the innate response. Thirteen members of the TLR family have now been identified. TLR1-9 are common to human and mouse. TLR10 appears to be functional only in humans and TLR11-13 occur only in the mouse [5–7]. TLRs signal via their Toll/interleukin (IL)-1 receptor-domains to promote pro-inflammatory cytokine responses via two signalling pathways – the canonical MyD88-dependent pathway, and the MyD88-independent pathway [5].

Original article

Stimulation of the interleukin-1 receptor and Toll-like receptor 2 inhibits hepatitis B virus replication in hepatoma cell lines in vitro

Alex J Thompson1,2,3, Danni Colledge1, Sally Rodgers1, Rachel Wilson1, Peter Revill1, Paul Desmond3, Ashley Mansell4, Kumar Visvanathan5 and Stephen Locarnini1*

1Victorian Infectious Diseases Reference Laboratory, North Melbourne, VIC, Australia2Duke University Medical Centre, Durham, NC, USA3Department of Gastroenterology, St Vincent’s Hospital, Fitzroy, VIC, Australia4Centre for Innate Immunity and Infectious Diseases, Monash Institute of Medical Research, Monash University, Clayton, VIC, Australia5Department of Medicine, Monash Medical Centre, Monash University, Clayton, VIC, Australia

*Corresponding author: e-mail: Stephen.Locarnini@mh.org.au

Introduction

Locarnini.indd 797 22/9/09 10:28:44

©2009 International Medical Press798

AJ Thompson et al.

The MyD88-dependent pathway is used by all TLRs with the exception of TLR3. Downstream signal-ling activates the transcription factors nuclear factor (NF)-κB and activator protein-1, which in turn induces the expression of genes involved in the inflammatory response, including tumour necrosis factor (TNF)-α, IL-1, IL-6 and IL-8. TLR7/8 and TLR9 are also able to use MyD88 to induce the expression of type-1 IFN. By contrast, the MyD88-independent pathway is used selectively by TLR3 and TLR4, which use the TIR-do-main-containing adaptor inducing IFN-β (TRIF) to acti-vate downstream interferon regulatory factor-3, subse-quently inducing the expression of IFN-β. TRIF is also able to induce a late phase pro- inflammatory cytokine response via an alternate pathway of NF-κB activation.

Not surprisingly, pathogens have evolved strategies to manipulate these receptors and signalling pathways to circumvent the host immune response. The recent identification of the expression of TLRs on human hepatocytes by Visvanathan et al. [8] indicated that the hepatocyte itself might play a role in innate immune defence against the HBV in vivo. The observation that hepatocyte TLR2 expression was down-regulated in the setting of hepatitis B e antigen (HBeAg)-positive CHB, but up-regulated in HBeAg-negative CHB [8], raised the possibility that TLR2 signalling might be an anti-viral strategy employed by the hepatocyte to control HBV replication. Consequently, it also added increased weight to a possible role for the precore (PC) protein of HBV (HBeAg) in modulating TLR2 expression and signalling in vivo.

There are a number of human hepatoma cell lines that support HBV replication in vitro [9]. These include the ATCC HepG2 cell line, which supports the recom-binant HBV baculovirus (BCV) transduction model [10] and the Huh-7 cell line, which supports transient transfection of infectious clones of HBV complimentary DNA (cDNA) in vitro [9]. Presently, there is little infor-mation concerning the use of cell-culture-based models for the study of innate immune responses or TLR-medi-ated events in the setting of HBV replication.

We have recently performed a comprehensive char-acterization of TLR signalling pathways in the ATCC HepG2 and Huh-7 cell lines [11]. A limited reper-toire of TLR expression was observed, consistent with two previous studies of TLR gene expression [12,13]. HepG2 cells expressed IL-1 receptor (IL-1R), TLR1, TLR2, TLR3 and TLR6 messenger RNA (mRNA), but minimal TLR4, TLR7 and TLR9 mRNA and no TLR5, TLR8 or TLR10 mRNA [11]. TLR2 forms heterodimers with either TLR1 or TLR6 to recognize specific ligands [5–7]. A similar expression profile was noted in the Huh-7 cell line [11]. In both cell lines, only the IL-1R and TLR2 were functional, however, and specific stimulation of these receptors resulted in

MyD88-dependent NF-κB activation and induction of the pro-inflammatory cytokines TNF-α and IL-8 [11]. A number of different steps in the pathway were inves-tigated to confirm specificity and physiological signifi-cance, including phosphorylation of NF-κB, increase of TNF-α, IL-8 and IL-6 mRNA expression, and TNF-α protein translation. A hierarchy of ligand response was noted, with potency of IL-1β>Pam-2-Cys (P2C; specific for the TLR2/TLR6 heterodimer) >P3C (specific for the TLR2/TLR1 heterodimer), respectively. No type-1 IFN response could be stimulated, consistent with previous reports that TLR2 does not activate the MyD88-inde-pendent signalling pathway in these cells [14]. Real time quantitative PCR (RT-qPCR) for induction of TNF-α mRNA was found to be the most sensitive assay, and the TNF-α enzyme immunoassay (EIA) was the least sensitive. Specifically, TNF-α protein was only detect-able by EIA, at levels near the lower limit of detection (LLD) of the assay (8 pg/ml) following stimulation with IL-1β, but not P2C or P3C. This was believed to reflect the sensitivity of the EIA relative to the RT-PCR or reporter assays. The potency of IL-1β was approxi-mately 10-fold higher than that of P2C, as measured by RT-qPCR, and by extrapolation of TNF-α production following P2C stimulation was predicted to fall below the LLD of the EIA.

Therefore, although hepatocyte cell lines were dem-onstrated to initiate a pro-inflammatory signalling cas-cade in response to TLR2 ligand stimulation, the physi-ological significance of these pathways was uncertain, given the absence or only very low level of detectable cytokine production at the protein level. Thus, the aim of this study was to investigate the potential antiviral effects of TLR2-specific signalling, within two human hepatoma cell lines, on HBV replication, in the pres-ence or absence of HBeAg production.

