This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lewellyn, E. B.
Right arrow Articles by Loeb, D. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lewellyn, E. B.
Right arrow Articles by Loeb, D. D.

 Previous Article  |  Next Article 

Journal of Virology, June 2007, p. 6207-6215, Vol. 81, No. 12
0022-538X/07/$08.00+0     doi:10.1128/JVI.00210-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Base Pairing between cis-Acting Sequences Contributes to Template Switching during Plus-Strand DNA Synthesis in Human Hepatitis B Virus{triangledown}

Eric B. Lewellyn and Daniel D. Loeb*

McArdle Laboratory for Cancer Research, University of Wisconsin School of Medicine and Public Health, 1400 University Ave., Madison, Wisconsin 53706

Received 30 January 2007/ Accepted 26 March 2007


arrow
ABSTRACT
 
Hepadnaviruses utilize two template switches (primer translocation and circularization) during synthesis of plus-strand DNA to generate a relaxed-circular (RC) DNA genome. In duck hepatitis B virus (DHBV) three cis-acting sequences, 3E, M, and 5E, contribute to both template switches through base pairing, 3E with the 3' portion of M and 5E with the 5' portion of M. Human hepatitis B virus (HBV) also contains multiple cis-acting sequences that contribute to the accumulation of RC DNA, but the mechanisms through which these sequences contribute were previously unknown. Three of the HBV cis-acting sequences (h3E, hM, and h5E) occupy positions equivalent to those of the DHBV 3E, M, and 5E. We present evidence that h3E and hM contribute to the synthesis of RC DNA through base pairing during both primer translocation and circularization. Mutations that disrupt predicted base pairing inhibit both template switches while mutations that restore the predicted base pairing restore function. Therefore, the h3E-hM base pairing appears to be a conserved requirement for template switching during plus-strand DNA synthesis of HBV and DHBV. Also, we show that base pairing is not sufficient to explain the mechanism of h3E and hM, as mutating sequences adjacent to the base pairing regions inhibited both template switches. Finally, we did not identify predicted base pairing between h5E and the hM region, indicating a possible difference between HBV and DHBV. The significance of these similarities and differences between HBV and DHBV will be discussed.


arrow
INTRODUCTION
 
Human hepatitis B virus (HBV) is a major health concern. It is estimated that 350 million people worldwide are chronically infected and have a 15 to 25% chance of premature death due to cirrhosis or hepatocellular carcinoma (29). HBV is a member of the Hepadnaviridae family, which is characterized by tropism for the liver; a double-stranded, relaxed-circular (RC) DNA genome; and DNA replication that involves reverse transcription of an RNA intermediate (for a review, see reference 6). Much of what is known about hepadnavirus biology has come from the study of duck hepatitis B virus (DHBV) and other HBV relatives. Study of animal HBV models has led to many insights, but significant differences exist between HBV and its hepadnavirus relatives. For example, HBV requires about twice as many cis-acting sequences as DHBV does for the synthesis of plus-strand DNA, and the two viruses share no discernible primary nucleotide sequence homology (14, 16, 18, 20, 21). Also, development of pharmaceuticals that are targeted to a specific step during replication will obviously require confirmation that the step is present in HBV. Therefore, validation of these models requires direct testing in HBV.

Reverse transcription of HBV requires three template switches to generate the mature RC DNA genome. This process occurs within the viral nucleocapsid. The template for DNA synthesis is the pregenomic RNA (pgRNA), a greater-than-genome-length molecule with multiple direct repeats of an 11-nucleotide (nt) sequence (DR1 and DR2) and a stem-loop structure near the 5' end ({varepsilon}) (Fig. 1A). Genome replication begins when the viral polymerase (P) associates with {varepsilon} and the complex becomes packaged into a nucleocapsid (2, 3, 12). Tyrosine 63 of the P protein serves as the primer to initiate synthesis of the first strand (minus strand) of DNA (26, 27). The bulge of {varepsilon} serves as the template (24) (Fig. 1A). After the first 3 (or 4) nt are synthesized, the nascent minus-strand DNA switches template to the 3' copy of DR1 and continues DNA synthesis from there (1, 24) (Fig. 1B). RNase H activity of P digests the pgRNA as minus-strand DNA is synthesized. This process produces a single-stranded (SS) DNA replicative intermediate but leaves ~17 nt at the 5' end of the pgRNA undigested (4, 19) (Fig. 1C). This remaining segment of the pgRNA then serves as the primer for plus-strand DNA synthesis, which can initiate at one of two sites (19). In the predominant pathway, the primer initiates plus-strand DNA synthesis at DR2 following a template switch from DR1 to DR2. This process is termed primer translocation (Fig. 1D) (28). The nascent plus-strand DNA is then extended to the 5' end of the minus-strand template. Again, it switches templates, this time from the 5' end (5'r) to the 3' end (3'r) of the minus-strand template, in a process termed circularization (Fig. 1E) (28). Synthesis of plus-strand DNA resumes, and the mature RC DNA genome is made subsequently (Fig. 1F). In the minor pathway, plus-strand DNA synthesis can initiate directly from DR1 in a process called in situ priming (Fig. 1G) (23). This minor pathway produces a double-stranded, duplex linear (DL) DNA (Fig. 1H). Virions containing RC DNA genomes are thought to be efficient at propagating an infection, while virions containing DL DNA genomes are not (6, 31, 32).


