Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Li, K. Articles by Tong, S. Search for Related Content PubMed PubMed Citation Articles by Li, K. Articles by Tong, S.

 Previous Article  |  Next Article 

Journal of Virology, September 2007, p. 9202-9215, Vol. 81, No. 17
0022-538X/07/$08.00+0     doi:10.1128/JVI.00390-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Critical Role of the 36-Nucleotide Insertion in Hepatitis B Virus Genotype G in Core Protein Expression, Genome Replication, and Virion Secretion

Ke Li,¹ Fabien Zoulim,^2,3* Christian Pichoud,² Karen Kwei,¹ Stéphanie Villet,^2,3 Jack Wands,¹ Jisu Li,¹ and Shuping Tong^1*

Liver Research Center, Rhode Island Hospital, Brown University, Providence, Rhode Island,¹ INSERM, U871, 69003 Lyon, France,² Université Lyon 1, IFR62 Lyon-Est, 69008 Lyon, France³

Received 23 February 2007/ Accepted 31 May 2007

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Frequent coinfection of hepatitis B virus genotype G with genotype A suggests that genotype G may require genotype A for replication or transmission. In this regard, genotype G is unique in having a 12-amino-acid extension in the core protein due to a 36-nucleotide insertion near the core gene translation initiation codon. The insertion alters base pairing in the lower stem of the pregenome encapsidation signal, which harbors the core gene initiator, and thus has the potential to affect both core protein translation and pregenomic RNA encapsidation. Genotype G is also unusual for possessing two nonsense mutations in the precore region, which together with the core gene encode a secreted nonstructural protein called hepatitis B e antigen (HBeAg). We found that genotype G clones were indeed incapable of HBeAg expression but were competent in RNA transcription, genome replication, and virion secretion. Interestingly, the 36-nucleotide insertion markedly increased the level of core protein, which was achieved at the level of protein translation but did not involve alteration in the mRNA level. Consequently, the variant core protein was readily detectable in patient blood. The 12-amino-acid insertion also enhanced the genome maturity of secreted virus particles, possibly through less efficient envelopment of core particles. Cotransfection of genotypes G and A did not lead to mutual interference of genome replication or virion secretion. Considering that HBeAg is an immunotolerogen required for the establishment of persistent infection, its lack of expression rather than a replication defect could be the primary determinant for the rare occurrence of genotype G monoinfection.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Hepatitis B virus (HBV) can be classified into eight genotypes with distinct geographic distributions, target populations, and modes of transmission (15, 19, 29). Genotype G was first recognized in French patients in 2000, although it had been described in earlier literature as a viral variant (4, 38, 42). This genotype is unique in that it is frequently detected in homosexual men, who may suffer from immune suppression due to infection with human immunodeficiency virus (4, 26, 44). At the molecular level, different isolates of genotype G display remarkable sequence conservation (>99%) (21). They harbor the A1762T/G1764A mutations in the core promoter region, which for other genotypes do not arise until the late stage of chronic infection (3, 8, 31, 33). Importantly, all genotype G clones have a 36-nucleotide (nt) insertion not found in any other HBV genotypes. This insertion, at the 5' end of the core gene, adds 12 amino acids (aa) to the core protein immediately following the initiating methionine: DRTTLPYGLFGL. At the RNA level, the insertion is located close to a hairpin structure at the 5' end of the pregenomic (pg) RNA called the encapsidation () signal, which directs the pg RNA into nascent core particles for initiation of DNA replication (Fig. 1A). In fact, the first 3 nt inserted alter base pairing at the lower stem of the signal (Fig. 1A) and thus could potentially affect the efficiency of pg RNA encapsidation. In addition to serving as the genome precursor, the pg RNA, prior to its encapsidation, functions as mRNA for the translation of the core protein, the building block for the core particle, as well as DNA polymerase, the enzyme responsible for the conversion of pg RNA into double-stranded DNA. In this regard, the core gene AUG initiator is located at the lower stem of the signal. Since the RNA secondary structure can impede translation initiation, alteration of base pairing by the 36-nt insertion has the potential to alter the efficiency of core protein translation.

View larger version (14K): [in this window] [in a new window]

FIG. 1. (A) Predicted secondary structure of the signal for genotypes A and G. The core gene translation initiation codon and the 5' end of the HBV sequence in the CMV-core constructs (see panel B) are indicated. Also shown for genotype G are the first 3 nt of the 36-nt insertion, the double nucleotide change at the loop that prevents pg RNA encapsidation (– mutant), and a G to A change of the initiation codon that abolishes core protein translation (core– mutant). (B) HBV genes, mRNAs, and constructs used for the present study. pc, precore; pg, pregenomic; sg, subgenomic; pol, polymerase; L, M, S, large, middle, and small envelope proteins, respectively. There are three in-frame translation initiation sites in the pre-S/S gene, leading to the expression of L protein from the 2.4-kb sg RNA and M and S proteins from the 2.1-kb sg RNA. The two initiation sites in the precore/core gene lead to the expression of HBeAg from pc RNA and core protein/polymerase from the slightly shorter pg RNA. For the same reason, CMV-core constructs produce core protein, whereas CMV-precore constructs generate HBeAg. The 1.1-mer constructs cannot express HBeAg, because the 3.5-kb mRNA produced mimics pg RNA, not pc RNA.

The core gene, together with the preceding precore region, codes for the precore/core protein, which is converted to hepatitis B e antigen (HBeAg) following cleavage of the signal peptide and an arginine-rich sequence at the carboxyl terminus (32). Although other genotypes can evolve into HBeAg-negative variants at a later stage of chronic infection, genotype G always contains a defective precore region due to two nonsense mutations. Surprisingly, most genotype G patients are still HBeAg positive. This puzzling observation has prompted careful molecular epidemiological studies, which have revealed frequent coinfection of genotype G with genotype A, the likely source of HBeAg detected in such patient sera (20). Genotype G gradually replaces genotype A as the patients seroconvert to anti-HBe (20, 21, 42). A more recent study revealed coinfection of genotype G with genotype H in Mexico (36), which seems to reinforce the requirement for a coinfecting genotype, whether in the form of genotype A or genotype H.

In the present study, we performed transfection experiments of genotype G in a human hepatoma cell line in order to characterize its genome replication, protein expression, and virion secretion properties. We found that genotype G clones were rather competent in genome replication and virion secretion. They failed to express bona fide HBeAg but produced a much higher level of the variant core protein, which could be detected in patient sera. The 36-nt insertion was indispensable for the replication efficiency of genotype G and was responsible for the high core protein level as well as the increased genome maturity of secreted virions. When cotransfected to cells, genotypes A and G did not significantly interfere with each other in terms of genome replication or virion secretion.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
HBV DNA constructs. Clone D2, containing 1.1 copies (3.5 kb) of genotype D genome under control of the cytomegalovirus (CMV) promoter, was a kind gift of Christoph Seeger, Fox Chase Cancer Institute, Philadelphia, PA. To study HBV replication driven by the strong actin promoter, HBV sequences were amplified from patient sera as two overlapping DNA fragments and cloned into the NotI-XhoI sites of pTriEx vector (Novagen) as 1.1 copies (1.1-mers), as detailed elsewhere (13). Clones G1 to G3 and G4 to G6 of genotype G were derived from two separate patients. The complete nucleotide sequences of clones G1 and G6 were determined by customer sequencing at the Keck Laboratory, Yale University (GenBank accession numbers EF634481 and EF634480, respectively). Clone G1 in the pTriEx vector was used to generate several additional constructs. Replacement of its 2.6-kb NotI-NcoI fragment (nucleotides 1806 to 1375) with the corresponding sequence from Kca1.6, a genotype C clone (D. Gewaily, unpublished data), produced a G1/C1 chimera. This construct expressed genotype C-specific core protein and envelope proteins. Removal of the 36 nt unique to genotype G produced G1d36. The core-negative mutants of G1 and G1d36 were generated by converting the initiation codon to ATA, while the packaging-negative mutants harbored the G1879T/T1880A double mutation in the signal (Fig. 1A) known to abolish genome replication (2). All mutations (substitutions, insertions, and deletions) were created by overlap extension PCR followed by restriction fragment exchange, as previously described (1, 2, 17, 22, 33). To generate an SphI dimer of G1 or G1d36, the full-length HBV genome was reamplified from the pTriEx plasmids with primers 3 and 4, located in the precore region, as previously described (33), except that a single nucleotide change was introduced into primer 4 to accommodate the TAA nonsense mutation at the second codon of the precore region. The DNA was digested with SapI and ligated at a high insert/vector ratio with a modified pUC18 vector containing compatible SapI ends to generate dimeric HBV constructs. The plasmids were further digested with SphI, and the unit-length HBV genome was religated with SphI-cut pUC18 vector to generate SphI dimers. Clones 2A, 4B, 5.4, and 6.2 of genotype A have been used previously as EcoRI dimers (33). In the present study, clone 2A was converted into an SphI dimer for better comparison with SphI dimers of genotype G. Insertion of the 36 nt into clone 4B generated 4Bins36, which was subsequently reconverted to an EcoRI dimer. An env^– mutant, a genotype A mutant defective in virion secretion, was created by a G261A nonsense mutation in the envelope gene of clone N16, as described previously (22). For core protein expression independent of the signal, a 0.6-kb DNA fragment was amplified by the sense primer 5'-AGCACCTCGAGAAGCTGTGCCTTGGGTGGCTTTG-3' (XhoI site underlined) and the antisense primer 5'-AAAGCGAATTCAAGTTTCCCACCTTATGAGTCC-3' (EcoRI site underlined) and cloned into the XhoI-EcoRI sites of pcDNA3.1 zeo(–) vector (Invitrogen). Such constructs direct core protein expression from the CMV promoter. For HBeAg expression under control of the CMV promoter, an upstream sense primer was used to include the entire precore region (5'-CCGAACTCGAGGCATAAATTGGTCTGCGCACCAGCACCATGCAACTTTTTCACCTCTGCC-3' [XhoI site underlined]).