Methods

Cell culture, transductions, transfections and ligand stimulationTwo cell lines were examined, namely the HepG2 (ATCC HepG2) and Huh-7 (Huh-7-SL) cell lines. HepG2 cell lines were maintained in minimum essential medium (Invitrogen, Carlsbad, CA, USA) supplemented with 10% heat-inactivated fetal calf serum, penicillin/ streptomycin and L-glutamine in a humidified 37°C, 5% CO2 incubator. Replication- competent genotype D 1.3 unit length recombinant HBV–BCV (rHBV–BCV; see HBV infectious clones), was generated and puri-fied as previously described [10,15]. HepG2 cells were seeded into 60 mm plates (Becton Dickinson, San Jose, CA, USA). After overnight (16 h) incubation, the cells were transduced with rHBV–BCV at a multiplicity of infection (MOI) of 50 plaque-forming units (PFU) per

Locarnini.indd 798 22/9/09 10:28:45

Stimulation of the IL-1 receptor and TLR2 inhibits HBV replication

Antiviral Therapy 14.6 799

cell [15]. Immediately after transduction, fresh medium was added. Cells were stimulated with particular treat-ments (control/antiviral drug/cytokine/TLR ligand) for 6, 24, 48 or 72 h prior to harvest. Each treatment was diluted in fresh medium, and added at the time of media change, rather than being added to each plate individually, to minimize dose variation. Consecutive passages of the same HepG2 subclone were used for each experimental set, to minimize experimental varia-tion. Cell culture media was changed on days 1, 3 and 4 post-transduction and cultures harvested on day 5 according to protocol.

The Huh-7-SL cell line (kindly provided by Stan Lemon, UTMB, Galveston, Texas) was maintained in Dulbecco’s modified Eagle’s medium (Gibco-BRL, Grand Island, NY, USA) supplemented with 10% heat-inactivated fetal calf serum, penicillin/streptomycin in a humidified 37°C, 5% CO2 incubator [14]. For transfec-tion, cells were seeded into 60 mm dishes and allowed to adhere overnight. On the following day, when the cells were 60–80% confluent, the culture medium was replaced with fresh medium. Sets of replicate cultures were transiently transfected [16] with 5 µg of each HBV construct by using 8 µl of FuGENE 6 transfec-tion reagent (Roche Diagnostics, Mannheim, Ger-many), according to the manufacturer’s instructions. The culture medium was changed on day 2 or day 3, and the cells were harvested on day 5 post-transfection. Cell viability was monitored using the MTT assay as described previously [15,17].

Plasmids and cytokinesThe small interfering RNA (siRNA) for TLR2 express-ing plasmid (psiRNA) was purchased from InVivoGen (San Diego, CA, USA). IL-1β and TNF-α were purchased from Roche (Indianapolis, IN, USA). The TLR2 ligand P2C was purchased from InVivoGen. 3TC was pur-chased from Moravek (Brea, CA, USA). Human IFN-α was obtained from Sigma (Clayton, VIC, Australia).

HBV infectious clonesA replication competent plasmid containing a 1.3× genome-length HBV construct (genotype D, subtype ayw) was cloned into the multiple cloning region of BCV transfer vector pBlueBac4.5 (Invitrogen) to create a new plasmid, designated the wild-type parent (pWT) [10,17]. Two additional HBV transfer plasmids, each containing different mutations, were generated from pWT by site-directed mutagenesis using a Quik-Change Site-Directed Mutagenesis kit (Stratagene, La Jolla, CA, USA) as described by Chen et al. [15] and included the PC stop codon (G1896A) and the double basic core promoter (BCP) mutant, A1762T, G1764A [15]. Both HBV mutants are replication com-petent in vitro [15]. Transduction of HepG2 cells with

the double BCP mutant resulted in a reduced titre of HBeAg in the cell cultures supernatant (50 Paul Ehr-lich International Units (PEIU)/ml versus 150 PEIU/ml for wild-type [WT] HBV) whilst the PC (G1896A PC stop codon mutant) HBV mutant produces no HBeAg at all.

Analysis of replicative intermediatesHBV core particles were isolated from HepG2 and Huh-7 cells as described previously [9,10]. Briefly, after removing the supernatant, cells were washed with phosphate-buffered saline (PBS) and then lysed with 800 µl of 0.5% Nonidet P-40 in PBS for 20 min at 4°C. Cell lysates were then transferred to microfuge tubes and spun at 13,000 rpm for 5 min to remove the nuclei. Unprotected DNA was removed by incubating the sam-ples with 20 U of RNase-free DNase I (Roche Diagnos-tics, Branchburg, NJ, USA) for 1 h at 37°C. The incu-bation mixture was then adjusted to 10 mM EDTA, 1% SDS and 100 mM NaCl. Proteinase K was then added to the samples at a final concentration of 0.5 mg/ml and incubated for a further 4 h at 37°C. Replicative intermediates were then isolated following sequential phenol/chloroform extractions and isopropanol pre-cipitation as described previously [9,10]. Precipitated nucleic acids were resuspended in 5 mM EDTA and digested with 20 U of DNase-free RNase I (Roche Diag-nostics, Branchburg, NJ, USA) for 1 h at 37°C [18]. Replicative intermediates were then analysed by elec-trophoresis, Southern blotting and autoradiography as described previously [9,15,19].

Analysis of secreted hepatitis B e antigenMedium from HepG2 cells was collected, centrifuged at 3,000 rpm to remove cellular debris and transferred to clean tubes. The amount of HBeAg secreted into the cell culture medium was then qualitatively determined using a commercially available EIA (Abbott Labora-tories, Chicago, IL, USA) and quantitatively using the Abbott Architect platform with cross-reference to the Paul Ehrlich HBeAg reference standard (Paul Ehrlich Institute, Langen, Germany) so that the amount of HBeAg was expressed in PEIU/ml similarly as described by Fried et al. [20].

Analysis of hepatitis B core antigen and nucleocapsid productionHepatitis B core antigen (HBcAg) was detected in cells by immunoblot using anti-core antibodies pur-chased from Abcam (Cambridge, MA, USA; mouse monoclonal) and Dako (Glostrup, Denmark; rabbit- polyclonal), followed by an anti-rabbit or anti-mouse horseradish peroxidase secondary antibody. Protein bands were visualised using chemiluminescence. Two measures of HBcAg protein were determined: total

Locarnini.indd 799 22/9/09 10:28:45

AJ Thompson et al.

©2009 International Medical Press800

intracellular HBcAg protein production and HBV nucleocapsid formation.