Figure 1
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 1. The reverse transcription pathway of HBV. (A) The pgRNA (thin black line) is the template for synthesis of minus-strand DNA. pgRNA is greater than genome length, so the termini are redundant (~110 nt). DR1 and DR2 are 11-nt direct repeat sequences (open boxes), and there are two copies of DR1 due to its location in the terminal redundancy. The stem-loop, {varepsilon}, is indicated. P protein forms a complex with {varepsilon}, and the two are coencapsidated into a newly formed nucleocapsid. DNA synthesis takes place within the nucleocapsid. P protein acts as primer and polymerase to initiate the synthesis of minus-strand DNA. First, the first 3 (or 4) nt are synthesized using the bulge of {varepsilon} as the template. (B) The nascent minus-strand DNA switches template to complementary sequence within DR1, and the synthesis of minus-strand DNA resumes. RNase H activity of P protein degrades the pgRNA as minus-strand DNA is synthesized. (C) The SS DNA intermediate contains a complete minus-strand DNA, and the 5' portion of the pgRNA is not digested. This short RNA (~17 nt) serves as the primer for the initiation of plus-strand DNA synthesis. (D) In the major pathway, the RNA primer switches template from DR1 to DR2 in a process termed primer translocation. Plus-strand DNA synthesis initiates from DR2. (E) When the nascent plus-strand DNA has extended to the 5' end of the minus strand, it switches template from 5'r to 3'r in a process termed circularization. (F) The plus-strand DNA is extended from 3'r to subsequently form the mature RC DNA genome. (G) In the minor pathway, the RNA primer is used to initiate plus-strand DNA synthesis from DR1 in a process termed in situ priming. (H) The mature DL DNA genome results from fully elongated plus-strand DNA primed at DR1.

RC DNA genomes are synthesized more frequently than DL DNA genomes, despite the fact that RC DNA synthesis requires additional template switches. This observation indicates that hepadnaviruses use one or more specific mechanisms for promoting template switching and/or repressing in situ priming. Understanding these mechanisms requires identification and characterization of all cis-acting sequences outside DR1, DR2, 5'r, and 3'r (the donor and acceptor sites for primer translocation and circularization, respectively). In DHBV, cis-acting sequences 3E, M, and 5E contribute to both template switches through base pairing with one another (Fig. 2B) (10, 11, 18, 20, 21). The 3' portion of M base pairs with 3E, and the 5' portion of M base pairs with 5E (18). Previous analyses have also identified multiple cis-acting sequences that are necessary for the accumulation of RC DNA in HBV (Fig. 2A) (16). Although there is no discernible nucleotide sequence similarity between DHBV and HBV, three of these cis-acting sequences in HBV occupy positions equivalent to those of 3E, M, and 5E in DHBV. For ease of presentation, we named these HBV sequences h3E, hM, and h5E, respectively (Fig. 2A). We asked whether these HBV sequences contributed to function through mechanisms similar to those of the DHBV 3E, M, and 5E sequences. We found similarities but also significant differences between HBV and DHBV. Also, we found that base pairing between h3E and hM was necessary for synthesis of RC DNA, a finding previously seen for DHBV (18). We found that h3E and hM contributed to function by means other than base pairing with each other, a finding not reported previously for DHBV (18). We found that h3E and hM contributed to synthesis of RC DNA during both primer translocation and circularization, a finding seen previously with DHBV (18). Finally we found that h5E also contributes to both primer translocation and circularization, but no base pairs between h5E and hM were identified, indicating a significant difference from DHBV. Our findings with HBV reinforce the concept that the topology of the minus-strand template makes one or more important contributions to the synthesis of plus-strand DNA. However, the inability to identify base pairs between hM and h5E and the presence of more cis-acting sequences in HBV than in DHBV strongly suggest that the role of base pairing cannot be to simply juxtapose the end of the minus-strand template to facilitate template switching.


Figure 2
View larger version (12K):
[in this window]
[in a new window]

  		FIG. 2. Locations of cis-acting sequences that contribute to the accumulation of RC DNA in HBV and DHBV relative to minus-strand DNA. (A) The locations of HBV cis-acting sequences that contribute to the accumulation of RC DNA are indicated (14, 16). Accumulation of RC DNA is less than 12% (black) or less than 75% (dark gray) of the WT reference level. Light gray indicates the previously understood boundaries of h3E and hM prior to this study. Previous mapping studies did not investigate the regions from nt 1845 to 1913 and 1745 to 1814 due to the presence of {varepsilon} and other cis-acting sequences needed for minus-strand DNA synthesis (1, 2) (B) As a comparison, the locations of DHBV cis-acting sequences 3E, M, and 5E, which contribute to the accumulation of RC DNA (18). The M region can be further subdivided into M3 and M5, which make somewhat different contributions to the accumulation of RC DNA.


arrow
MATERIALS AND METHODS
 
Molecular clones. All molecular clones were derived from the HBV ayw subtype genome (GenBank accession number V01460). The C of the EcoRI site (GAATTC) was designated nucleotide position 1.

The plasmid that expressed our wild-type (WT) HBV reference contained 1.1 copies of the HBV genome. This plasmid expressed pgRNA under the control of the cytomegalovirus immediate-early promoter. This plasmid, named NL84, did not express C, P, X, or envelope proteins and was described previously (1). All variants were derived from NL84. The HBV P, C, and X proteins were expressed from the plasmid LJ96 (16).

Variants were generated by oligonucleotide, site-directed mutagenesis via PCR. All molecular clones were confirmed by sequencing across the entire PCR insert. Deletion variants are indicated by the "{Delta}" prefix and substitution variants by the "sub" prefix. The nucleotide coordinates for each variant indicate the first and last nucleotide altered. In all substitution variants, each altered nucleotide was changed to its Watson-Crick pairing partner (i.e., A to T, C to G, etc.).

The internal standard used in our primer extension analyses was derived from a plasmid containing HBV sequence from nt 1067 to 1983. This plasmid was digested with the restriction endonuclease StyI, which cuts at HBV nt 1641 and 1884.