Transfection. Huh7 cells were maintained in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum. Transient transfection was performed on cells seeded in six-well plates or 10-cm dishes, with cells reaching 40 to 80% confluence at the time of transfection. A typical transfection for one well of the six-well plates consisted of 2 µg of HBV DNA, 5 ng of cDNA encoding secreted alkaline phosphatase (SEAP) to serve as control for transfection efficiency, 6 µl of LT1 reagent (Mirus), and 200 µl of serum-free medium. Medium was changed on the second day after transfection, and cells were harvested 4 days later (day 5 posttransfection). For Northern blot analysis, 1.2 µg of HBV construct was cotransfected with 0.8 µg of SEAP cDNA, and cells were harvested at day 3 rather than day 5 posttransfection.

Analysis of HBV DNA replication. HBV DNA was extracted from intracellular core particles for Southern blot analysis (1, 2, 17, 33). Briefly, cells were scrapped off each well of the six-well plates and lysed in 80 µl of 10 mM HEPES (pH 7.5), 100 mM NaCl, 1 mM EDTA, and 1% NP-40. Two-fifths of the cell lysate was treated at 37°C for 15 min with 7.5 U of mung bean nuclease and 0.5 U of DNase I in a total volume of 100 µl, in the presence of 10 mM CaCl₂ and 12 mM MgCl₂. Next, core particles were precipitated by polyethylene glycol solution. After another treatment with nucleases, HBV DNA was released by proteinase K digestion, extracted with phenol, and precipitated with ethanol. DNA was run in 1.2% agarose gels in the presence of ethidium bromide, transferred to GeneScreen Plus membranes (PerkinElmer), and hybridized with ³²P-labeled full-length HBV DNA amplified by primers 3 and 4 (33). For unbiased detection of both genotype A and genotype G, genotype A and genotype G DNA was mixed at 1:1 ratio for labeling. After hybridization, the blots were washed with 2x SSC (1x SSC is 0.15 M NaCl plus 0.015 M sodium citrate)-0.1% sodium dodecyl sulfate (SDS) buffer at 60°C for 2 h. For sequential hybridization of the same blots with different probes, the old probe was removed by boiling the blots for 30 min in 0.1x SSC-1% SDS.

Analysis of virions and naked core particles released to culture supernatant. Viral and subviral particles secreted to culture supernatant were precipitated with a horse polyclonal anti-HBs (Ad/Ay) antibody (Abcam) that had been preconjugated to protein G-agarose beads (Roche) by rotation overnight at 4°C with a ratio of 1.5 µl antibody per 10 µl bed volume of beads. Next, 10 µl bed volume of loaded beads was incubated with 1.5 ml of precleared culture supernatant at 4°C for two overnights with rotation. The beads were brought down by low-speed centrifugation, washed once with 1 ml of phosphate-buffered saline, and spun down again. The immobilized particles were digested at 37°C for 15 min with 1 u of DNase I and 1.5 U of mung bean nuclease in 50 µl of 10 mM Tris (pH 7.5)-6 mM MgCl₂-8 mM CaCl₂ solution, followed by digestion with 0.5 mg/ml of proteinase K. Subsequent to phenol extraction, DNA was precipitated with ethanol by using glycogen (20 µg) as a carrier. Purified DNA was dissolved in Tris-EDTA buffer for Southern blot analysis. To detect naked core particles, 1.5 ml of precleared culture supernatant was incubated overnight at 4°C with 1.5 µl of polyclonal rabbit anti-core antibody (Dako), followed by the addition of 10 µl bed volume of protein G beads and a further incubation of 3 to 5 h. The subsequent steps were identical to those for viral particles.

Differentiation between genotypes A and G during mixed transfection experiments. DNA extracted from core particles or virions was treated at 37°C for 2 h in high-salt restriction enzyme buffer (Roche) with 1 U of Klenow fragment in the presence of a 100 µM concentration of deoxynucleoside triphosphates (dNTP) to repair the single-stranded region. The enzyme was inactivated at 75°C for 10 min. Next, one-third of the DNA extracted from intracellular core particles or extracellular virions was digested at 37°C for 4 h with 5 U of EcoRI, while another one-third was digested with 5 U each of BglI and XhoI. Digested and undigested DNA was heated at 85°C for 10 min, chilled on ice, and separated in agarose gels for Southern blot analysis.

Northern blot analysis. Cells were lysed with TRIzol (Invitrogen), and total RNA was extracted according the manufacturer's protocol. The RNA pellet was dissolved in nuclease-free water, and 8 µg was dissolved in loading buffer containing 2.1 M formaldehyde, heated at 95°C for 3 min, and separated in a 1.2% agarose gel containing 0.8 M formaldehyde. The blot was hybridized with mixed genotype A/G probe in the same manner as for Southern blotting. After stripping, the same blot was hybridized with a 1.5-kb SEAP cDNA probe, which had been PCR amplified with the sense primer 5'-TGGGCCTGAGGCTACAGCTC-3' and the antisense primer 5'-TATCTTATCATGTCTGCTCGAAGC-3'.

Primer extension experiments. Huh7 cells grown in six-well plates were harvested at day 3 posttransfection with 1 ml of TRIzol solution (Invitrogen) and RNA was extracted. A commercial kit (AMV Reverse Transcriptase; Promega) was used for the primer extension assay. The RNA (10 µg) was incubated at 58°C for 20 min with 300 fmol of an antisense primer labeled with [-³²P]ATP by T4 polynucleotide kinase. The primers used were 5'-GACTCTAAGGCTTCTCGATACAGAG-3' (positions 2031 to 2007), for genotype A samples, and 5'-GACTCTAAGGATTCCCGGTACAAAG-3' (positions 2067 to 2043), for genotype G samples. The annealed oligonucleotide was extended by AMV Reverse Transcriptase at 42°C for 30 min. The reaction was stopped by the addition of loading buffer, and the product was heated at 90°C for 10 min before separation in a 5% acrylamide gel containing 7 M urea with 1x Tris-borate-EDTA buffer. As a molecular size marker, HaeIII-digested X174 DNA was labeled with [-³²P]ATP by T4 polynucleotide kinase and run in parallel. The gel was dried, and primer extension products were revealed by autoradiography.

Detection of core protein and HBeAg from cell lysate or culture supernatant. HBeAg present in culture supernatant was quantified by the ETI-EBK PLUS enzyme immunoassay (DiaSorin). Western blot analysis was performed as described previously (17). Proteins from about 30 µl of cell lysate were separated by 0.1% SDS-12% polyacrylamide gel electrophoresis (PAGE) and transferred to polyvinylidene difluoride membranes. The blots were blocked at room temperature for 1 h with 3% bovine serum albumin (BSA) in Tris-buffered saline-0.05% Tween 20 (TBST) buffer and incubated overnight at 4°C with a polyclonal rabbit anti-core antibody (Dako) diluted 1:2,000 in 3% BSA-TBST. After a wash at room temperature for 40 min, the blots were incubated at room temperature for 1 h with a 1:20,000 to 1:40,000 dilution of anti-rabbit antibodies conjugated with horseradish peroxidase (HRP) (Amersham). The blots were washed again for 40 min, and signals were revealed by enhanced chemiluminescence. The core protein expressed from clone 4B of genotype A, which was not reactive to the rabbit antibody from Dako, was detected by a monoclonal antibody (14E11; Abcam) at a 1:2,000 dilution, followed by anti-mouse antibodies conjugated with HRP (1:20,000 dilution). In some experiments, the blots were treated with stripping buffer (Pierce) at 37°C for 25 min with agitation, washed five times with TBST for a total of 50 min, and blocked again in 3% BSA-TBST. The membranes were incubated with mouse GAPDH antibody (MAB374; Chemicon) at a 1:5,000 dilution, followed by a 1:10,000 dilution of anti-mouse antibody conjugated to HRP.