Hepatitis B core protein detectionCells were harvested after 5–7 days following trans-duction/transfection by dissolving in a 2% SDS lysis buffer (fully denaturing) containing complete protein inhibitors and mixed with 5× Laemmli loading dye as previously described [9,10,15]. Proteins were then electrophoresed through a 12% polyacrylamide gel, transferred to Hybond-C membrane and detected using either monoclonal or polyclonal antibodies against HBcAg (anti-HBc) as described above in order to measure the total amount of cellular HBcAg pro-tein synthesised.

HBV nucleocapsid detectionNucleocapsid particles were isolated from transfected cells following cell lysis in a buffer containing 0.5% Nonidet P-40, 10 mM Tris pH 7.5, 50 mM NaCl and 1 mM EDTA and the nuclei were removed by cen-trifugation at 13,000 g. These nuclei-free cell lysates were subjected to digestion with 20 U DNase I and were stored at 4°C. A quantity of 50 µg of each lysate was then subjected to electrophoresis through a 1.2% (w/v) non-denaturing agarose gel in Tris-acetate EDTA buffer as described by Melegari et al. [21]. Nucleocapsids were detected by western blot analysis using Dako polyclonal anti-HBc as per standard pro-cedures (see Analysis of hepatitis B core antigen and nucleocapsid production) [9,10,15].

siRNA for TLR2Huh-7 cells were co-transfected with a total of 4 µg of plasmid DNA comprising either pWT (HBV):plasmid empty vector (pEV) in a 2:2 ratio or pWT:pEV:psiRNA in a 2:1:1 ratio using FuGENE 6 Transfection Reagent (Roche Diagnostics, Mannheim, Germany). siRNA vectors obtained from InVivoGen have been validated by the company and guarantee a minimum knock-down of target mRNA>70%; this has been independ-ently confirmed for TLR2 knockdown [22]. Cells were treated for 3 days, with fresh media containing either P2C (1,000 ng/ml), IFN-α (103 IU/ml), IL-1β (10 ng/nl), TNF-α (10 ng/ml) or 3TC (0.01 µM). Nucleo-capsid particles were isolated as described (see HBV nucleocapsid detection). Following DNase I diges-tion, the sample was divided in two, with half being used for nucleocapsid detection and the remaining half being processed for the isolation of replicative intermediates and subsequent detection by Southern blot hybridization as described (see Analysis of repli-cative intermediates). Aliquots of these samples were also harvested for total HBcAg by dissolving the cells in Laemmli buffer and processing for polyacrylamide

gel electrophoresis as described above (see Analysis of replicative intermediates).

Quantitative PCRThe expression of TNF-α mRNAs following ligand stimulation was tested as previously described [11]. In brief, for each of the cell lines, live cells (2×105 per well) were seeded into 6-well plates; 24 h later, cells were stimulated before harvesting at the indi-cated times. Total cellular RNA was extracted using the RN easy MiniKit (Qiagen, Germantown, MD, USA) according to the manufacturer’s instructions and eluted into 50 µl of nuclease-free water. cDNA was synthesized using the ABI High capacity RT kit (Applied Biosystems, Foster City, CA, USA) as per the manufacturer’s instructions. TaqMan real-time PCR was performed with 50 µg of cDNA using the Assays-On-Demand Gene Expression Products (Applied Bio-systems) and an ABI Prism 7000 Sequence Detection System (Applied Biosystems). The relative quantities of mRNA were determined using the comparative cycle threshold (Ct) method, where the amount of target DNA was normalized to the internal control, RPLP0 ribosomal cDNA and expressed relative to the baseline unstimulated control cDNA (2-∆∆Ct). RPLP0 was chosen as the internal control after comparison with 18 S and GAPDH. Probe sets resulting in Ct>35 were not analysed and scored as undetectable.

Enzyme immunoassayTNF-α was measured in cell culture supernatant by cap-ture EIA using the OptEIA Set (BD Biosciences, Fran-klin Lakes, NJ, USA) according to the manufacturer’s specifications. The LLD of the assay was 0.008 ng/ml.

Statistical analysesTwo-tailed Student’s t-tests were used to compare results between quantifiable experiments where possible. An α-value of 0.05 was used to determine deviations between treatments that could not be explained by nor-mal sampling variation alone. Standard deviations are shown as error bars for most experiments presented.

Results

IL-1/TLR2 signalling inhibits HBV replication in hepatoma cells in vitroStimulation with IL-1β inhibited replication of HBVThe effect of MyD88 signalling on HBV replication was firstly investigated using IL-1β, which has a common downstream signalling pathway with TLRs via its Toll/IL-1 receptor domain. IL-1β was chosen as proof-of-principle and positive control, as hepatocytes have pre-viously been shown to posses intact IL-1R and previ-ous reports suggested that, in an alternative cell culture

Locarnini.indd 800 22/9/09 10:28:45

Stimulation of the IL-1 receptor and TLR2 inhibits HBV replication

Antiviral Therapy 14.6 801

model, HBV replication was inhibited in vitro [23]. A timecourse experiment was performed using the dose of IL-1β identified to induce maximal TNF mRNA expres-sion (data not shown). HepG2 cells were transduced with WT rHBV–BCV constructs at an MOI of 50 PFU per cell. All cells were harvested at 4 days post-trans-duction; cells were stimulated with IL-1β (10 ng/ml) for 6, 24, 48 or 72 h prior to harvest. 3TC was used as the positive control at the 50% inhibitory concentration in this system of 0.01 µM [15,24]. IL-1β was found to exert a modest antiviral effect, in which maximal reduc-tion of intracellular single-stranded HBV replicative intermediates was observed after 24–48 h of stimulation (Figure 1). The inhibition of replication was attenuated beyond 48 h, perhaps consistent with the development of tolerance. As expected, the 3TC effect was seen to cumulatively increase throughout the timecourse [25]. Thus, 48 h was chosen as the duration of stimulation with ligands for all subsequent experiments.