Cell culture and transfection. The human hepatoblastoma cell line HepG2 was used in all analyses. These cells were grown in RPMI 1640 medium supplemented with 10% fetal bovine serum, 0.01 M HEPES, and 0.1 mM nonessential amino acids for minimal essential medium. Cells were grown to approximately 75% confluence in 60-mm cell culture dishes, and medium was changed to Dulbecco modified Eagle medium-F-12 with 5% fetal bovine serum several minutes prior to transfection. We transfected each cell culture with 4.5 µg pgRNA expression plasmid and 4.5 µg HBV replication protein expression plasmid LJ96. To monitor the efficiency of each transfection, 1 µg of a green fluorescent protein expression plasmid was included. We used a CaPO4 precipitation procedure as described previously (17). Medium was changed back to RPMI 1640 with supplementation 15 h after the transfection and changed every 2 days until the cultures were harvested at day 6. All cell culture reagents were purchased from Gibco/Invitrogen (Carlsbad, CA).

Isolation of viral replicative intermediates. Isolation of viral replicative intermediate DNA was performed as described previously (16). Cell cultures were washed with a solution of HEPES-buffered saline plus EGTA (2 mM HEPES, 150 mM NaCl, 0.5 mM EGTA [pH 7.45]) and then frozen to –70°C. Later, cells were thawed to room temperature and lysed with 0.2% NP-40 in 50 mM Tris-HCl and 1 mM EDTA (pH 8.0) for 10 min. Then nuclei were removed by centrifugation at 16,000 x g for 4 minutes at 4°C. Supernatants were adjusted to 2 mM CaCl2 and then incubated with 44 units of micrococcal nuclease (Worthington, Lakewood, NJ) to digest transfected plasmid DNA and unencapsidated HBV RNA. After 2 hours, these reaction mixtures were supplemented with EDTA to 10 mM, sodium dodecyl sulfate (SDS) to 0.4%, and pronase to 400 µg/ml (Roche, Basel, Switzerland). These additions terminated the micrococcal nuclease digestion and digested nucleocapsids and the P protein. Viral DNA was then extracted with a phenol-chloroform mixture and precipitated with ethanol.

Southern blot analyses. Viral DNA was electrophoresed through an 11-cm x 14-cm x 0.5-cm 1.25% agarose gel in 1x Tris-borate-EDTA buffer at 24 V for 24 h. DNA was then transferred by capillary action to a Hybond-N membrane (Amersham, Buckinghamshire, United Kingdom) and cross-linked to the membrane with UV light. Membranes were incubated in Church hybridization solution (10 mM EDTA, 1% bovine serum albumin, 0.5 M NaHPO4, 7% SDS [pH 7.2]) for 1 hour at 65°C prior to hybridization of the HBV probe. A 350-nt riboprobe, specific for the minus strand from nt 3096, was used. We used this probe to avoid detection of spliced RNAs that have been encapsidated and reverse transcribed (7, 13, 15, 25, 30). The riboprobe was labeled with [{alpha}-32P]UTP and transcribed in vitro with T7 RNA polymerase. The membrane and probe were hybridized at 65° overnight in Church hybridization solution. Membranes were then washed in Church wash solution (1% SDS, 20 mM Na2HPO4, 1 mM EDTA) until no radioactivity was detected in the wash. Autoradiographic images of the membrane were obtained using a GE Healthcare Typhoon phosphorimager and quantitated using ImageQuant 5.2 software (GE Healthcare).

Primer extension analysis. Three oligonucleotide DNA primers were end labeled with [{gamma}-32P]ATP. Oligonucleotide 1 (5'-CTCTTGGACTCTCAGCAATGTCAAC-3') annealed to minus-strand DNA beginning at nt 1661 and was used to measure the level of minus-strand DNA. Oligonucleotide 2 (5'-GCATGGTGCTGGTGC-3') annealed to plus-strand DNA at nt 1815 and was used to measure the level of plus-strand DNA that had undergone primer translocation and extended to at least nt 1815. Oligonucleotide 3 (5'-AGGACATGAACAAGAGATGATTAG-3') annealed to plus-strand DNA at nt 1859 and was used to measure the level of plus-strand DNA that had undergone primer translocation and circularization and extended to at least nt 1859. For analyses of each variant by primer extension, 600 pg of internal standard DNA was added to each viral DNA. Next, the mixture of DNA was heated to 95°C for 5 min, treated with 3 µg RNase A for 1 h at 37°C, precipitated with ethanol, and resuspended in H2O. The rehydrated sample was split into three equal portions, and primer extension analysis was performed on each using the three oligonucleotides, one per reaction, with Vent Exo– polymerase (New England Biolabs, Ipswich, MA). The sizes of products for oligonucleotide 1, oligonucleotide 2, and oligonucleotide 3 extended on viral DNA are 155 nt, 175 nt, and 219 nt, respectively. Products for oligonucleotide 1, oligonucleotide 2, and oligonucleotide 3 extended on internal standard DNA are 224 nt, 217 nt, and 261 nt, respectively. Products of the primer extension reactions were electrophoresed through 5% polyacrylamide gels with 7.6 M urea. Gels were dried, and autoradiography was performed using a phosphorimager as described for Southern blot analysis.

Statistical analysis. We conducted statistical analyses on three parameters, RC DNA accumulation efficiency (determined by Southern blotting), primer translocation efficiency (determined by primer extension), and circularization efficiency (also determined by primer extension). For each parameter, we compared each variant to the WT reference using a two-sided Wilcoxon signed-rank test. Paired samples were defined as a WT reference and a variant that were transfected at the same time and analyzed in the same experiment. If multiple culture dishes were transfected with WT reference genomes and analyzed in a single experiment, we calculated the average value for the WT reference. Signed-rank tests were performed on the data before normalization to the WT reference, and all samples had n = 6 or more.

An additional comparison was made for substitution variants in set C and set D (see Fig. 4A) that disrupt, then restore, putative base pairing between h3E and hM. For each parameter, we compared each single substitution to the corresponding double substitution using the Wilcoxon signed-rank test as described above. A pair was defined as a single substitution and corresponding double substitution from the same transfection that were analyzed in the same experiment. Again, all samples had n = 6 or more. For all tests, we considered P < 0.05 to be significant. All samples were statistically analyzed, unless otherwise mentioned.