For Western blot analysis of secreted HBeAg and core protein, 1 ml of culture supernatant was rotated at 4°C overnight with 1 µl of rabbit polyclonal anti-core antibody (Dako), followed by the addition of 10 µl bed volume of protein G-agarose beads and a further incubation of 5 h. The beads were brought down by low-speed centrifugation and washed once with phosphate-buffered saline, and associated proteins were separated by SDS-15% PAGE. The subsequent Western blot analysis was identical to that for the core protein. For detection of core protein and HBeAg from patient sera, the immunoprecipitation step was performed at room temperature (3 h of incubation with the antibody followed by 2 h of incubation with protein G beads) to avoid precipitation of serum proteins.

Detection of core protein by pulse-chase experiments. Huh7 cells in six-well plates were rinsed with Hanks solution 2 days after transfection and starved for 2 h in methionine-free/cysteine-free Eagle MEM lacking calf serum. Next, cells were incubated for 3 h with 500 µl/well of the same medium supplemented with 0.12 mCi/ml of Express Protein Labeling Mix (PerkinElmer), followed by a wash with complete DMEM supplemented with extra methionine (10 times more than in the original medium). Cells were either harvested immediately by scraping or continued to incubate in methionine-fortified complete medium before harvest. The cell pellet was lysed in 500 µl of lysis buffer (10 mM HEPES [pH 7.5], 100 mM NaCl, 1 mM EDTA, and 1% NP-40) supplemented with protease inhibitor cocktail (Roche) and incubated with 10 µl bed volume of protein G beads at 4°C for 4 h, followed by a brief spin to remove proteins that directly bound to beads. The precleared lysate was incubated at 4°C overnight with 0.5 µl of rabbit polyclonal anti-core antibody (Dako) and incubated for 3 h more following the addition of 10 µl bed volume of protein G beads. The beads were collected by low-speed centrifugation and washed once with lysis buffer. Proteins bound to beads were separated in an SDS-12% polyacrylamide gel, which was treated sequentially with 10% acetic acid-25% methanol solution and Enlightening solution (New England Nuclear). The gel was dried and radioactive signals were revealed by exposure to X-ray films. Quantitative analysis was based on scanning of lightly exposed X-ray film followed by analysis with NIH Image software.

Detection of core particles. Intracellular core particles were detected by native agarose gels (24) with minor modifications. Huh7 cell lysate (10 µl) was diluted with a one-sixth volume of 6x DNA loading buffer (0.25% bromophenol blue, 40% sucrose) and applied to a 1% agarose gel. The gel was run in Tris-acetate-EDTA buffer at 90 V, and proteins were blotted overnight with 10x SSC solution to a polyvinylidene difluoride membrane which had been soaked successively in methanol, water, and 2x SSC before use. Subsequent detection of core protein on the blot was identical to that by Western blot analysis.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Replication and expression constructs. Two types of constructs were employed for study of the replication of genotype G: 1.1-mers of the viral genome cloned in the pTriEx vector downstream of the actin promoter and dimers of the viral genome inserted into a promoterless cloning vector (Fig. 1B). Robust transcription of pg RNA by the actin promoter makes it easy to monitor core protein expression, genome replication, and virion secretion. The biological relevance of the 36-nt insertion was evaluated by its removal from clone G1. Furthermore, we created core protein-deficient and encapsidation-defective mutants of G1 and its deletion mutant (G1d36) to separate the cis effect (36-nt insertion in the genome) from the trans effect (12-aa insertion in the core protein). Core protein expression was prevented by mutating the core AUG to AUA, while pg RNA encapsidation was abolished by the G1879T/T1880A double mutation at the loop of the signal (Fig. 1A).

It should be pointed out that the degree of replication observed from 1.1-mer genomes may not necessarily reflect the in vivo situation, where viral replication is driven by the endogenous core promoter and subject to modulation by the two enhancers, HBx protein, and naturally occurring mutations in the core promoter (6, 33). Moreover, such constructs do not transcribe the precore RNA required for HBeAg expression, because their 5' ends do not include the entire precore region (Fig. 1B). Therefore, tandem dimers of clone G1 and its deletion mutant were cloned into pUC18 vector to study viral replication in a more natural setting. Core gene constructs under control of the CMV promoter (CMV-core) (Fig. 1B) permitted us to examine the impact of the 36-nt insertion on core protein expression without complication by the secondary structure of the signal (Fig. 1A). Considering that genotype G has a nonfunctional precore region, the potential impact of the 36-nt insertion on HBeAg expression was studied in genotype A clones either as tandem dimers or as CMV-precore expression constructs (Fig. 1B).

Experimental conditions for unbiased detection of the two genotypes. Thanks to a >12% sequence divergence between genotypes A and G at the nucleotide level (21), a potential complication in comparative studies of the two HBV genotypes is preferential hybridization of the probe to DNA or RNA of the same genotype. We prepared a blot containing 30 pg to 1 ng of HBV DNA of genotypes A, D, and G for sequential hybridization with probes of genotypes A, G, and A/G at 1:1 ratio. The genotype A probe coupled with a low-salt washing buffer (0.5x SSC-0.1% SDS) was quite biased against the other genotypes, while the mixed probe coupled with a high-salt washing buffer (2x SSC-0.1% SDS) was able to detect both genotypes A and G at similar efficiencies (Fig. 2). These conditions were adopted for all of the Southern and Northern blot analyses.

View larger version (18K): [in this window] [in a new window]

FIG. 2. Identification of hybridization conditions for unbiased detection of both genotypes A and G. pTriExD1 and pTriExG1 were linearized by XhoI digestion, while an EcoRI dimer of genotype A (clone 4B) in the pUC18 vector was linearized with HindIII. Increasing amounts of digested DNA (30 pg, 100 pg, 300 pg, and 1,000 pg) were separated in an agarose gel, and the blot was hybridized successively with probes of genotype A, genotype G, and A/G at a 1:1 ratio. The blots were washed at 60°C for a total of 2 h with high-salt (2x SSC-0.1% SDS) or low-salt (0.5x SSC-0.1% SDS) buffer.

The rabbit polyclonal antibody against core protein (Dako) was capable of detecting core protein derived from different genotypes except for one genotype A clone (4B), due to an E77Q missense mutation (K. H. Kim et al., unpublished data). A mouse monoclonal antibody (Abcam) was used instead for clone 4B and its derivatives.

Genotype G clones are competent in the transcription of pregenomic and subgenomic RNAs. Huh7 cells were transfected with 1.1-mers of HBV genomes cloned in the pTriEx vector or tandem dimers (EcoRI or SphI) cloned in the pUC18 vector, together with SEAP cDNA. According to Northern blot analysis, two of the five genotype G clones in the pTriEx vector (G4 and G6) (Fig. 3A, lanes 6 and 7) produced amounts of pg RNA comparable to those of genotype A and D clones (Fig. 3A, lanes 1 and 2). The slightly lower HBV transcript levels associated with three clones from one patient (G1, G1d36, and G2) (Fig. 3A, lanes 3 to 5) correlated with reduced SEAP mRNA levels (Fig. 3B). Conversion of clone G1 into a tandem SphI dimer markedly reduced the transcription of pg RNA (Fig. 3, lanes 10 versus 3). Indeed, dimers of G1 and its d36 mutant produced less precore (pc)/pg RNA and subgenomic (sg) RNA than did clone 2A, a low-replicating genotype A clone (33) (Fig. 3, lanes 10, 11, and 9). Clone 4B, a high-replicating genotype A clone with core promoter mutations (33), generated the highest level of pc/pg RNA among the dimer constructs (Fig. 3, lane 8). Therefore, under endogenous promoters, genotype G may be less efficient in RNA transcription than genotype A.

View larger version (32K): [in this window] [in a new window]

FIG. 3. HBV RNA transcription in transfected Huh7 cells. The 1.1-mers were cloned in the pTriEx vector with transcription of the 3.5-kb pg RNA under control of the chicken actin promoter. The dimers were cloned into pUC18 vector, with transcription of all viral mRNAs under endogenous promoters. Huh7 cells were cotransfected with HBV constructs and SEAP cDNA, and 8 µg of RNA extracted at day 3 posttransfection was separated in a denaturing agarose gel. (A) Hybridization of the blot with a mixed probe of genotypes A and G. (B) Hybridization of the same blot with a SEAP probe following stripping of the HBV probe.