Stimulation with the TLR2 ligand P2C inhibited replication of HBVThe rHBV–BCV-HepG2 system was then used to test for an antiviral effect of the TLR2 ligand P2C. HepG2 cells were again transduced with WT rHBV–BCV con-structs at an MOI of 50 PFU per cell. After 2 days of culture, cells were stimulated for 48 h with a range of concentrations of P2C (10, 100 and 1,000 ng/ml; Figure 2A). The cells were also separately treated with TNF-α (10 ng/ml) and 3TC (0.01 µM) as positive con-trols, with TNF-α inhibiting replication via an alter-nate pathway of NF-κB induction [24], and 3TC via HBV polymerase inhibition [15,16]. The results, sum-marized in Figure 2B, show a dose–response inhibi-tion of HBV viral replication using P2C. P2C exerted a moderate inhibition of WT HBV replication, but the maximum concentration of 1,000 ng/ml was less potent than either TNF-α or 3TC as an inhibitor of WT HBV replication.

1Lane 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

RC

Time 6 h

Moc

k

24 h 48 h 72 h

DSL

SS

RCDSL

SS

IL-1

β

3TC

Moc

k

IL-1

β

3TC

Moc

k

IL-1

β

Moc

k

IL-1

β

3TC

1.2

0.8

1

0.6

0.4

0.2

0Mock 6 h

HB

V D

NA

fold

red

uctio

n

24 h

Time course

a

a

a

a

a

48 h

IL-1 5 ng/ml3TC 0.01 µM

72 h

Figure 1. Stimulation with 5 ng/ml IL-1β inhibits wild-type HBV replication using the recombinant HBV–BCV system

(A) Southern blot of hepatitis B virus (HBV) DNA showing that maximal inhibition of replication was observed at 24–48 h of stimulation (lane 20 is molecular weight markers). (B) The graph presents mean densitometry values, calculated on the single-stranded (SS) HBV DNA from three separate experiments performed in duplicate. aSignificant difference from the untreated control (Mock; Mann–Whitney U P<0.05). BCV, baculovirus; DSL, double-stranded linear HBV DNA; IL, interleukin; RC, relaxed circular HBV DNA; 3TC, lamivudine.

A

B

Locarnini.indd 801 22/9/09 10:28:47

AJ Thompson et al.

©2009 International Medical Press802

Testing of HBV mutants defective for HBeAg productionThe experiment was then repeated using viruses defective for HBeAg production: the G1896A PC variant, which does not produce HBeAg, and the A1762T/G1764A BCP variant, which produces approximately 30% of the HBeAg that the WT virus produces. The inhibitory effect of P2C on both viruses was of similar magnitude (Fig-ure 3A to 3D), with the effect on the BCP variant being greater than the PC variant (80% reduction versus 35% reduction at P2C at a dose of 1,000 ng/ml). The P2C antiviral effect on the PC variant (35% inhibition at P2C of 1,000 ng/ml; Figure 3A and 3B) was of a similar order of magnitude to its effect on the WT virus (45% inhibi-tion, at P2C of 1,000 ng/ml; Figure 2A and 2B).

HBeAg secretion was not inhibited by the TLR2 ligand P2CThe cell culture supernatants from the rHBV–BCV experiments were then assayed for HBeAg titre. Despite

the Southern blots showing significant decreases in HBV replicative intermediates, no effect on HBeAg production was evident, for either WT HBV or BCP Mutant HBV (Figure 2C and 3E). This was consistent with an antiviral block to the replication cycle at a post-transcriptional level.

The TLR2 ligand P2C inhibited HBV capsid formation but not the total amount of HBcAg produced in cellsCytokines, including TNF-α, IFN-α/β and IFN-γ have previously been shown to inhibit HBV replication by inhibiting nucleocapsid formation [26–28]. In order to gauge the effect of P2C on HBV nucleocapsid for-mation, treated cells were harvested and processed in either Laemmli buffer (total HBV core protein produc-tion) or nonidet P-40 (NP-40; nucleocapsid formation) and analysed by denaturing or non-denaturing gel elec-trophoresis. Unfortunately, very little core protein was reliably detected following transduction of HBV, likely

1Lane 2

Moc

k

3 4 5 6 7 8 9 10 11 12

P2C

10

ng/m

l

P2C

100

ng/

ml

P2C

1,0

00 n

g/m

l

TNF-

α 10

ng/

ml

3TC

0.0

1 µm

Moc

k

P2C 1

0 ng

/ml

P2C 1

00 n

g/m

l

P2C 1

,000

ng/

ml

TNF-

α 10

ng/m

l

3TC 0

.01

µmM

ock

P2C 1

0 ng

/ml

P2C 1

00 n

g/m

l

P2C 1

,000

ng/

ml

TNF-

α 10

ng/m

l

3TC 0

.01

µm

RCDSLSS

1.21

0.80.60.4

aa

aa

0.20

HB

V D

NA

fold

red

uctio

n

Stimulus/ligand Stimulus/ligand

2.0

1.5

1.0

0.5

0

HB

eAg

fold

red

uctio

n

Figure 2. Stimulation with the TLR2 ligand P2C inhibits wild-type HBV replication using the recombinant HBV–BCV system

(A) Southern blot results of hepatitis B virus (HBV) DNA showing that at a dose of 1,000 ng/ml, up to 45% inhibition of viral replication was achieved. (B) The graph presents mean densitometry values, calculated on the single-stranded (SS) HBV DNA from three separate experiments performed in duplicate. (C) Mean hepatitis B e antigen (HBeAg) concentration in the cell culture supernatant, showing that HBeAg secretion was not reduced despite the potent suppression of core-associated HBV DNA. aSignificant difference from untreated control (Mock; Mann–Whitney U P<0.05). BCV, baculovirus; DSL, double-stranded linear HBV DNA; P2C, Pam-2-Cys; RC, relaxed circular HBV DNA; TLR, Toll-like receptor; TNF, tumour necrosis factor; 3TC, lamivudine.