Figure 4
View larger version (14K):
[in this window]
[in a new window]

  		FIG. 4. Base pairing between h3E and hM is necessary for the accumulation of RC DNA. (A) Predicted base pairing between h3E and hM. The substitution variants at either h3E or hM to test the contribution of the predicted base pairs are shown. (B) Southern blot analysis of substitution variants. Novel splice forms are indicated by asterisks. (C) Histogram of RC DNA accumulation efficiency for each variant. Error bars indicate standard deviations. Each variant was analyzed at least six times.


arrow
RESULTS
 
Rationale. Previous analyses identified five regions in the HBV genome, besides the donor and acceptor sites, that contribute to the accumulation of RC DNA (Fig. 2A) (14, 16, 22). However, this analysis did not determine how these sequences were acting. Our study focused on the regions from nt 1833 to 1844 (h3E), nt 2806 to 2971 (hM), and nt 1511 to 1566 (h5E) (Fig. 2A). First, we wanted to define further the boundaries of hM and h5E. Because h3E previously was determined to lie within an 11-nt region (16), we did not pursue finer mapping of its boundaries. Then, we wanted to determine whether these regions base paired with one another and whether they contributed to primer translocation and/or circularization.

To study DNA synthesis, HepG2 cells were cotransfected with separate plasmids expressing pgRNA or replication proteins. This strategy was used to prevent mutant proteins from being expressed. Newly synthesized viral DNA was then isolated 6 days later and analyzed by Southern blotting or primer extension.

hM mapped to nt 2820 to 2868. Previously, the hM region was found to lie within nt 2806 to 2994 (16). To determine the boundaries of the hM region more precisely, we generated a series of substitution and deletion variants that collectively represent this region (Fig. 3A). We analyzed these variants by Southern blotting and determined their ability to accumulate RC DNA (Fig. 3B). We used a hybridization probe that would not detect DNA synthesized from encapsidated, spliced RNAs (7, 13, 15, 25, 30). Detection of these DNA forms would complicate the analyses. The relative level of accumulation of RC DNA for each variant was calculated using the relationship shown in Fig. 3C. Deletions or substitutions within nt 2820 to 2868 accumulated less RC DNA than the WT reference whereas mutations within nt 2806 to 2819 and 2870 to 2994 did not affect accumulation of RC DNA (Fig. 3B). Mutations between nt 2832 and 2847 had the largest decreases in RC DNA (~12% of the level of the WT reference) (Fig. 3B, lanes 5 to 7). The variant sub2830-2831 accumulated RC DNA at 50% of the level of the WT reference (Fig. 3B, lane 4). Variants {Delta}2820-2829 and {Delta}2853-2868 accumulated RC DNA at greater than 80% of the level of the WT reference (Fig. 3B, lanes 3 and 9). Interestingly, two variants within this region, sub2832-2836 and {Delta}2848-2853, synthesized two new DNA forms that migrated faster than DL DNA and faster than SS DNA, respectively (Fig. 3B, lanes 5 and 8). These DNA forms were identified as reverse transcripts of a novel spliced RNA with a splice junction at nt 2445/2845 (data not shown). Thus, the variants sub2832-2836 and {Delta}2848-2853 could not be accurately quantitated. However, it was clear by visual inspection of the Southern blot that sub2832-2836 accumulated little to no RC DNA (Fig. 3B, lane 5). In summary, our analyses indicated that hM was primarily contained within nt 2832 to 2847 with a very minor contribution made by the 10 to 20 nt on either side.


Figure 3
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 3. The hM region of HBV is localized to a short sequence. (A) Positions of deletion and substitution variants within the minus-strand DNA. The region from nt 2806 to 2994 is expanded to show relative sizes and positions of deletions and substitutions. The level of accumulation of RC DNA for each variant relative to the level of the WT reference and its standard deviation are shown on the right. Each variant was independently analyzed at least six times. NA, not available. (B) Representative Southern blot of hM region variants. RC, DL, and SS DNA forms are indicated. Asterisks indicate a new DNA form, seen only in some hM region variants, that is derived from a spliced RNA that has been encapsidated and reverse transcribed. The hybridization probe detected 350 nt of minus-strand DNA starting at nt 3096. (C) Calculation to determine the level of accumulation of RC DNA of each variant relative to the WT reference. The value "RC DNA" is the measurement of the DNA form labeled RC. The value "total minus-strand DNA" is the measurement of all DNA species between (and including) RC and SS DNA.

Base pairing between h3E and hM is necessary for the synthesis of RC DNA. With the refined map of the hM region in hand, we identified a short region of potential base pairing between h3E and hM (Fig. 4A). To determine whether this predicted base pairing contributes to accumulation of RC DNA, we generated four sets of variants (A to D). Each set contained substitutions at either h3E, hM, or both such that base pairing was disrupted and then restored (Fig. 4A). Southern blot analysis showed that single substitutions at either h3E or hM caused a pronounced decrease in the accumulation of RC DNA in all four sets. Double substitutions of variant sets A, B, and C restored accumulation of RC DNA, as seen in Fig. 4B (lanes 4, 7, and 10). The double substitution h3ED/hMD partially restored accumulation of RC DNA (Fig. 4B, lane 13). We did not measure the magnitude of defects in sets A and B due to the presence of the aforementioned new DNA forms, but sets C and D were quantitated. The double substitution h3EC/hMC restored the accumulation of RC DNA to near-WT levels. In contrast, h3ED/hMD restored accumulation of RC DNA but not to the level of the WT reference. In addition, variant {Delta}2841-2847, whose mutation lies outside the region of base pairing, accumulated little to no RC DNA (Fig. 3B, lane 7). These two variants, h3ED/hMD and {Delta}2841-2847, indicated that hM base pairing with h3E does not completely explain the mechanism by which hM works. Overall, these findings indicate that base pairing between h3E and hM is necessary for the accumulation of RC DNA but that additional factors may also be required.