The 36-nt insertion in genotype G is critical for efficient core protein expression. Due to the strong actin promoter, the 1.1-mer genomes in the pTriEx vector were generally efficient in core protein expression, with the exception of a chimeric construct with a core gene derived from genotype C (G1/C1) (Fig. 4A, upper panel, lane 10). The different mobilities of core proteins generated from various genotypes are compatible with their sizes: 183 aa for genotypes C and D, 185 aa for genotype A, and 195 aa for genotype G. The levels of core particles assembled generally correlated with core protein levels (Fig. 4A; compare upper and lower panels), suggesting that core particle assembly was not adversely affected by the 12-aa insertion unique to genotype G. In this regard, other insertions at the N terminus of the core protein were also tolerated in terms of particle formation (18, 34). Removal of the genotype G-specific 36 nt from clone G1 markedly reduced core protein expression (Fig. 4A, upper panel; compare lanes 4 and 9). When the core protein was expressed from tandem SphI dimers under the endogenous promoter, G1 expressed a much higher level of core protein than did a genotype A clone (2A) (Fig. 4B, left panel). Removal of the 36-nt insertion from clone G1 reduced core protein to undetectable levels. In this regard, the first 3 nt of the 36-nt insertion alters base pairing at the lower stem of the signal, which may potentially modulate the efficiency of translation initiation or RNA packaging (Fig. 1A). However, encapsidation-deficient mutants of G1 or G1d36 produced levels of core protein similar to those of their parental 1.1-mer genomes (Fig. 5C, lanes 3 and 5 versus lanes 4 and 6). Furthermore, we generated core protein expression constructs driven by the CMV promoter, which produced core protein in the absence of a preceding signal (Fig. 1). Again clone G1 produced a much higher level of core protein than did its deletion mutant (Fig. 4C, left panel).

View larger version (24K): [in this window] [in a new window]

FIG. 4. Core protein expression from 1.1-mer genomes (A), tandem dimers (B), and CMV-core expression constructs (C). Cells were harvested at day 5 posttransfection, and core protein was detected from lysates by a rabbit polyclonal antibody (Dako), except for clone 4B and its derivatives (panels B and C, right). For these samples, a mouse monoclonal antibody was used instead. The predicted core protein size is 183 aa for genotypes C and D, 185 aa for genotype A, and 195 aa for genotype G. For panel C, two independent clones of G1d36 and 4Bins36 were analyzed. The lower part of panel A shows intracellular core particles that have been separated in a native agarose gel and probed with the rabbit polyclonal antibody.

View larger version (24K): [in this window] [in a new window]

FIG. 5. (A to C) Relative contribution of the 36-nt insertion in the genome versus the 12-aa insertion in the core protein to HBV DNA replication and virion secretion. The 1.1-mer genomes were used. The core-negative (core–) mutants of G1 and G1d36 (1 µg) were cotransfected with same amount of core-expressing constructs: -negative (–) mutants or CMV-core (lanes 7 to 14). For controls, 1 µg of the -negative mutants, CMV-core constructs, or parental clones was transfected with 1 µg of pcDNA3.1 vector DNA (lanes 1 to 6). About half of the cell lysate was used for detection of DNA replication (A), while another aliquot was used for successive detection of core protein and GAPDH (C). The supernatant was employed for detection of secreted virus particles (B). The EcoRI/RsrII digest of an EcoRI dimer served as size markers of 3.2, 1.7, and 1.5 kb. RC, relaxed circular; PDS, partially double stranded; SS, single stranded. (D) Impact of the 36-nt/12-aa insertion on release of HBV as virions or naked core particles. Pooled supernatants from transfected Huh7 cells were divided into two equal parts, one for immunoprecipitation with anti-S antibody and another for precipitation with anti-core antibody. The env– mutant of genotype A did not express envelope proteins and therefore failed to secrete virions.

In a complementary approach, we inserted the 36 nt into clone 4B of genotype A, either in the form of an EcoRI dimer or a CMV-core construct. In both cases, the insertion markedly enhanced core protein expression (Fig. 4B and C, right panels). The fact that a different antibody was used to detect core protein produced from clone 4B suggests that the 36-nt insertion in genotype G increased core protein expression or stability rather than its recognition by the polyclonal antibody from Dako.

The 36-nt insertion increased core protein translation but did not enhance mRNA abundance. Pulse-chase experiments were conducted to determine whether the 36-nt insertion increased de novo core protein translation or the 12-aa insertion slowed core protein degradation. Only genotype G constructs were examined, because the murine monoclonal antibody failed to immunoprecipitate core protein from clone 4B or 4Bins36. During a short labeling period of 3 h, about seven times more core protein was synthesized from the 1.1-mer genome of G1 than from G1d36 (Fig. 6A, lanes 3 and 4). Similarly, the CMV-core construct of G1 produced a much higher level of core protein than did its deletion mutant (Fig. 6A, lanes 1 and 2). Identical results were obtained following 1 h of pulse with [³⁵S]methionine (data not shown). Culturing of labeled cells for 68 h more resulted in a reduction of labeled core protein for both G1 and G1d36 to about 50% of the original level (Fig. 6A, lanes 6 and 7), suggesting similar protein half-lives. Thus, up-regulation of core protein by the 36-nt insertion is mediated by increased protein synthesis.

View larger version (20K): [in this window] [in a new window]

FIG. 6. The 36-nt insertion in the core gene increased core protein translation without up-regulating its steady-state mRNA level. (A) Pulse-chase experiment. Huh7 cells were transfected with the indicated constructs and labeled with [35S]methionine for 3 h. One set of the samples was harvested immediately, while another set was cultured for an additional 68 h in the absence of labeled methionine. Core protein was immunoprecipitated from cell lysates and revealed by SDS-PAGE and fluorography. Positions of prestained molecular size markers are also indicated (they underestimate the sizes of the endogenous proteins by at least 10%). The high-molecular-weight bands unique to pTriex constructs have the size of core protein dimers. (B) Primer extension assay. RNA (10 µg) extracted from day 3 posttransfection was annealed with an end-labeled antisense primer of genotype A (lanes 2 to 8) or genotype G (lanes 9 to 13), and the negative-stranded cDNA was synthesized with AMV Reverse Transcriptase. The product was heated and separated in a 5% denaturing acrylamide gel. End-labeled, HaeIII-digested X174 DNA served as molecular size markers. The predicted sizes of the primer extension products are indicated.

One potential mechanism whereby the 36-nt insertion increases core protein translation is by up-regulating the corresponding mRNA, either at the transcriptional or posttranscriptional level. However, primer extension assays excluded this possibility. For example, the 4Bins36 dimer did not produce more pg RNA than the parental clone, 4B (Fig. 6B, lanes 3 and 2), nor did the corresponding CMV-core construct (Fig. 6B, lanes 5 and 4). The CMV-core construct of G1 produced a higher transcript level than did G1d36 (Fig. 6B, lanes 11 and 12), which may partly explain the much higher core protein expression. However, this was not the case for 1.1-mer genome (Fig. 6B, lanes 9 and 10). Consistent with the primer extension data, Northern blot analysis also failed to reveal higher pg/pc RNA levels in G1 than in G1d36 (Fig. 3A, lanes 3, 4, 10, and 11). Therefore, the 36 nt unique to genotype G apparently make the corresponding mRNA more efficient for core protein translation.

The 36-nt insertion is essential for efficient replication of genotype G. The genotype G clones as 1.1-mer genomes in the pTriEx vector were replication competent (Fig. 7A, upper panel, lanes 4 to 8). Removing the 36 nt from clone G1 not only markedly reduced core protein expression, as described, but also suppressed genome replication to a similar extent (Fig. 7A, lane 9 versus lane 4). We employed a trans-complementation assay in an attempt to sort out whether the reduced encapsidation/replication efficiency of the G1d36 pg RNA or the small amount of core protein produced restricted genome replication. The core-negative mutants of G1 and G1d36 were able to replicate their respective genomes as long as core protein was provided in trans. Their -negative mutants could not package the pgRNA but were nevertheless competent in the expression of core protein in addition to DNA polymerase and envelope proteins. The CMV-core constructs differ from the -negative mutants in expressing the core protein alone. A representative result is shown in Fig. 5A. We found that replication of the G1 core-negative mutant (1 µg) could be rescued by 1 µg of the -negative mutant of either G1 or G1d36 to the level achievable by 1 µg of wild-type G1 (Fig. 5A; compare lanes 7 and 8 with lane 5). Rescue by the CMV-core constructs was less efficient, especially that of G1d36 (Fig. 5A, lanes 9 and 10). Replication of the G1d36 core-negative mutant was rescued more efficiently by the -negative mutant of G1 than of G1d36 (Fig. 5A, lanes 11 and 12), although even the latter combination reached higher replication capacity than did the parental clone, G1d36 (Fig. 5A, lane 6). Therefore, the trans-complementation experiments implicate a combination of cis and trans defects in the lower replication capacity of G1d36. Certainly, the trans-complementation assay may have its limitations, because the pg RNA of the core- and -negative mutants served as pregenome and mRNA respectively, rather than both pregenome and mRNA for the parental construct. With the CMV-core construct, translation of core protein was no longer coordinated with expression of DNA polymerase, another component required for pg RNA encapsidation.

View larger version (27K): [in this window] [in a new window]

FIG. 7. HBV DNA replication and virion secretion from the 1.1-mer constructs (A) and dimer constructs (B and C). Cells and culture supernatants were harvested at day 5 posttransfection. Core particles extracted from cell lysates were used for the detection of DNA replication, while virus particles were immunoprecipitated from culture supernatants with anti-S antibodies prior to Southern blot analysis. For the lower part of panel B, virus particles secreted from clone 2A contained primarily single-stranded genome. For panel C, SphI dimers were cotransfected with CMV-core constructs at a 1:1 ratio. In addition to DNA replication and virion secretion (upper and middle panels), core protein expression was also monitored (lower panel). The rabbit polyclonal antibody used (Dako) failed to detect core protein expressed from clone 4B.