A

B C

Locarnini.indd 802 22/9/09 10:28:48

Stimulation of the IL-1 receptor and TLR2 inhibits HBV replication

Antiviral Therapy 14.6 803

reflecting the lower replication yield of the rHBV–BCV system relative to transfection systems [9,15,16]. These experiments were then repeated using transient trans-fection of WT HBV-expressing plasmids, in the Huh-7 cell line. This transient transfection system achieves higher levels of HBV replication than the rHBV–BCV model [9,15,16], and core protein could be detected by

immunoblot. TNF-α, IFN-α and 3TC were included as controls. Treatment of the transfected Huh-7 cells with P2C or any of the other cytokines had no effect on the total level of HBV core protein produced in the Huh-7 cells (data not shown). However, when the cells were treated with P2C and the lysates analysed for nucleo-capsid formation, a decrease (≥30% reduction) in the

RCDSLSS

1Lane 2 3 4 5 6 7 8 9 10 11 12

1Lane 2 3 4 5 6 7 8 9 10 11 12

1.21

0.80.60.4

a

a

a

a a

a

a

a

0.20

HB

V D

NA

fold

red

uctio

n

1.21

0.80.60.40.2

0

HB

V D

NA

fold

red

uctio

n

2.0

1.5

1.0

0.5

0

HB

eAg

fold

red

uctio

n

Moc

k

P2C

10

ng/m

l

P2C

100

ng/

ml

P2C

1,0

00 n

g/m

l

TNF-

α 10

ng/

ml

3TC

0.0

1 µm

Moc

k

P2C 1

0 ng

/ml

P2C 1

00 n

g/m

l

P2C 1

,000

ng/

ml

TNF-

α 10

ng/m

l

3TC 0

.01

µm

Moc

k

P2C

10

ng/m

l

P2C

100

ng/

ml

P2C

1,0

00 n

g/m

l

TNF-

α 10

ng/

ml

3TC

0.0

1 µm

Moc

k

P2C 1

0 ng

/ml

P2C 1

00 n

g/m

l

Stimulus/ligand Stimulus/ligand

P2C 1

,000

ng/

ml

TNF-

α 10

ng/m

l

3TC 0

.01

µmM

ock

P2C 1

0 ng

/ml

P2C 1

00 n

g/m

l

P2C 1

,000

ng/

ml

TNF-

α 10

ng/m

l

3TC 0

.01

µm

Figure 3. Stimulation with the TLR2 ligand P2C also inhibits HBeAg-defective HBV replication using the recombinant HBV–BCV system

(A) Southern blot analysis of hepatitis B virus (HBV) DNA showing that Pam-2-Cys (P2C) inhibits replication of the G1896A precore mutant HBV to a level of approximately 35% at a dose of 1,000 ng/ml. (B) The graph presents mean densitometry values, calculated on the single-stranded (SS) HBV DNA from three separate experiments performed in duplicate. (C) P2C inhibits replication of the A1762T/G1764A basal core promoter variant HBV at a much greater level of approximately 80% at a dose of 1,000 ng/ml. (D) The graph presents mean densitometry values, calculated on the SS HBV DNA from three separate experiments performed in duplicate. (E) Mean hepatitis B e antigen (HBeAg) concentration in the cell culture supernatant, showing that HBeAg secretion was not reduced despite the potent suppression of core-associated HBV DNA. aSignificant difference from untreated control (Mock; Mann–Whitney U P<0.05). BCV, baculovirus; DSL, double-stranded linear HBV DNA; RC, relaxed circular HBV DNA; TNF, tumour necrosis factor; 3TC, lamivudine.

A B

C

De

Locarnini.indd 803 22/9/09 10:28:50

AJ Thompson et al.

©2009 International Medical Press804

Figure 4. Stimulation with the TLR2 ligand P2C inhibits HBV capsid formation

Huh7 cells were transfected with a wild-type (WT) hepatitis B virus (HBV) construct and simulated 24 h post-transfection with 100 ng/ml Pam-2-Cys (P2C), 1,000 IU/ml interferon (IFN)-α, 10 ng/ml interleukin (IL)-1β, 10 ng/ml tumour necrosis factor (TNF)-α, 0.01 µM lamivudine (3TC) or untreated control (Mock). Cells were treated daily for 3 days with fresh media containing the stimulus of interest. Cells were harvested on day 4 post-transfection. NP-40 lysates were electrophoresed through a native agarose (non-denaturing) gel, before probing using antibodies against hepatitis B core antigen (Dako, Glostrup, Denmark), as described in the Methods. (A) A representative western blot. (B) The summary results of two experiments performed in duplicate. aSignificant difference from Mock (Mann–Whitney U P<0.05).

Nucleocapsid

P2C100 ng/ml

IFN-α1,000 IU/ml

IL-1β10 ng/ml

TNF-α10 ng/ml

3TC0.01 µM Mock

1.2

0.8

1

0.6

0.2

0.4

0P2C

a

a

a

IFN-α IL-1β TNF-α 3TC Mock

HB

V n

ucle

ocap

sid

form

atio

nfo

ld c

hang

e

A

B

-

+

Core protein

P2C IFN-α TNF-α 3TC Mock

1.41.21.00.80.60.40.20.0

P2C- +

IFN-α- +

TNF-α- +

3TC- +

Mock- +

HB

V c

ore

pro

tein

fold

cha

nge

P2C IFN-α TNF-α 3TC Mock

-

+Nucleocapsids

1.2

1

0.8

0.6

0.4

0.2

0

P2C

pWT:pEV

pWT:psiRNA

- +IFN-α- +

TNF-α- +

3TC- +

Mock- +

HB

V n

ucle

ocap

sid

form

atio

nfo

ld c

hang

e

Figure 5. The antiviral effect of P2C was blocked in the presence of TLR2 siRNA

Huh-7 cells were transfected with either wild type parent (pWT):plasmid empty vector (pEV) or pWT:plasmid expressing small interfering RNA for Toll-like receptor 2 (psiRNA) and then treated for 3 days as described in the Methods. (A) Fully denaturing western blot of the cell lysates demonstrating the total level of the hepatitis B virus (HBV) core protein production. (B) Non-denaturing western blot results from the same cell lysates showing blocking of the antiviral effect of siRNA-Toll-like receptor (TLR)2 on HBV nucleocapsid formation. (C) Graphs represent mean densitometry values of replicate samples, calculated on the p25 core protein (A) and the nucleocapsid band (B). Values are expressed as the fold change between untreated (Mock) and treated cultures. In the presence of siRNA for TLR2, the inhibitory effect of Pam-2-Cys (P2C) is markedly reduced. IFN, interferon; TNF, tumour necrosis factor; 3TC, lamivudine; +, transfection with pWT:psiRNA, that is, transfection in the presence of siRNA to TLR2; -, transfection with pWT:pEV.

A B

Locarnini.indd 804 22/9/09 10:28:52

Stimulation of the IL-1 receptor and TLR2 inhibits HBV replication

Antiviral Therapy 14.6 805

levels of viral capsids was observed (Figure 4). These findings were reproducible. A post-transcriptional antiviral effect of TNF-α and IFN-α on nucleocapsid formation was also observed, whereas 3TC was shown to have no effect, as previously described [15,24] (Figure 4).