h3E and hM regions contribute to primer translocation. In DHBV, mutations in either 3E or M affected both primer translocation and circularization (18). First, we asked whether h3E or hM contributes to primer translocation in HBV. We used a primer extension assay to measure primer translocation. In this assay, two oligonucleotide DNA primers (oligonucleotides 1 and 2) were extended on a mixture of viral DNA and an internal standard DNA. Oligonucleotides 1 and 2 were used in separate primer extension reactions. Oligonucleotide 1 was used to measure the level of minus-strand DNA, and oligonucleotide 2 was used to measure the level of plus-strand DNA that had undergone primer translocation and extended to at least nt 1815 (the circularization point). The relative annealing sites of oligonucleotides 1 and 2 are indicated in Fig. 5A. First, the level of viral DNA measured by each primer was normalized to the level of the internal standard. Then the level of primer translocation was determined by dividing the amount of total priming from DR2 (oligonucleotide 2) by the level of minus-strand DNA (oligonucleotide 1) in the sample. The primer translocation for each variant was then compared to the WT reference. This relationship is shown in Fig. 5G. A representative primer extension analysis with substitution set C and the WT reference is shown in Fig. 5.


Figure 5
View larger version (23K):
[in this window]
[in a new window]

  		FIG. 5. h3E and hM contribute to primer translocation and circularization. (A) Schematic representation of RC, DL, and SS DNA with relative positions of oligonucleotides used for primer extension analysis. (B to D) Primer extension gel with oligonucleotides 1, 2, and 3, respectively. The sequencing ladder is on the left. Bands representing replicative intermediate (R.I.) and internal standard (I.S.) DNA are indicated. Each replicative intermediate signal is normalized to the internal standard signal. The sizes of products for oligonucleotide 1, oligonucleotide 2, and oligonucleotide 3 extended on viral DNA are 155 nt, 175 nt, and 219 nt, respectively. Products for oligonucleotide 1, oligonucleotide 2, and oligonucleotide 3 extended on internal standard DNA are 224 nt, 217 nt, and 261 nt, respectively. (E and F) Graphical representations of primer translocation (P.T.) efficiency (E) and circularization efficiency (F) are shown. Each variant was analyzed at least six times. (G) Calculation to determine the relative level of primer translocation. We defined primer translocation as the proportion of replicative intermediates containing minus-strand DNA that had initiated plus-strand DNA from DR2 and elongated it up to at least the circularization point (about 217 nt from DR2). (H) Calculation to determine the relative level of circularization. We defined circularization as the fraction of plus-strand DNA having undergone primer translocation that also circularized and extended to nt 1859 (34 nt from the circularization point).

The level of primer translocation was determined for substitution sets C and D and deletion {Delta}2841-2847 (Fig. 5E). The substitutions at either h3E or hM and the variant {Delta}2841-2847 had defects in primer translocation, approximately 17% to 25% of the level of the WT reference. As predicted, the double substitution of set C showed almost complete restoration of primer translocation (Fig. 5C). In summary, these findings indicate that h3E and hM contribute to primer translocation. Whether the lack of RC DNA seen by Southern blotting could be attributed entirely to a defect in primer translocation remained to be determined.

The h3E and hM regions also contribute to circularization. Next, we asked whether h3E or hM also contributes to circularization. Again, we utilized primer extension to determine the level of circularization with oligonucleotide 2 (previously mentioned) and oligonucleotide 3. Oligonucleotide 3 annealed to plus-strand DNA at nt 1859 (Fig. 5A). First, the level of viral DNA was normalized to the internal standard for oligonucleotides 2 and 3. Then, the level of DNA measured by oligonucleotide 3 was divided by the level of DNA measured by oligonucleotide 2. The level of circularization for each variant was compared to the WT reference. This relationship is shown in Fig. 5H.

The level of circularization was affected for all variants analyzed (Fig. 5F). For variant {Delta}2841-2847, the level of circularization was 3% of the level of the WT reference. All mutations in the hM and h3E regions resulted in low but detectable levels of circularization, ranging from 6% (h3ED) to 27% (hMC) of the level of the WT reference. As predicted, the double substitution variant of set C almost completely restored the level of circularization (86% of the WT reference versus 25% and 27% for each single substitution). The double substitution of set D had a modest restoration in circularization compared to the constituent single substitutions (23% versus 6% and 11%). This analysis indicates that h3E and hM contribute to circularization. Overall, these regions contributed to both template switches, as was found in DHBV (10, 18).

h5E contributes to primer translocation and circularization. Liu et al. (16) identified a cis-acting sequence located within nt 1511 to 1568 (h5E). To understand this region better, we analyzed {Delta}1511-1568 as well as three smaller deletions, {Delta}1511-1532, {Delta}1533-1553, and {Delta}1554-1566, by Southern blotting and primer extension (Fig. 6A). First, we found that these four variants accumulated less RC DNA than the WT reference. Variants {Delta}1511-1568, {Delta}1511-1532, and {Delta}1533-1553 accumulated RC DNA to less than 50% of the level of the WT reference. Variant {Delta}1554-1566 accumulated RC DNA to 78% of the level of the WT reference (Fig. 6B). When we examined these variants for their ability to carry out primer translocation, we found that all four variants were affected. The levels of primer translocation for {Delta}1511-1568, {Delta}1511-1532, {Delta}1533-1553, and {Delta}1554-1566 were 45%, 56%, 53%, and 79%, respectively, relative to the level of the WT reference (Fig. 6C). Next, we measured the level of circularization for each variant by primer extension. We found that all four variants had little to no decrease in their levels of circularization. Three of the variants, {Delta}1511-1568, {Delta}1511-1532, and {Delta}1533-1553, had small decreases in their levels of circularization (77%, 84%, and 77%, respectively) (Fig. 6D). The level of circularization for {Delta}1554-1566 was not different from that of the WT reference (Fig. 6D). These findings indicate that the h5E cis-acting sequence contributes to the synthesis of RC DNA at both primer translocation and circularization, but the impact on primer translocation is greater.