The SphI dimer of clone G1 replicated to an extent similar to that of clone 2A of genotype A, whereas G1d36 was nearly defective in genome replication (Fig. 7B, upper panel). This defect could be rescued by cotransfection of the CMV-core construct of clone G1, although the rescue was incomplete (Fig. 7C, top panel; compare lane 6 with lane 1).

The 12-aa insertion in the core protein of genotype G enhanced the genome maturity of secreted virus particles. Interestingly, all of the genotype G clones secreted virus particles of much higher genome maturity than clones of other genotypes, as evidenced by the paucity of single-stranded DNA and the abundance of relaxed circular (RC) genome (Fig. 7A, lower panel, lanes 4 to 8). The less efficient virion secretion of clone G3 was correlated with much-reduced HBsAg secretion (unpublished data). Replacement of the core gene of clone G1 with genotype C (G1/C1) reduced the genome maturity of secreted particles (Fig. 7A, lower panel, lane 11 versus lane 4). Removal of the 36-nt insertion from G1 produced a similar effect, but the efficiency of virion secretion was enhanced as evidenced by the ratio of extracellular HBV DNA/intracellular viral DNA (Fig. 7A, lane 9). Whether the genome maturity of secreted virus particles was controlled by the 36-nt insertion in the genotype G DNA or the 12-aa extension in the core protein was established by trans-complementation assay, as described above. We found that genome maturity was determined primarily by the core protein expression construct. Thus, the G1 -negative mutant generated virions of mature genome whether cotransfected with the G1 core-negative or the G1d36 core-negative mutant (Fig. 5B, lanes 7 and 11), whereas the G1d36 -negative mutant ensured more efficient virion secretion with a less mature genome (Fig. 5B, lanes 8 and 12). Similar results were obtained with core protein expressed from CMV-core constructs (Fig. 5B, lanes 9, 13, 10, and 14).

The 36-nt/12-aa insertion increases secretion of naked core particles. By simultaneous detection of both virions and naked core particles in the culture supernatant, it became apparent that genotype G secreted more naked core particles than did genotype A (Fig. 5D, lanes 1, 2, 5, and 6). Moreover, naked core particles of genotype G contained a less mature genome than the corresponding virions (Fig. 5D; compare lanes 6 and 2). Deletion of the 36 nt not only increased virion secretion (Fig. 5D, lane 3) but also markedly suppressed core particle release (Fig. 5D, lane 7). The virus particles of G1d36 had a genome maturity similar to that of the naked core particles of G1 (Fig. 5D; compare lanes 3 and 6). Therefore, the 36-nt insertion enhanced the genome maturity of secreted virions through inhibition of core particle envelopment.

Independent genome replication and virion secretion of genotypes A and G. Considering the frequent coinfection of genotype G with genotype A, we performed cotransfection experiments to test for possible mutual enhancement or interference of replication/virion secretion. The 1.1-mer genomes in the pTriEx vector (A1 and G4) were employed for such studies because the level of virion secretion from dimers was too low. Differentiation between the progeny DNAs of the two genotypes was made possible by the single EcoRI site at position 1 of genotype A (but not of genotype G) and single sites of BglI (position 1925, inside the 36-nt insertion) and XhoI (position 2907) on genotype G (but not genotype A) (Fig. 8A). To minimize the effect of the single-stranded region on the banding of digestion products, we repaired the single-stranded gap. Furthermore, digested or undigested DNA was heated at 85°C for 10 min prior to gel electrophoresis to convert RC DNA into a linear form. As shown in Fig. 8B and C, digestion with EcoRI converted A1 DNA into fragments of approximately 1.8 kb and 1.4 kb (lane 3) but did not affect G4 DNA (lane 15 in panel A and lane 12 in panel B). Double digestion with BglI/XhoI converted G4 DNA into fragments of 2.2 kb and 1 kb (lane 14 in panel A and lane 11 in panel B) but did not affect the migration of A1 DNA (lane 2). With ratios of input DNA at 1:3, 1:1, and 3:1, both genotypes were able to replicate and secrete virus particles, with their relative abundance proportional to the input (Fig. 8B and C). Assuming that most cells were transfected with both genotypes, the results suggest no competition or enhancement between the two genotypes, at least for replication driven by a strong exogenous promoter.

View larger version (21K): [in this window] [in a new window]

FIG. 8. Lack of significant interference between genotypes A and G with regard to genome replication and virion secretion. (A) Cartoon view of the EcoRI site of genotype A and the BglI/XhoI sites of genotype G. Shown are linear and RC forms with the single-stranded region repaired. EcoRI digestion converts the linear form of genotype A into two bands of 1.8 kb and 1.4 kb, whereas double digestion with BglI and XhoI produces 2.2-kb and 1-kb bands for the linear form of genotype G. The heating step prior to electrophoresis melts the base pairing between the 5' ends of the positive and negative strands in the RC DNA, thus generating the same migration pattern as the linear DNA. The numbers 1 and 2 inside boxes refer to DR1 and DR2 sequences. (B) HBV DNA associated with intracellular core particles. (C) HBV DNA of extracellular virions. Huh7 cells grown in 10-cm dishes were transfected with a total of 10 µg of 1.1-mer genomes at the ratios indicated. DNA harvested at day 5 posttransfection was treated with Klenow fragment in the presence of dNTP. A one-third aliquot of the DNA was digested with EcoRI, while another aliquot with treated with BglI and XhoI. Digested and undigested DNA was heated at 85°C for 10 min before Southern blot analysis with mixed A/G probe. For intracellular DNA (B), the smaller fragments of EcoRI and BglI/XhoI digestion (1.3 kb and 1.0 kb) were less distinct.

The 12-aa insertion in the precore protein is able to reduce HBeAg production. Genotype G harbors the C1817T and G1896A nonsense mutations in the precore region, which are capable of terminating HBeAg translation. However, it has been speculated that the 12-aa insertion at the amino terminus of the core protein may serve as a signal peptide to generate a variant HBeAg from the core gene alone. First, the impact of the 12-aa insertion on HBeAg expression was studied in genotype A, which has a functional precore region. The 36-nt insertion caused about a fivefold drop in HBeAg expression from both an EcoRI dimer (clone 4B) and a CMV-precore construct (clone 6.2) (Fig. 9C, left and middle panels). Since the insertion did not down-regulate the corresponding mRNA (Fig. 6B, lanes 6 and 7), it is likely that the 12-aa insertion impaired HBeAg processing or secretion. Immunoprecipitation and Western blot analysis revealed reduced rather than accelerated mobility of the variant HBeAg (Fig. 9C, right panel), suggesting that the original N-terminal signal peptide of 19 residues, rather than the 12 inserted residues, was recognized and removed by the signal peptidase.

View larger version (26K): [in this window] [in a new window]

FIG. 9. Core protein and HBeAg in the culture supernatant of transfected Huh7 cells (A to C) and patient blood (D). (A) "HBeAg" from cells transfected with 1.1-mer viral genomes. "HBeAg" was measured from 15 µl of culture supernatant harvested at day 5 posttransfection. Values (with A1 arbitrarily set at 1) were averaged from three transfection experiments. (B) Immunoprecipitation-Western blot analysis of core protein and HBeAg secreted to the culture supernatant. Clones K81 and 5.4 were defective in HBeAg expression due to frameshift mutations in the precore region and the core gene, respectively. (C) Impact of the 36-nt insertion on HBeAg expression from two genotype A clones as an EcoRI dimer (4B) or CMV-precore construct (6.2). The left and middle panels show HBeAg values of the insertion mutants relative to the parental constructs, which were arbitrarily set at 1. Results are based on three transfection experiments. The right panel shows immunoprecipitation-Western blot analysis. (D) Immunoprecipitation-Western blot analysis of core protein and HBeAg in sera of patients infected with genotype A, G, or both. The culture supernatant of transfected Huh7 cells (sup) served as a positive control. The HBeAg values shown at the bottom were measured from 1.5 µl of serum sample or culture supernatant.

Core protein overexpression leads to its release into the culture supernatant. We next investigated whether the 12-aa insertion as part of the core protein could be cleaved by a signal peptidase to generate variant HBeAg. The 1.1-mer genomes, which were able to express core protein but not precore protein, were utilized. Interestingly, a low level of "HBeAg" was detected, but this was not unique to genotype G (Fig. 9A). Failure of the G1 core-negative mutant to secrete such "HBeAg" (Fig. 9A, last lane) confirmed it as a core gene product. The amount of "HBeAg" released by the 1.1-mer genomes was less than 5% of the HBeAg produced by the SphI dimer of a genotype A clone (2A). Immunoprecipitation and Western blot analysis revealed intact core protein (21 kDa for most genotypes and 23 kDa for genotype G) in such supernatants (Fig. 9B, lanes 1 to 9). If variant HBeAg were produced from genotype G core protein by cleavage around the 12-aa insertion followed by C-terminal cleavage, it should have been smaller than the classic HBeAg secreted by the SphI dimer of clone 2A (Fig. 9B, lane 11). Since nonparticulate core protein is known to possess HBe antigenicity (37), the "HBeAg" produced by the 1.1-mer HBV genomes may reflect the release of free core protein or disassembly of released core/virus particles.