The antiviral effect of P2C was blocked in the presence of siRNA for TLR2In order to demonstrate the specificity of the antiviral effect on HBV nucleocapsid assembly following TLR2 stimulation, Huh-7 cells were again cotransfected with HBV DNA plus an expression vector of siRNA for TLR2 (psiRNA). The same experimental protocol described in Methods used and the NP-40 lysates from the cells were prepared, harvested and processed for HBV core (fully denaturing immunoblot; Figure 5A) levels as well as nucleocapsid (non-denaturing immuno-blot; Figure 5B) formation. Figure 5A shows that there was no difference (that is, <30% reduction) in the pres-ence or absence of siRNA for TLR2 for the amount of core protein produced following any of the treatments (Figure 5A). However, in the presence of the siRNA TLR2, under non-denaturing conditions (Figure 5B), the antiviral effect of TLR2 stimulation on nucleocap-sid production was abolished.

The antiviral effect of P2C was not caused by the secondary induction of TNF-α in HepG2 or Huh-7 cellsTo exclude the possibility that secondary TNF-α induc-tion might be responsible for the inhibition of viral replication observed following TLR2 stimulation, the HepG2 and Huh-7 cell lines were exposed to increasing doses of TNF-α. The output measured was the induc-tion of TNF-α mRNA itself, as measured by RT-PCR, as TNF-α had been previously shown to induce expres-sion of TNF-α mRNA via the TNF receptor (positive feedback) [11], and in this cell culture system induction of TNF-α was the most sensitive marker for activation of NF-κB signalling. The dose range was chosen on the basis of the concentration of TNF-α protein detected by EIA in the cell culture supernatant in the initial experi-ments, when maximal doses of IL-1β resulted in a peak concentration of 0.008 ng/ml TNF-α in the cell culture supernatant. Induction of TNF-α mRNA was detect-able only at doses of ≥0.10 ng/ml TNF-α in both cell lines (Figure 6A for HepG2 and Figure 6B for Huh-7 cells), indicating that secondary induction of TNF-α was probably not responsible for the antiviral effect observed in either cell culture system.

The antiviral effect of P2C was not caused by cytotoxicityBecause cytokine treatment can cause toxicity directly and thereby affect viral replication, three parallel sets

of HepG2 cells were continuously exposed to P2C (10, 100 and 1,000 ng/ml), IL-1β (0.01, 0.1, 1 and 10 ng/ml) and TNF-α (10 ng/ml) for 7 days. The cell monolayers remained intact for the entire dura-tion of the assay, and daily microscopic examination revealed no evidence of cytotoxicity. The MTT uptake by viable cells was measured by spectrophotometry as described in the Methods. The agents were not cytotoxic at the concentrations used, confirming that the observed inhibitory effect of on HBV replication was not an artefact caused by cytotoxicity (data not shown). Similar results were found with the Huh-7 cells (data not shown).

100.00

80.00

60.00

40.00

20.00

010

TNF-α, ng/ml

HepG2

a

a

a

a

a

a

a

a

TNF-

α m

RN

A fo

ld c

hang

e

10.10.010.0010

700

600

500

400

300

200

100

010

TNF-α, ng/ml

Huh-7

TNF-

α m

RN

A fo

ld c

hang

e

10.10.010.0010

Figure 6. Secondary induction of TNF-α mRNA was only induced by TNF-α at doses ≥0.10 ng/ml

(A) ATCC HepG2 cells were stimulated with tumour necrosis factor (TNF)-α over a dose range of 0.001–10 ng/ml. Cells were harvested for real-time PCR 60 min post-stimulation; the output was fold change in TNF-α messenger RNA (mRNA) relative to mock-stimulated cells. (B) A similar experiment to (A), but with Huh-7 cells seeded into 6-well plates at 2×105 cells per well. After 24 h, total cellular RNA was extracted using RNeasy MiniKit (Qiagen, Germantown, MD, USA), following stimulation for 1 h with increasing 10-fold concentrations of TNF-α (0.001–10 ng/ml). The relative quantity of TNF-α mRNA was determined and expressed relative to the baseline unstimulated control. aSignificant differences from mock-treated control (Mann–Whitney U P<0.05).

A

B

Locarnini.indd 805 22/9/09 10:28:52

AJ Thompson et al.

©2009 International Medical Press806

Discussion

The HepG2 cells and the Huh-7-SL cell lines used in this study have been previously shown by our group to respond to the TLR2 ligand P2C in functional assays [11]. The finding of functional TLR2 expression on HepG2 and Huh-7-SL cells provided the opportunity to examine human cell culture models that would allow testing of the direct effect of TLR2 signalling on the replication of HBV in vitro. In the rHBV–BCV model, 48 h of stimulation with P2C significantly inhibited WT HBV DNA replication. A dose-dependent effect was observed, with core-associated HBV DNA replication inhibited up to 90% at maximal dose. In the transient transfection system, nucleocapsid formation was also significantly reduced. The HBeAg defective strains of HBV were also similarly inhibited by P2C.

Antiviral cytokines have been shown to inhibit viral replication post-transcriptionally. In the case of TNF-α, this can occur via destabilization of HBV capsid for-mation, a crucial step in the viral replication pathway [26]. Type-1 IFNs have also been shown to interfere with capsid formation [28]. The observation in the rHBV–BVC HepG2 system, that HBV DNA replica-tion was inhibited, yet HBeAg secretion was preserved, suggested that the TLR2 anti-HBV effect was acting at the post-transcriptional level. Post-transcriptional viral replication events were therefore investigated. This was not possible in the HepG2 cells, as the level of core protein generated by the rHBV–BCV model was below the threshold of sensitivity for western blot. Using the transient transfection model in Huh-7 cells, TLR2 stimulation was shown to reduce HBV nucleo-capsid formation, but not the total amount of HBcAg produced by these cells. The inhibitory effect attribut-able to P2C was comparable to the known anti-HBV cytokine TNF-α [24], but less effective than the potent nucleoside analogue 3TC. That the antiviral hierarchy observed in the rHBV–HCV HepG2 system (TNF-α>P2C) was not observed in the nucleocapsid assay was interesting. Although this might have related to differences in TNF-α-stimulated signalling pathways between the two cell lines, it was felt more likely that this reflected differences between the models. In the Huh-7 transfection model, HBV replication was driven by the powerful cytomegalovirus promoter rather than endogenous promoters; in addition, the nucleocapsid assay was less sensitive than the Southern blot for core-associated HBV DNA. As a result, a more subtle dis-tinction between the P2C and TNF-α antiviral effect was likely not possible. Collectively, the data suggests that activation of the TLR2 signalling pathway leads to reduced nucleocapsid formation and pregenomic RNA packaging, thus blocking subsequent reverse transcrip-tion and DNA replication [29].