Figure 6
View larger version (13K):
[in this window]
[in a new window]

  		FIG. 6. Efficiency of RC accumulation, primer translocation, and circularization for h5E region variants. (A) Positions of deletion and substitution variants within the minus-strand DNA. The region from nt 1511 to 1597 is expanded to show relative sizes and positions of deletions. (B to D) Histograms indicate the level of accumulation of RC DNA (B), primer translocation (PT) (C), and circularization (D) relative to the level of the WT reference. Error bars indicate standard deviations. Each variant was analyzed at least six times.


arrow
DISCUSSION
 
In this study, we characterized three HBV cis-acting sequences, h3E, hM, and h5E, that are necessary for efficient synthesis of RC DNA. We found that base pairing between two of these sequences, namely, h3E and hM, contributes to both primer translocation and circularization during plus-strand DNA synthesis. Additionally, we found that nucleotides within hM that do not form base pairs with h3E are also necessary for efficient template switching. Therefore, it appears that hM and h3E contribute to template switching through both base pairing with one another and additional mechanisms that may include base pairing with other as-of-yet-unidentified complementary regions of the genome or non-base-pairing mechanisms. Finally, we found that h5E also contributes to primer translocation and, to a lesser extent, circularization. We were unable to find base pairing between h5E and hM, indicating a significant difference between HBV and DHBV.

When the cis-acting sequences in HBV and DHBV are compared, several similarities and differences are noteworthy. First, efficient synthesis of RC DNA in both viruses requires cis-acting sequences outside the donor and acceptor sites for template switching. However, HBV appears to require at least five cis-acting sequences (16) while DHBV requires only three (11, 18, 20) (Fig. 2). Interestingly, three of the HBV cis-acting sequences occupy positions equivalent to the 3E, M, and 5E positions in DHBV (16, 18). Furthermore, mutation of h3E or hM in HBV and 3E or M in DHBV reduces the synthesis of RC DNA to undetectable levels during both primer translocation and circularization (Fig. 5B) (11, 18, 20). Mutation of h5E in HBV and 5E in DHBV causes a partial reduction in the synthesis of RC DNA; however, the defect in DHBV occurs primarily during circularization while the defect in HBV occurs primarily during primer translocation (Fig. 6A) (11, 18, 20).

When the mechanisms of h3E and hM in HBV and 3E and M in DHBV are compared, similarities are apparent. In both these viruses, the ability of base pairs to form between the 3' end and the middle of the minus-strand DNA is critically important for template switching (Fig. 5) (18). This finding suggests that base pairing between h3E and hM in HBV and base pairing between 3E and M in DHBV may perform similar functions. Since DHBV and HBV are distantly related hepadnaviruses, this function may be a conserved feature of most, if not all, hepadnaviruses. Consistent with this prediction, we found putative base pairing regions akin to h3E and hM in woolly monkey hepatitis B virus, woodchuck hepatitis B virus, ground squirrel hepatitis B virus, heron hepatitis B virus, and Ross goose hepatitis B virus, although the functional relevance of these putative base pairs has not been tested.

An apparent difference between HBV and DHBV is that we did not identify base pairing between the HBV h5E and hM regions. Although hM has a 5' portion that does not base pair with h3E, we could not identify potential base pairs between the hM and h5E regions. This apparent lack of base pairing between hM and h5E suggests that DHBV 5E and HBV h5E may not perform the same role in template switching or that HBV h5E has evolved a mechanism other than base pairing with hM to carry out its function.

Although we have established that base pairing between h3E and hM contributes to both plus-strand template switches, how this base pairing contributes is unclear. During template switching, several events must occur. Base pairing at the donor site must be disrupted, the primer or nascent strand must transfer to the opposite end of the genome and reanneal, and the viral polymerase must reorient and initiate DNA synthesis. While many of these steps likely require catalysis by a protein or proteins within the capsid, Liu et al. (18) suggested that a tertiary DNA conformation that juxtaposed the donor and acceptor sites may also be necessary, and base pairing between cis-acting sequences may establish that conformation. This model seemed a reasonable explanation for DHBV since 5E and 3E were near the donor and acceptor sites, and when both of them base paired with M, the donor and acceptor sites were brought into proximity (18). However, in light of our findings presented here, the applicability of this putative model for HBV must be reconsidered. First, since there does not appear to be base pairing between h5E and hM in HBV, it is unclear how the base pairing between only h3E and hM could juxtapose the donor and acceptor sites. Therefore, to assert that h3E, hM, and/or h5E contributes to template switching by stabilizing a specific DNA conformation, one would also need to assert that additional interactions were involved. These interactions could involve DNA-protein interaction or other, as-of-yet-unidentified DNA-DNA interactions, and our findings that cis-acting sequences that do not form base pairs are also necessary for template switching are consistent with this idea. However, no such interactions have been identified to date.

An alternative model is that the formation of base pairs between h3E and hM serves to offset the net loss of ~6 base pairs during primer translocation. This net loss occurs because the ~17-nt RNA primer can form only 11 base pairs at DR2. However, if the formation of ~8 base pairs between h3E and hM was coupled to this process via an as-of-yet-unidentified mechanism, there would be a net gain of base pairs. Therefore, the process would be energetically favorable.