Natural infection with genotype G is associated with core protein release. The inability of genotype G to secrete bona fide HBeAg (Fig. 9B), coupled with recent findings of genotype G coinfection with genotype A or F (20, 36), suggests genotype A or F as the source of the HBeAg detected in clinical samples. We collected five samples infected with genotype G, three of which had high HBeAg titers (Fig. 9D, lanes 1, 3, and 4). Both direct sequencing and the INNO-LiPA test (Innogenetics) identified sample 1 as genotype G. Sample 3 was found to be genotype G by direct sequencing but was a mixture of A and G genotypes according to INNO-LiPA. The INNO-LiPA assay also identified sample 4 as being mixed A/G infection. Immunoprecipitation and Western blot analysis revealed a strong band of HBeAg in the two HBeAg⁺ genotype A samples, a finding which was comparable to that from culture supernatant of transfected Huh7 cells (Fig. 9D, lanes 8, 9, and 10). In contrast, samples 1 and 3 of genotype G produced only a weak or barely visible band at the position of classic HBeAg. Interestingly, both samples displayed an additional band corresponding to genotype G-specific core protein (Fig. 9D, lanes 1 and 3), suggesting core protein release as a consequence of increased production.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The frequent coinfection of genotype G with another HBV genotype suggests a possible defect of genotype G in replication or transmission. In this regard, the salient features of genotype G sequence include the 36-nt insertion at the 5' end of the core gene, a double nonsense mutation in the precore region, and a double core promoter mutation. The precore and core promoter mutations do develop in other HBV genotypes, albeit at a late stage of infection, in response to anti-HBe immunity (3, 7, 8, 28, 31, 40). The precore mutations that abrogate the expression of HBeAg do not necessarily impair HBV replication, and the core promoter mutations actually enhance viral replication (5, 6, 33, 39, 41, 43, 45). Thus, genotype G could be defective in genome replication due to the extra 36 nt not found in other genotypes. Our experiments demonstrated that genotype G clones were replication competent whether driven by the strong actin promoter (1.1-mer genomes in pTriEx vector) or by the endogenous promoter (SphI dimers in pUC18 vector), as evidenced by RNA transcription, envelope protein expression (our unpublished observations), core protein expression and particle assembly, genome replication, and virion secretion. These findings confirm an earlier report of the replication capacity of an SpeI dimer of genotype G (25) and are fully compatible with a recent report of genotype G monoinfection in a blood donor and several transfusion recipients (12). Those authors were unable to detect genotype A sequence in blood samples by a variety of sensitive methods, thus demonstrating convincingly that genotype G can be transmitted and propagated in patients without the need for a helper virus.

The transfection experiments revealed several interesting biological features of genotype G, including more efficient core protein expression, secretion of virus particles with high genome maturity, and, surprisingly, dependence of its replication on the 36-nt insertion. Furthermore, the same 36-nt insertion underlies both efficient core protein expression and secretion of virions with a more mature genome. The exact mechanism whereby the 36-nt insertion increases core protein levels remains to be worked out, but it has to do with more efficient protein synthesis rather than extension of protein half-life (Fig. 6A). Moreover, the increased translation occurred in the absence of up-regulation of the corresponding mRNA (Fig. 6B). This finding is reminiscent of our recent report revealing marked down-regulation of core protein expression by several point mutations in the precore region, such as G1862T and T1863A (in the bulge of the signal), apparently without reducing pg RNA abundance (17). Based on experiments with the CMV-core constructs, the 36-nt insertion can impact core protein translation even in the absence of the signal. It will be interesting to examine whether the 36-nt insertion increases the distribution of the corresponding mRNA toward polysomes, the ribosome fraction actively involved in translation. One possible scenario is that core protein binds to its own mRNA to inhibit protein translation (for coordinated core particle assembly and pg RNA packaging), which is weakened by the 36-nt insertion on the mRNA or the 12-aa insertion in core protein. In this regard, dihydrofolate reductase protein has been found to down-regulate its own translation by binding to the cognate mRNA (11, 14). Further experiments are needed to test our hypothesis.

Since the 36-nt insertion is needed for both efficient core protein expression and replication of genotype G (both as 1.1-mer genomes and SphI dimers), it is natural to ask whether the two phenotypes are connected. In the present study, we performed detailed trans-complementation assays only on the 1.1-mer genomes, which are certainly different from natural infection in overproduction of the pg RNA and, consequently, core protein. The results were not clear-cut, but it appears that both the loss of the 36 nt in the pg RNA and the loss of the 12 aa in the core protein contributed to reduced replication of G1d36. Therefore, cotransfection of the G1d36 core-negative mutant with its -negative mutant gave rise to lower replication than did cotransfection of the G1 core-negative mutant with the -negative mutant of either G1 or G1d36. Another observation was the much less efficient rescue of the core-negative mutants by CMV-core constructs, which differ from the -negative mutants by not expressing polymerase and envelope proteins. The same G1 CMV-core construct only partially rescued the replication defect of the SphI dimer of G1d36. However, this experiment differed from that of the 1.1-mer genomes in that core protein with the 12-aa deletion was still expressed. Considering that the SphI dimers of G1 and G1d36 produced less pg/pc RNA than did a genotype A clone (2A) (Fig. 3), we favor the hypothesis that the 36-nt insertion is required to compensate for the low abundance of pg RNA to sustain genotype G replication. The fact that the SphI dimer of clone G1 replicated to a degree similar to that of clone 2A suggests that up-regulation of core protein translation alone is sufficient to augment replication. Alternatively, since both the core protein and DNA polymerase are translated from pg RNA, the 36-nt insertion may also up-regulate the translation of polymerase. This possibility should be tested experimentally.

While the question of whether the 36-nt insertion in the pg RNA or the 12-aa insertion in the core protein up-regulates core protein translation remains open, we have compelling evidence to suggest that the 12-aa insertion in the core protein leads to secretion of virus particles with a much more mature genome and higher RC DNA content than other genotypes. This point was demonstrated convincingly by the trans-complementation experiments. Whether for the G1 core-negative mutant or the G1d36 core-negative mutant, cotransfection with expression constructs of G1 core protein (the -negative mutant or the CMV-core construct) led to secretion of mature genome, whereas cotransfection with G1d36 core constructs ensured secretion of less mature genome but at higher efficiency. Simultaneous analysis of naked core particles released to the culture supernatant from both G1 and G1d36 mutants suggests that the 12-aa insertion reduced the efficiency of core particle envelopment, leading to (i) secretion of more naked core particles, (ii) a moderate increase in the genome maturity of intracellular core particles, and (iii) markedly increased genome maturity of particles that were eventually secreted as virions. It will be of interest to repeat the experiments with HepG2 cells, which in our hand produced more mature genomes than did Huh7 cells.

Two artificial HBV mutants with 10- and 23-aa insertions at the N terminus of the core protein failed to secrete virus particles (18), while a single naturally occurring P5T mutation near the N terminus of the core protein reduced virion secretion (27). A series of N-terminal insertion mutants of duck HBV secreted virus particles of reduced rather than increased genome maturity, which was associated with preferential degradation of particles harboring mature genome (23). Therefore, N-terminal insertion in the core protein modulates virion secretion, although the exact outcome is sequence specific.

The 12-aa insertion has been proposed to drive HBeAg production in the blood of patients infected with genotype G. According to this hypothesis, a fraction of the core protein enters the secretory pathway through a novel signal peptide created by the 12-aa insertion, followed by removal of both the signal peptide and the arginine-rich sequence at the carboxyl terminus. However, we failed to observe protein bands with sizes comparable to such variant HBeAg even with core protein overexpression driven by the strong actin promoter. Insertion of the same 36 nt into genotype A clones with a functional precore region suppressed HBeAg secretion but did not alter the cleavage site. This result agrees with the lack of HBeAg or anti-HBe antibody in patients infected with genotype G alone (12). Nevertheless, a high titer of HBeAg could be detected in the sera of two patients infected with genotype G, one with no indication of coinfection with another genotype (at least based on direct sequencing and the INNO-LiPA assay). Immunoprecipitation-Western blot analysis of two serum samples infected with genotype G did not reveal a sufficient amount of HBeAg to account for the extremely high HBeAg titers (Fig. 9D, lanes 1 and 3). Although both samples contained core protein, our experience with 1.1-mer genomes revealed extremely low HBe antigenicity of secreted core protein. An alternative possibility is that the HBeAg from these two samples, derived from coinfecting genotype A isolates, harbored mutations that rendered it poorly recognizable by the polyclonal antibody (Dako) used in this study. We found that the core protein and the HBeAg from clone 4B of genotype A were not recognized by the Dako antibody (Fig. 7C) due to an E77Q mutation; this mutation is very common among genotype A clones isolated from the late HBeAg⁺ phase of infection.