The mechanism whereby activation of TLR2 inhibits HBV replication remains unresolved. Whether this is via direct targeting of DNA replication by a TLR2 signal-ling intermediate or the induction of a TLR2 dependent gene is yet to be determined. A direct role for second-ary cytokine induction is possible, but unlikely from the data presented. TNF-α is a canonical pro- inflammatory cytokine pathway induced by the MyD88- dependent signalling pathway, and is known to inhibit HBV in vitro [24]. However, the finding that no TNF-α was detectable in either the cell culture supernatant or cel-lular lysates following stimulation with P2C suggests that a novel antiviral cytokine might be involved. Fur-thermore, when the HepG2 cells were stimulated with TNF-α, no pro-inflammatory response was seen until the doses used were at least 10-fold higher than that observed in cell culture supernatant following high-dose IL-1β stimulation. It is therefore reasonable to conclude that the antiviral effect of P2C, a weaker pro-inflammatory stimulus than IL-1β, was unlikely to have been mediated by secondary TNF-α induction. We have previously shown that IL-8 expression is also induced by TLR2 stimulation in these cell lines [11]. Although IL-8, a chemokine, is likely to play a role in the immu-nopathogenesis of HBV infection in vivo, there is no literature to support a direct anti-HBV effect of IL-8 on hepatocytes. Furthermore, we have previously shown in these cell lines that the fold-increase in IL-8 gene expression observed in response to P2C stimulation was less than that for TNF-α [11], which itself was below the threshold for antiviral effect. Therefore, we believe that IL-8 is an unlikely candidate as a direct antiviral cytokine. As no type-1 IFN response occurs following TLR2 stimulation in these cell lines, the possibility of a novel antiviral mechanism is worthy of further investi-gation. It was notable that IL-1β, in contrast to P2C, did not significantly reduce nucleocapsid accumulation, consistent with distinct antiviral mechanisms, and sug-gesting distinct intracellular signalling downstream of TLR2 and IL-1R, despite sharing the canonical MyD88-dependent pathway of NF-κB induction. The molecular mechanism responsible for the antiviral effect of TLR2 therefore requires further investigation.

This is the first study that demonstrates a direct role for hepatocyte TLR signalling as an anti-HBV mechanism in human hepatoma cell lines. Previ-ous work using the HBeAg-positive HBV-transgenic mouse model has identified only an indirect role for liver non-parenchymal cell TLR signalling as an anti-viral mediator via induction of type-1 IFNs [30]. TLR ligand stimulation of the HBV–Met cell line, an immortalized mouse hepatocyte cell line that con-tains the HBV transgene, did not have any antiviral efficacy [30,31]. The conclusion was drawn that, at least in these mouse models, hepatocyte TLRs were

Locarnini.indd 806 22/9/09 10:28:53

Stimulation of the IL-1 receptor and TLR2 inhibits HBV replication

Antiviral Therapy 14.6 807

not functional. There might therefore be interspecies variation in the expression and functional signifi-cance of TLRs. It should also be noted that the HBV transgenic mouse is a fundamentally different model of HBV replication, in which HBV is transcribed off the HBV transgene, cccDNA is not produced, and the intracellular milieu of transcription and translation factors substantially differs from the hepatoma cell culture models used in this study.

We have previously observed that patients with HBeAg-negative CHB display increased TLR2 expres-sion and function on peripheral blood mononuclear cells, Kupffer cells and hepatocytes, and decreased lev-els of viral replication, relative to patients with HBeAg-positive CHB [8] We interpret this as suggesting that the production of the HBeAg represents an important immune evasion mechanism in vivo. In the in vitro mod-els studied here, then, we were surprised to observe that the antiviral effect of P2C on HBV replication, was simi-lar for all three viral constructs, including the PC and BCP variant viruses which are associated with either an absolute or relative decrease in HBeAg production, respectively. Greater potency of antiviral effect was not observed of the TLR2 ligand effect in vitro. Several pos-sibilities might explain this discrepancy. First, it might be that HBeAg does play a direct immunoevasive role in vivo, but that the target is not the hepatocyte. The peripheral monocyte is a potential candidate, consistent with the inhibitory effect of PC conditioned media on monocyte TLR2 signalling previously described [8]. Sec-ond, it is possible that the HBeAg PC effect occurs only in the absence of other potentially immunostimulatory HBV proteins, such as HBsAg, as occurs in the extracel-lular environment.

In conclusion, we have shown that hepatocytes can play an important innate immune surveillance role directly within the liver itself and can no longer be con-sidered as merely passive bystander cells. The finding of an antiviral effect in vitro provides support for the clinical observation that TLR2 plays an important role in the pathogenesis of CHB. It is also proof of principle that TLR ligands might have therapeutic potential. It will now be possible to interrogate different cell cul-ture models in more detail for the molecular basis of the observed antiviral effect. These cell culture models should now prove a valuable tool for the further devel-opment of TLR ligands as novel antiviral agents.

Acknowledgements

This work was supported in part from a National Health and Medical Research Council of Australia (NHMRC) postgraduate scholarship (Alex Thompson), the NHMRC Grant number 491017 and the Bristol–Myers Squibb Freedom To Discover Award to SL.

The authors wish to thank Ms Tania Candy for edi-torial assistance.

Disclosure statement

The authors declare no competing interests.

References1. Locarnini S, Mason WS. Cellular and virological

mechanisms of HBV drug resistance. J Hepatol 2006; 44:422–431.