Finally, it is possible that the base pairing is actually contributing to initiation (or reinitiation) of DNA synthesis after the respective template switches. Studies of the TY1 retrotransposon, which also undergoes template switching during its reverse transcription, have shown that base pairing between the ends of the TY1 genomic RNA likely contributes to initiation of minus-strand DNA synthesis (5). Since our primer extension assay does not distinguish between annealing of the transferring strand to its new template and subsequent initiation of DNA synthesis, this possibility cannot be ruled out for HBV.

Throughout these studies, we consistently found that decreases in the accumulation of RC DNA were not accompanied by equivalent increases in the accumulation of DL DNA. This observation suggests that viral replicative intermediates that do not follow the RC DNA synthesis pathway do not necessarily follow the DL DNA synthesis pathway by default. While this could indicate that there is some inherent inefficiency associated with in situ priming, DL DNA elongation, or the stability of DL DNA, it could also indicate that there are mechanisms to specifically inhibit priming of plus-strand DNA at DR1. In DHBV, a DNA stem-loop at the 5' side of DR1 inhibits in situ priming (8, 9). No equivalent stem-loop is seen in HBV. However, we observed by Southern blotting that mutations within either h3E or hM had up-to-threefold increases in the accumulation of DL DNA (data not shown). Since the h3E-hM base pairing region is within 9 nt of DR1, it may act as an inhibitor of in situ priming in HBV in addition to its role in promoting primer translocation and circularization.

In the context of ongoing efforts to develop new treatments for chronic HBV infection, one needs to consider the potential implications of our study. We have identified a previously unknown process in HBV replication, base pairing between h3E and hM, which is indispensable for mature genome formation. A drug or small molecule that would interfere with this base pairing should have profound effects on the synthesis of RC DNA. Such a drug or small molecule would complement the existing anti-HBV therapies which are comprised largely of nucleoside analog chain terminators.


arrow
ACKNOWLEDGMENTS
 
We thank Brian Olsen and Chris Black for aid in generating some HBV variants. We are particularly grateful to Teresa Abraham, Cathy Grutzmacher, Katy Haines, Thomas Lentz, and Megan Maguire for helpful discussion throughout the course of these studies and the thoughtful and critical review of the manuscript.

This work was supported by NIH grant CA22443.


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: McArdle Laboratory for Cancer Research, University of Wisconsin School of Medicine and Public Health, 1400 University Ave., Madison, WI 53706. Phone: (608) 262-1260. Fax: (608) 262-2824. E-mail: loeb{at}oncology.wisc.edu Back