What is responsible for the rare monoinfection by genotype G? One explanation is its relatively inefficient replication. The SphI dimer of G1 replicated to a level comparable to that of clone 2A, a low-replicating genotype A clone, despite the presence of a double core promoter mutation (A1762T/G1764A). In some transfection experiments, the 1.1-mer genomes of genotype G also replicated less efficiently than the corresponding genotype A or D clones. Secondly, genotype G and genotype A (or genotype F) may complement each other for better replication or transmission. For example, the high level of core protein produced by genotype G may increase the replication of another genotype. Further cotransfection experiments using tandem dimers rather than 1.1-mer genomes, as shown in Fig. 8, may provide additional information. It will be also interesting to study the fitness of genotype G alone or in a mixed infection with genotype A in differentiated HepRG cells (16). However, we reason that the major impediment to monoinfection by genotype G may be its inability to express HBeAg, a molecule believed to play an immunomodulatory function critical for the establishment of persistent infection (10, 30). It has been reported that children with maternal transmission of pure wild-type virus became chronically infected, while those infected with a mixture of wild-type virus and an HBeAg-negative mutant resolved the acute infection (35). Similarly, the precore-defective mutant of woodchuck hepatitis virus induced only transient infection in the animals (9). Consistent with the immunomodulatory role of HBeAg, genotype G frequently infects homosexual men, who, due to immune suppression, are less likely to clear the viral infection (4, 44). In such hosts, immune escape mechanisms are probably not essential for the establishment of persistent infection, as opposed to the case with immunocompetent individuals. But even in such a group, the HBeAg-producing genotype is the dominant viral species during the early stage of infection; selection of genotype G coincides with the emergence of anti-HBe (20, 21, 42).

In conclusion, genotype G is replication competent. Efficient replication of genotype G requires the 36-nt insertion in the core gene, which may modulate the efficiency of core protein expression and virion secretion, as well as the genome maturity of virus particles. Our findings provide a molecular explanation for the presence of the 36-nt insertion in genotype G. Considering that this insertion has never been found in other genotypes, whether its artificial introduction into other genotypes will modulate genome replication warrants further investigation.

 
This work was supported by grants CA109733, CA95490, CA35711, DK066950, and DK28614 from the National Institutes of Health; by grant RSG 06-059-01-MBC from the American Cancer Society; and by a grant from the Agence Nationale de Recherche Contre le SIDA (ANRS) to F.Z. K.L. was supported in part by the Tan Yan Kee Foundation.

 
* Corresponding author. Mailing address for Fabien Zoulim: INSERM Unit 871, 151 Cours Albert Thomas, 69003 Lyon, France. Phone: 33472681970. Fax: 33472681971. E-mail: zoulim{at}lyon.inserm.fr. Mailing address for Shuping Tong: Liver Research Center, Rhode Island Hospital, Brown University, 55 Claverick St., 4th Floor, Providence, RI 02903. Phone: (401) 444-7365. Fax: (401) 444-2939. E-mail: Shuping_Tong_MD{at}Brown.edu