2. World Health Organization. Hepatitis B: fact sheet. (Accessed 30 March 2009.) Available from http://www.who.int/mediacentre/factsheets/fs204/en/index.html

3. Chisari FV, Ferrari C. Hepatitis B virus immunopathogenesis. Annu Rev Immunol 1995; 13:29–60.

4. Lok AS. Chronic hepatitis B. N Engl J Med 2002; 346:1682–1683.

5. Akira S, Uematsu S, Takeuchi O. Pathogen recognition and innate immunity. Cell 2006; 124:783–801.

6. Takeda K, Akira S. Toll-like receptors in innate immunity. Int Immunol 2005; 17:1–14.

7. Takeda K, Kaisho T, Akira S. Toll-like receptors. Annu Rev Immunol 2003; 21:335–376.

8. Visvanathan K, Skinner NA, Thompson AJ, et al. Regulation of Toll-like receptor-2 expression in chronic hepatitis B by the precore protein. Hepatology 2007; 45:102–110.

9. Delaney WE IV, Isom HC. In vitro model systems for studying hepatitis B virus replication. In Lai CL, Locarnini S (Editors). Hepatitis B virus. 1st ed. London: International Medical Press 2002; pp. 55–80.

10. Delaney WE, Isom HC. Hepatitis B virus replication in human HepG2 cells mediated by hepatitis B virus recombinant baculovirus. Hepatology 1998; 28:1134–1146.

11. Preiss S, Thompson A, Chen X, et al. Characterization of the innate immune signalling pathways in hepatocyte cell lines. J Viral Hepat 2008; 15:888–900.

12. Liu S, Gallo DJ, Green AM, et al. Role of toll-like receptors in changes in gene expression and NF-kappa B activation in mouse hepatocytes stimulated with lipopolysaccharide. Infect Immun 2002; 70:3433–3442.

13. Nishimura M, Naito S. Tissue-specific mRNA expression profiles of human toll-like receptors and related genes. Biol Pharm Bull 2005; 28:886–892.

14. Li K, Chen Z, Kato N, Gale M, Jr, Lemon SM. Distinct poly(I-C) and virus-activated signaling pathways leading to interferon-beta production in hepatocytes. J Biol Chem 2005; 280:16739–16747.

15. Chen RY, Edwards R, Shaw T, et al. Effect of the G1896A precore mutation on drug sensitivity and replication yield of lamivudine-resistant HBV in vitro. Hepatology 2003; 37:27–35.

16. Chin R, Shaw T, Torresi J, et al. In vitro susceptibilities of wild-type or drug-resistant hepatitis B virus to (-)-beta-d-2,6-diaminopurine dioxolane and 2′-fluoro-5-methyl-beta-l-arabinofuranosyluracil. Antimicrob Agents Chaemother 2001; 45:2495–2501.

17. Delaney WE. Edwards R, Colledge D, et al. Cross-resistance testing of antihepadnaviral compounds using novel recombinant baculoviruses which encode drug-resistant strains of hepatitis B virus. Antimicrob Agents Chemother 2001; 45:1705–1713.

18. Sambrook J, Fritsch EF, Maniatis T. Molecular cloning: a laboratory manual. 1st ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press 1989.

Locarnini.indd 807 22/9/09 10:28:53

AJ Thompson et al.

©2009 International Medical Press808

19. Warner N, Locarnini S, Kuiper M, et al. The L80I substitution in the reverse transcriptase domain of the hepatitis B virus polymerase is associated with lamivudine resistance and enhanced viral replication in vitro. Antimicrob Agents Chemother 2007; 51:2285–2292.

20. Fried MW, Piratvisuth T, Lau GK, et al. HBeAg and hepatitis B virus DNA as outcome predictors during therapy with peginterferon alfa-2a for HBeAg-positive chronic hepatitis B. Hepatology 2008; 47:428–434.

21. Melegari M, Wolf SK, Schneider RJ. Hepatitis B virus DNA replication is coordinated by core protein serine phosphorylation and HBx expression. J Virol 2005; 79:9810–9820.

22. Banerjee P, Biswas A, Biswas T. Porin-incorporated liposome induces Toll-like receptors 2- and 6-dependent maturation and type 1 response of dendritic cell. Int Immunol 2008; 20:1551–1563.

23. Togashi H, Ohno S, Matsuo T, et al. Interferon-gamma, tumor necrosis factor-alpha, and interleukin 1-beta suppress the replication of hepatitis B virus through oxidative stress. Res Commun Mol Pathol Pharmacol 2000; 107:407–417.

24. Biermer M, Puro R, Schneider RJ. Tumor necrosis factor alpha inhibition of hepatitis B virus replication involves disruption of capsid Integrity through activation of NF-kappaB. J Virol 2003; 77:4033–4042.

25. Delaney WE, Miller TG, Isom HC. Use of the hepatitis B virus recombinant baculovirus-HepG2 system to study the effects of (-)-beta-2′,3′-dideoxy-3′-thiacytidine on replication of hepatitis B virus and accumulation of covalently closed circular DNA. Antimicrob Agents Chemother 1999; 43:2017–2026.

26. Puro R, Schneider RJ. Tumor necrosis factor activates a conserved innate antiviral response to hepatitis B virus that destabilizes nucleocapsids and reduces nuclear viral DNA. J Virol 2007; 81:7351–7362.

27. Robek MD, Boyd BS, Wieland SF, Chisari FV. Signal transduction pathways that inhibit hepatitis B virus replication. Proc Natl Acad Sci U S A 2004; 101:1743–1747.

28. Wieland SF, Eustaquio A, Whitten-Bauer C, Boyd B, Chisari FV. Interferon prevents formation of replication-competent hepatitis B virus RNA-containing nucleocapsids. Proc Natl Acad Sci U S A 2005; 102:9913–9917.

29. Feld JJ, Colledge D, Sozzi V, Edwards R, Littlejohn M, Locarnini SA. The phenylpropenamide derivative AT-130 blocks HBV replication at the level of viral RNA packaging. Antiviral Res 2007; 76:168–177.

30. Isogawa M, Robek MD, Furuichi Y, Chisari FV. Toll-like receptor signalling inhibits hepatitis B virus replication in vivo. J Virol 2005; 79:7269–7272.

31. Wu J, Lu M, Meng Z, et al. Toll-like receptor-mediated control of HBV replication by nonparenchymal liver cells in mice. Hepatology 2007; 46:1769–1778.

Accepted for publication 4 May 2009

Locarnini.indd 808 22/9/09 10:28:53