{triangledown} Published ahead of print on 4 April 2007. Back


arrow
REFERENCES
 
    1
  1. Abraham, T. M., and D. D. Loeb. 2006. Base pairing between the 5' half of {varepsilon} and a cis-acting sequence, {Phi}, makes a contribution to the synthesis of minus-strand DNA for human hepatitis B virus. J. Virol. 80:4380-4387.[Abstract/Free Full Text]
    2. 2
  2. Bartenschlager, R., M. Junker-Niepmann, and H. Schaller. 1990. The P gene product of hepatitis B virus is required as a structural component for genomic RNA encapsidation. J. Virol. 64:5324-5332.[Abstract/Free Full Text]
    3. 3
  3. Bartenschlager, R., and H. Schaller. 1992. Hepadnaviral assembly is initiated by polymerase binding to the encapsidation signal in the viral RNA genome. EMBO J. 11:3413-3420.[Medline]
    4. 4
  4. Chen, Y., and P. L. Marion. 1996. Amino acids essential for RNase H activity of hepadnaviruses are also required for efficient elongation of minus-strand viral DNA. J. Virol. 70:6151-6156.[Abstract]
    5. 5
  5. Cristofari, G., C. Bampi, M. Wilhelm, F. X. Wilhelm, and J. L. Darlix. 2002. A 5'-3' long-range interaction in Ty1 RNA controls its reverse transcription and retrotransposition. EMBO J. 21:4368-4379.[CrossRef][Medline]
    6. 6
  6. Ganem, D., and R. J. Schneider. 2001. Hepadnaviridae: the viruses and their replication, p. 2923-2957. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th ed., Lippincott Williams & Wilkins, Philadelphia, PA.
    7. 7
  7. Gunther, S., G. Sommer, A. Iwanska, and H. Will. 1997. Heterogeneity and common features of defective hepatitis B virus genomes derived from spliced pregenomic RNA. Virology 238:363-371.[CrossRef][Medline]
    8. 8
  8. Habig, J. W., and D. D. Loeb. 2003. The conformation of the 3' end of the minus-strand DNA makes multiple contributions to template switches during plus-strand DNA synthesis of duck hepatitis B virus. J. Virol. 77:12401-12411.[Abstract/Free Full Text]
    9. 9
  9. Habig, J. W., and D. D. Loeb. 2002. Small DNA hairpin negatively regulates in situ priming during duck hepatitis B virus reverse transcription. J. Virol. 76:980-989.[Abstract/Free Full Text]
    10. 10
  10. Havert, M. B., L. Ji, and D. D. Loeb. 2002. Analysis of duck hepatitis B virus reverse transcription indicates a common mechanism for the two template switches during plus-strand DNA synthesis. J. Virol. 76:2763-2769.[Abstract/Free Full Text]
    11. 11
  11. Havert, M. B., and D. D. Loeb. 1997. cis-acting sequences in addition to donor and acceptor sites are required for template switching during synthesis of plus-strand DNA for duck hepatitis B virus. J. Virol. 71:5336-5344.[Abstract]
    12. 12
  12. Hirsch, R. C., D. D. Loeb, J. R. Pollack, and D. Ganem. 1991. cis-acting sequences required for encapsidation of duck hepatitis B virus pregenomic RNA. J. Virol. 65:3309-3316.[Abstract/Free Full Text]
    13. 13
  13. Kock, J., M. Nassal, K. Deres, H. E. Blum, and F. von Weizsacker. 2004. Hepatitis B virus nucleocapsids formed by carboxy-terminally mutated core proteins contain spliced viral genomes but lack full-size DNA. J. Virol. 78:13812-13818.[Abstract/Free Full Text]
    14. 14
  14. Lee, J., M. K. Shin, H. J. Lee, G. Yoon, and W. S. Ryu. 2004. Three novel cis-acting elements required for efficient plus-strand DNA synthesis of the hepatitis B virus genome. J. Virol. 78:7455-7464.[Abstract/Free Full Text]
    15. 15
  15. Le Pogam, S., P. K. Chua, M. Newman, and C. Shih. 2005. Exposure of RNA templates and encapsidation of spliced viral RNA are influenced by the arginine-rich domain of human hepatitis B virus core antigen (HBcAg 165-173). J. Virol. 79:1871-1887.[Abstract/Free Full Text]
    16. 16
  16. Liu, N., L. Ji, M. L. Maguire, and D. D. Loeb. 2004. cis-acting sequences that contribute to the synthesis of relaxed-circular DNA of human hepatitis B virus. J. Virol. 78:642-649.[Abstract/Free Full Text]
    17. 17
  17. Liu, N., K. M. Ostrow, and D. D. Loeb. 2002. Identification and characterization of a novel replicative intermediate of heron hepatitis B virus. Virology 295:348-359.[CrossRef][Medline]
    18. 18
  18. Liu, N., R. Tian, and D. D. Loeb. 2003. Base pairing among three cis-acting sequences contributes to template switching during hepadnavirus reverse transcription. Proc. Natl. Acad. Sci. USA 100:1984-1989.[Abstract/Free Full Text]
    19. 19
  19. Loeb, D. D., R. C. Hirsch, and D. Ganem. 1991. Sequence-independent RNA cleavages generate the primers for plus-strand DNA synthesis in hepatitis B viruses—implications for other reverse transcribing elements. EMBO J. 10:3533-3540.[Medline]
    20. 20
  20. Mueller-Hill, K., and D. D. Loeb. 1996. Previously unsuspected cis-acting sequences for DNA replication revealed by characterization of a chimeric heron/duck hepatitis B virus. J. Virol. 70:8310-8317.[Abstract]
    21. 21
  21. Mueller-Hill, K., and D. D. Loeb. 2002. cis-acting sequences 5E, M, and 3E interact to contribute to primer translocation and circularization during reverse transcription of avian hepadnavirus DNA. J. Virol. 76:4260-4266.[Abstract/Free Full Text]
    22. 22
  22. Perri, S., and D. Ganem. 1997. Effects of mutations within and adjacent to the terminal repeats of hepatitis B virus pregenomic RNA on viral DNA synthesis. J. Virol. 71:8448-8455.[Abstract]
    23. 23
  23. Staprans, S., D. D. Loeb, and D. Ganem. 1991. Mutations affecting hepadnavirus plus-strand DNA synthesis dissociate primer cleavage from translocation and reveal the origin of linear viral DNA. J. Virol. 65:1255-1262.[Abstract/Free Full Text]
    24. 24
  24. Tavis, J. E., S. Perri, and D. Ganem. 1994. Hepadnavirus reverse transcription initiates within the stem-loop of the RNA packaging signal and employs a novel strand transfer. J. Virol. 68:3536-3543.[Abstract/Free Full Text]
    25. 25
  25. Terre, S., M. A. Petit, and C. Brechot. 1991. Defective hepatitis B virus particles are generated by packaging and reverse transcription of spliced viral RNAs in vivo. J. Virol. 65:5539-5543.[Abstract/Free Full Text]
    26. 26
  26. Wang, G. H., and C. Seeger. 1993. Novel mechanism for reverse transcription in hepatitis B viruses. J. Virol. 67:6507-6512.[Abstract/Free Full Text]
    27. 27
  27. Wang, G. H., and C. Seeger. 1992. The reverse transcriptase of hepatitis-B virus acts as a protein primer for viral DNA synthesis. Cell 71:663-670.[CrossRef][Medline]
    28. 28
  28. Will, H., W. Reiser, T. Weimer, E. Pfaff, M. Buscher, R. Sprengel, R. Cattaneo, and H. Schaller. 1987. Replication strategy of human hepatitis B virus. J. Virol. 61:904-911.[Abstract/Free Full Text]
    29. 29
  29. World Health Organization. 2004. Hepatitis B vaccines. Wkly. Epidemiol. Rec. 79:255-263.[Medline]
    30. 30
  30. Wu, H. L., P. J. Chen, S. J. Tu, M. H. Lin, M. Y. Lai, and D. S. Chen. 1991. Characterization and genetic analysis of alternatively spliced transcripts of hepatitis B virus in infected human liver tissues and transfected HepG2 cells. J. Virol. 65:1680-1686.[Abstract/Free Full Text]
    31. 31
  31. Yang, W. G., and J. Summers. 1995. Illegitimate replication of linear hepadnavirus DNA through nonhomologous recombination. J. Virol. 69:4029-4036.[Abstract]
    32. 32
  32. Yang, W. G., and J. Summers. 1998. Infection of ducklings with virus particles containing linear double-stranded duck hepatitis B virus DNA: illegitimate replication and reversion. J. Virol. 72:8710-8717.[Abstract/Free Full Text]

Journal of Virology, June 2007, p. 6207-6215, Vol. 81, No. 12
0022-538X/07/$08.00+0     doi:10.1128/JVI.00210-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Moreno, J. L., Zuniga, S., Enjuanes, L., Sola, I. (2008). Identification of a Coronavirus Transcription Enhancer. J. Virol. 82: 3882-3893 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lewellyn, E. B.
Right arrow Articles by Loeb, D. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lewellyn, E. B.
Right arrow Articles by Loeb, D. D.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»