The Duck Hepatitis B Virus Reverse Transcriptase Functions as a Full-length Monomer
2006; Elsevier BV; Volume: 281; Issue: 47 Linguagem: Inglês
10.1074/jbc.m608031200
ISSN1083-351X
Autores Tópico(s)Viral gastroenteritis research and epidemiology
ResumoHepadnaviral reverse transcription occurs within cytoplasmic capsid particles and is catalyzed by a virally encoded reverse transcriptase, but the primary structure and multimeric state of the polymerase during reverse transcription are poorly understood. We measured these parameters for the duck hepatitis B virus polymerase employing active enzyme translated in vitro and derived from intracellular core particles and mature virions. In vitro-translated polymerase immunoprecipitated as a monomer, and polymerase molecules with complementary defects in the enzymatic active site and tyrosine 96, which primes DNA synthesis, could not complement or inhibit each other in priming assays. Western analysis using antibodies recognizing epitopes throughout the polymerase combined with nuclease digestion of permeabilized virion-derived capsid particles revealed that only full-length polymerase molecules were in virions and that they were all covalently attached to large DNA molecules. Because DNA synthesis is primed by the polymerase itself and only one copy of the viral DNA is in each capsid, the polymerase must function as an uncleaved monomer. Therefore, a single polymerase monomer is encapsidated, primes DNA synthesis, synthesizes both DNA strands, and participates in the three-strand transfers of DNA synthesis, with all steps after DNA priming performed while the polymerase is covalently coupled to the product DNA. Because the N-terminal domain of the polymerase is displaced from the active site on the same molecule by the viral DNA during reverse transcription, P must be structurally dynamic during DNA synthesis. Therefore, non-nucleoside compounds that interfere with this change may be novel antiviral agents. Hepadnaviral reverse transcription occurs within cytoplasmic capsid particles and is catalyzed by a virally encoded reverse transcriptase, but the primary structure and multimeric state of the polymerase during reverse transcription are poorly understood. We measured these parameters for the duck hepatitis B virus polymerase employing active enzyme translated in vitro and derived from intracellular core particles and mature virions. In vitro-translated polymerase immunoprecipitated as a monomer, and polymerase molecules with complementary defects in the enzymatic active site and tyrosine 96, which primes DNA synthesis, could not complement or inhibit each other in priming assays. Western analysis using antibodies recognizing epitopes throughout the polymerase combined with nuclease digestion of permeabilized virion-derived capsid particles revealed that only full-length polymerase molecules were in virions and that they were all covalently attached to large DNA molecules. Because DNA synthesis is primed by the polymerase itself and only one copy of the viral DNA is in each capsid, the polymerase must function as an uncleaved monomer. Therefore, a single polymerase monomer is encapsidated, primes DNA synthesis, synthesizes both DNA strands, and participates in the three-strand transfers of DNA synthesis, with all steps after DNA priming performed while the polymerase is covalently coupled to the product DNA. Because the N-terminal domain of the polymerase is displaced from the active site on the same molecule by the viral DNA during reverse transcription, P must be structurally dynamic during DNA synthesis. Therefore, non-nucleoside compounds that interfere with this change may be novel antiviral agents. Hepadnaviruses are small DNA-containing viruses that replicate by reverse transcription and infect primates, rodents, and birds (reviewed in Ref. 1Ganem D. Schneider R.J. Knipe D.M. Howley P.M. Fields Virology. Lippincott Williams & Wilkins, Philadelphia2001: 2923-2969Google Scholar). Significant differences exist in the biology of these viruses, but they all share a very similar genetic organization, hepatic tropism, and nearly identical replication mechanisms. The human hepadnavirus, hepatitis B virus (HBV), 2The abbreviations used are: HBV, hepatitis B virus; DHBV, duck hepatitis B virus; pgRNA, pregenomic RNA; P, polymerase or reverse transcriptase; HIV, human immunodeficiency virus; LMH, chicken hepatoma cells; LMH-D2, LMH cells stably transfected with a DHBV dimer; HA, influenza hemagglutinin epitope tag; MAb, monoclonal antibody; nt, nucleotide(s). is a major cause of liver disease and liver cancer worldwide (2Hollinger F.B. Liang T.J. Knipe D.M. Howley P. Fields Virology. Lippincott Williams & Wilkins, Philadelphia2001: 2971-3036Google Scholar). Duck hepatitis B virus (DHBV) is a common model for HBV. Hepadnaviral reverse transcription (Ref. 3Summers J. Mason W.S. Cell. 1982; 29: 403-415Abstract Full Text PDF PubMed Scopus (1138) Google Scholar, and reviewed in Ref. 1Ganem D. Schneider R.J. Knipe D.M. Howley P.M. Fields Virology. Lippincott Williams & Wilkins, Philadelphia2001: 2923-2969Google Scholar) begins with the binding of P to an RNA stem-loop (ϵ) on the pregenomic RNA (pgRNA), and this ribonucleoprotein complex is then encapsidated (4Bartenschlager R. Junker-Niepmann M. Schaller H. J. Virol. 