Published ahead of print on 13 June 2007.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Ahn, S. H., A. Kramvis, S. Kawai, H. C. Spangenberg, J. Li, G. Kimbi, M. Kew, J. Wands, and S. Tong. 2003. Sequence variation upstream of precore translation initiation codon reduces hepatitis B virus e antigen production. Gastroenterology 125:1370-1378.[CrossRef][Medline]
2
2. Bang, G., K. H. Kim, M. Guarnieri, F. Zoulim, S. Kawai, J. Li, J. Wands, and S. Tong. 2005. Effect of mutating the two cysteines required for HBe antigenicity on hepatitis B virus DNA replication and virion secretion. Virology 332:216-224.[CrossRef][Medline]
3
3. Baptista, M., A. Kramvis, and M. C. Kew. 1999. High prevalence of 1762(T) 1764(A) mutations in the basic core promoter of hepatitis B virus isolated from black Africans with hepatocellular carcinoma compared with asymptomatic carriers. Hepatology 29:946-953.[CrossRef][Medline]
4
4. Bhat, R. A., P. P. Ulrich, and G. N. Vyas. 1990. Molecular characterization of a new variant of hepatitis B virus in a persistently infected homosexual man. Hepatology 11:271-276.[CrossRef][Medline]
5
5. Blum, H. E., T. J. Liang, E. Galun, and J. R. Wands. 1991. Persistence of hepatitis B viral DNA after serological recovery from hepatitis B virus infection. Hepatology 14:56-63.[CrossRef][Medline]
6
6. Buckwold, V. E., Z. Xu, M. Chen, T. S. Yen, and J. H. Ou. 1996. Effects of a naturally occurring mutation in the hepatitis B virus basal core promoter on precore gene expression and viral replication. J. Virol. 70:5845-5851.[Abstract]
7
7. Carman, W. F., M. R. Jacyna, S. Hadziyannis, P. Karayiannis, M. J. McGarvey, A. Makris, and H. C. Thomas. 1989. Mutation preventing formation of hepatitis B e antigen in patients with chronic hepatitis B infection. Lancet ii:588-591.
8
8. Chan, H. L., M. Hussain, and A. S. Lok. 1999. Different hepatitis B virus genotypes are associated with different mutations in the core promoter and precore regions during hepatitis B e antigen seroconversion. Hepatology 29:976-984.[CrossRef][Medline]
9
9. Chen, H. S., M. C. Kew, W. E. Hornbuckle, B. C. Tennant, P. J. Cote, J. L. Gerin, R. H. Purcell, and R. H. Miller. 1992. The precore gene of the woodchuck hepatitis virus genome is not essential for viral replication in the natural host. J. Virol. 66:5682-5684.[Abstract/Free Full Text]
10
10. Chen, M. T., J. N. Billaud, M. Sallberg, L. G. Guidotti, F. V. Chisari, J. Jones, J. Hughes, and D. R. Milich. 2004. A function of the hepatitis B virus precore protein is to regulate the immune response to the core antigen. Proc. Natl. Acad. Sci. USA 101:14913-14918.[Abstract/Free Full Text]
11
11. Chu, E., C. H. Takimoto, D. Voeller, J. L. Grem, and C. J. Allegra. 1993. Specific binding of human dihydrofolate reductase protein to dihydrofolate reductase messenger RNA in vitro. Biochemistry 32:4756-4760.[CrossRef][Medline]
12
12. Chudy, M., M. Schmidt, V. Czudai, H. Scheiblauer, S. Nick, M. Mosebach, M. K. Hourfar, E. Seifried, W. K. Roth, E. Grunelt, and C. M. Nubling. 2006. Hepatitis B virus genotype G monoinfection and its transmission by blood components. Hepatology 44:99-107.[CrossRef][Medline]
13
13. Durantel, D., S. Carrouee-Durantel, B. Werle-Lapostolle, M. N. Brunelle, C. Pichoud, C. Trepo, and F. Zoulim. 2004. A new strategy for studying in vitro the drug susceptibility of clinical isolates of human hepatitis B virus. Hepatology 40:855-864.[CrossRef][Medline]
14
14. Ercikan-Abali, E. A., D. Banerjee, M. C. Waltham, N. Skacel, K. W. Scotto, and J. R. Bertino. 1997. Dihydrofolate reductase protein inhibits its own translation by binding to dihydrofolate reductase mRNA sequences within the coding region. Biochemistry 36:12317-12322.[CrossRef][Medline]
15
15. Fung, S. K., and A. S. Lok. 2004. Hepatitis B virus genotypes: do they play a role in the outcome of HBV infection? Hepatology 40:790-792.[CrossRef][Medline]
16
16. Gripon, P., S. Rumin, S. Urban, J. Le Seyec, D. Glaise, I. Cannie, C. Guyomard, J. Lucas, C. Trepo, and C. Guguen-Guillouzo. 2002. Infection of a human hepatoma cell line by hepatitis B virus. Proc. Natl. Acad. Sci. USA 99:15655-15660.[Abstract/Free Full Text]
17
17. Guarnieri, M., K. H. Kim, G. Bang, J. Li, Y. Zhou, X. Tang, J. Wands, and S. Tong. 2006. Point mutations upstream of hepatitis B virus core gene affect DNA replication at the step of core protein expression. J. Virol. 80:587-595.[Abstract/Free Full Text]
18
18. Hui, E. K., Y. S. Yi, and S. J. Lo. 1999. Hepatitis B viral core proteins with an N-terminal extension can assemble into core-like particles but cannot be enveloped. J. Gen. Virol. 80:2647-2659.[Abstract/Free Full Text]
19
19. Kao, J. H. 2002. Hepatitis B viral genotypes: clinical relevance and molecular characteristics. J. Gastroenterol. Hepatol. 17:643-650.[CrossRef][Medline]
20
20. Kato, H., E. Orito, R. G. Gish, N. Bzowej, M. Newsom, F. Sugauchi, S. Suzuki, R. Ueda, Y. Miyakawa, and M. Mizokami. 2002. Hepatitis B e antigen in sera from individuals infected with hepatitis B virus of genotype G. Hepatology 35:922-929.[CrossRef][Medline]
21
21. Kato, H., E. Orito, R. G. Gish, F. Sugauchi, S. Suzuki, R. Ueda, Y. Miyakawa, and M. Mizokami. 2002. Characteristics of hepatitis B virus isolates of genotype G and their phylogenetic differences from the other six genotypes (A through F). J. Virol. 76:6131-6137.[Abstract/Free Full Text]
22
22. Khan, N., M. Guarnieri, S. H. Ahn, J. Li, Y. Zhou, G. Bang, K. H. Kim, J. R. Wands, and S. Tong. 2004. Modulation of hepatitis B virus secretion by naturally occurring mutations in the S gene. J. Virol. 78:3262-3270.[Abstract/Free Full Text]
23
23. Kock, J., S. Wieland, H. E. Blum, and F. von Weizsacker. 1998. Duck hepatitis B virus nucleocapsids formed by N-terminally extended or C-terminally truncated core proteins disintegrate during viral DNA maturation. J. Virol. 72:9116-9120.[Abstract/Free Full Text]
24
24. Koschel, M., D. Oed, T. Gerelsaikhan, R. Thomssen, and V. Bruss. 2000. Hepatitis B virus core gene mutations which block nucleocapsid envelopment. J. Virol. 74:1-7.[Abstract/Free Full Text]
25
25. Kremsdorf, D., F. Garreau, F. Capel, M. A. Petit, and C. Brechot. 1996. In vivo selection of a hepatitis B virus mutant with abnormal viral protein expression. J. Gen. Virol. 77:929-939.[Abstract/Free Full Text]
26
26. Lacombe, K., V. Massari, P. M. Girard, L. Serfaty, J. Gozlan, G. Pialoux, P. Mialhes, J. M. Molina, C. Lascoux-Combe, D. Wendum, F. Carrat, and F. Zoulim. 2006. Major role of hepatitis B genotypes in liver fibrosis during coinfection with HIV. AIDS 20:419-427.[Medline]
27
27. Le Pogam, S., T. T. Yuan, G. K. Sahu, S. Chatterjee, and C. Shih. 2000. Low-level secretion of human hepatitis B virus virions caused by two independent, naturally occurring mutations (P5T and L60V) in the capsid protein. J. Virol. 74:9099-9105.[Abstract/Free Full Text]
28
28. Li, J., S. Tong, L. Vitvitski, F. Zoulim, and C. Trepo. 1990. Rapid detection and further characterization of infection with hepatitis B virus variants containing a stop codon in the distal pre-C region. J. Gen. Virol. 71:1993-1998.[Abstract/Free Full Text]
29
29. Magnius, L. O., and H. Norder. 1995. Subtypes, genotypes and molecular epidemiology of the hepatitis B virus as reflected by sequence variability of the S-gene. Intervirology 38:24-34.[Medline]
30
30. Milich, D., and T. J. Liang. 2003. Exploring the biological basis of hepatitis B e antigen in hepatitis B virus infection. Hepatology 38:1075-1086.[CrossRef][Medline]
31
31. Okamoto, H., F. Tsuda, Y. Akahane, Y. Sugai, M. Yoshiba, K. Moriyama, T. Tanaka, Y. Miyakawa, and M. Mayumi. 1994. Hepatitis B virus with mutations in the core promoter for an e antigen-negative phenotype in carriers with antibody to e antigen. J. Virol. 68:8102-8110.[Abstract/Free Full Text]
32
32. Ou, J. H., O. Laub, and W. J. Rutter. 1986. Hepatitis B virus gene function: the precore region targets the core antigen to cellular membranes and causes the secretion of the e antigen. Proc. Natl. Acad. Sci. USA 83:1578-1582.[Abstract/Free Full Text]
33
33. Parekh, S., F. Zoulim, S. H. Ahn, A. Tsai, J. Li, S. Kawai, N. Khan, C. Trepo, J. Wands, and S. Tong. 2003. Genome replication, virion secretion, and e antigen expression of naturally occurring hepatitis B virus core promoter mutants. J. Virol. 77:6601-6612.[Abstract/Free Full Text]
34
34. Pumpens, P., and E. Grens. 1999. Hepatitis B core particles as a universal display model: a structure-function basis for development. FEBS Lett. 442:1-6.[CrossRef][Medline]
35
35. Raimondo, G., E. Tanzi, S. Brancatelli, S. Campo, M. A. Sardo, G. Rodino, M. Pernice, and A. R. Zanetti. 1993. Is the course of perinatal hepatitis B virus infection influenced by genetic heterogeneity of the virus? J. Med. Virol. 40:87-90.[Medline]
36
36. Sanchez, L. V., Y. Tanaka, M. Maldonado, M. Mizokami, and A. Panduro. 2007. Difference of hepatitis B virus genotype distribution in two groups of Mexican patients with different risk factors. High prevalence of genotype H and G. Intervirology 50:9-15.[CrossRef][Medline]
37
37. Seifer, M., and D. N. Standring. 1995. Assembly and antigenicity of hepatitis B virus core particles. Intervirology 38:47-62.[Medline]
38
38. Stuyver, L., S. De Gendt, C. Van Geyt, F. Zoulim, M. Fried, R. F. Schinazi, and R. Rossau. 2000. A new genotype of hepatitis B virus: complete genome and phylogenetic relatedness. J. Gen. Virol. 81:67-74.[Abstract/Free Full Text]
39
39. Tong, S. P., C. Diot, P. Gripon, J. Li, L. Vitvitski, C. Trepo, and C. Guguen-Guillouzo. 1991. In vitro replication competence of a cloned hepatitis B virus variant with a nonsense mutation in the distal pre-C region. Virology 181:733-737.[CrossRef][Medline]
40
40. Tong, S. P., J. S. Li, L. Vitvitski, and C. Trepo. 1990. Active hepatitis B virus replication in the presence of anti-HBe is associated with viral variants containing an inactive pre-C region. Virology 176:596-603.[CrossRef][Medline]
41
41. Tong, S. P., J. S. Li, L. Vitvitski, and C. Trepo. 1992. Replication capacities of natural and artificial precore stop codon mutants of hepatitis B virus: relevance of pregenome encapsidation signal. Virology 191:237-245.[CrossRef][Medline]
42
42. Tran, A., D. Kremsdorf, F. Capel, C. Housset, C. Dauguet, M. A. Petit, and C. Brechot. 1991. Emergence of and takeover by hepatitis B virus (HBV) with rearrangements in the pre-S/S and pre-C/C genes during chronic HBV infection. J. Virol. 65:3566-3574.[Abstract/Free Full Text]
43
43. Ulrich, P. P., R. A. Bhat, I. Kelly, M. R. Brunetto, F. Bonino, and G. N. Vyas. 1990. A precore-defective mutant of hepatitis B virus associated with e antigen-negative chronic liver disease. J. Med. Virol. 32:109-118.[Medline]
44
44. Vieth, S., C. Manegold, C. Drosten, T. Nippraschk, and S. Gunther. 2002. Sequence and phylogenetic analysis of hepatitis B virus genotype G isolated in Germany. Virus Genes 24:153-156.[CrossRef][Medline]
45
45. Yuan, T. T., A. Faruqi, J. W. Shih, and C. Shih. 1995. The mechanism of natural occurrence of two closely linked HBV precore predominant mutations. Virology 211:144-156.[CrossRef][Medline]

---

Journal of Virology, September 2007, p. 9202-9215, Vol. 81, No. 17
0022-538X/07/$08.00+0     doi:10.1128/JVI.00390-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

This article has been cited by other articles:

• Garcia, T., Li, J., Sureau, C., Ito, K., Qin, Y., Wands, J., Tong, S. (2009). Drastic Reduction in the Production of Subviral Particles Does Not Impair Hepatitis B Virus Virion Secretion. J. Virol. 83: 11152-11165 [Abstract] [Full Text]  
• Ito, K., Kim, K.-H., Lok, A. S.-F., Tong, S. (2009). Characterization of Genotype-Specific Carboxyl-Terminal Cleavage Sites of Hepatitis B Virus e Antigen Precursor and Identification of Furin as the Candidate Enzyme. J. Virol. 83: 3507-3517 [Abstract] [Full Text]  
• Osiowy, C., Gordon, D., Borlang, J., Giles, E., Villeneuve, J.-P. (2008). Hepatitis B virus genotype G epidemiology and co-infection with genotype A in Canada. J. Gen. Virol. 89: 3009-3015 [Abstract] [Full Text]  

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Li, K. Articles by Tong, S. Search for Related Content PubMed PubMed Citation Articles by Li, K. Articles by Tong, S.