1990; 64: 5324-5332Crossref PubMed Google Scholar, 5Hirsch R.C. Lavine J.E. Chang L.J. Varmus H.E. Ganem D. Nature. 1990; 344: 552-555Crossref PubMed Scopus (258) Google Scholar). Reverse transcription occurs within cytoplasmic capsid particles and is catalyzed by a virally encoded reverse transcriptase (abbreviated as "P" for "polymerase"). DNA synthesis is primed by a tyrosine in the N-terminal domain of P employing ϵ as a template, and thus, the viral DNA is covalently linked to P (6Weber M. Bronsema V. Bartos H. Bosserhoff A. Bartenschlager R. Schaller H. J. Virol. 1994; 68: 2994-2999Crossref PubMed Google Scholar, 7Zoulim F. Seeger C. J. Virol. 1994; 68: 6-13Crossref PubMed Google Scholar, 8Lanford R.E. Notvall L. Lee H. Beams B. J. Virol. 1997; 71: 2996-3004Crossref PubMed Google Scholar). P synthesizes negative polarity DNA by RNA-dependent DNA synthesis and then positive polarity DNA by DNA-dependent synthesis. This process employs three-strand transfer reactions that presumably are promoted by P, possibly with contributions from the viral capsid structure. P is synthesized by de novo translational initiation (9Schlicht H.J. Radziwill G. Schaller H. Cell. 1989; 56: 85-92Abstract Full Text PDF PubMed Scopus (101) Google Scholar, 10Chang L.J. Pryciak P. Ganem D. Varmus H.E. Nature. 1989; 337: 364-368Crossref PubMed Scopus (106) Google Scholar) rather than as a fusion protein such as the orthoretroviral reverse transcriptases (11Jacks T. Varmus H.E. Science. 1985; 230: 1237-1242Crossref PubMed Scopus (305) Google Scholar, 12Wilson W. Braddock M. Adams S.E. Rathjen P.D. Kingsman S.M. Kingsman A.J. Cell. 1988; 55: 1159-1169Abstract Full Text PDF PubMed Scopus (170) Google Scholar). P has four domains (Fig. 1) (13Arnold E. Ding J. Hughes S.H. Hostomsky Z. Curr. Opin. Struct. Biol. 1995; 5: 27-38Crossref PubMed Scopus (63) Google Scholar, 14Chang L.J. Hirsch R.C. Ganem D. Varmus H.E. J. Virol. 1990; 64: 5553-5558Crossref PubMed Google Scholar, 15Radziwill G. Tucker W. Schaller H. J. Virol. 1990; 64: 613-620Crossref PubMed Google Scholar). The terminal protein domain contains a tyrosine residue that primes DNA synthesis and covalently links P to the viral DNA (Tyr-96 in DHBV (6Weber M. Bronsema V. Bartos H. Bosserhoff A. Bartenschlager R. Schaller H. J. Virol. 1994; 68: 2994-2999Crossref PubMed Google Scholar, 7Zoulim F. Seeger C. J. Virol. 1994; 68: 6-13Crossref PubMed Google Scholar)). The spacer domain has no known function and can tolerate significant insertions and deletions. The reverse transcriptase domain contains the DNA polymerase activity. Mutation of an essential YMDD motif in the reverse transcriptase domain leads to nucleoside analog resistance or ablation of DNA synthesis activity (2Hollinger F.B. Liang T.J. Knipe D.M. Howley P. Fields Virology. Lippincott Williams & Wilkins, Philadelphia2001: 2971-3036Google Scholar, 16Shaw T. Bartholomeusz A. Locarnini S. J. Hepatol. 2006; 44: 593-606Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar). The RNaseH domain contains the RNaseH activity, which degrades the viral RNA following its conversion to DNA. Despite the importance of P as a target for three approved nucleoside analog drugs, the size of P in viral particles has been hard to discern. Various activity gel, Western, and immunoprecipitation analyses have revealed P sizes of 155, 109, 98, 85, 80, 78, 70, 68, and 63 kDa associated with DHBV cores (17Oberhaus S.M. Newbold J.E. J. Virol. 1993; 67: 6558-6566Crossref PubMed Google Scholar, 18Oberhaus S.M. Newbold J.E. Virology. 1996; 226: 132-134Crossref PubMed Scopus (1) Google Scholar, 19Bosch V. Bartenschlager R. Radziwill G. Schaller H. Virology. 1988; 166: 475-485Crossref PubMed Scopus (44) Google Scholar) and 100, 90, 70, and 65 kDa associated with HBV cores (20Bartenschlager R. Kuhn C. Schaller H. Nucleic Acids Res. 1992; 20: 195-202Crossref PubMed Scopus (39) Google Scholar, 21Bartenschlager R. Schaller H. EMBO J. 1992; 11: 3413-3420Crossref PubMed Scopus (285) Google Scholar, 22Bavand M. Feitelson M. Laub O. J. Virol. 1989; 63: 1019-1021Crossref PubMed Google Scholar, 23Bavand M.R. Laub O. J. Virol. 1988; 62: 626-628Crossref PubMed Google Scholar, 24Mack D.H. Bloch W. Nath N. Sninsky J.J. J. Virol. 1988; 62: 4786-4790Crossref PubMed Google Scholar, 25Shin H.J. Rho H.M. J. Biol. Chem. 1995; 270: 11047-11050Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). We have detected DHBV P in cores at its predicted 90-kDa mass by Western analysis (26Yao E. Gong Y. Chen N. Tavis J.E. J. Virol. 2000; 74: 8648-8657Crossref PubMed Scopus (33) Google Scholar), but the antibodies recognized epitopes only in the terminal protein domain and would not have detected P fragments lacking these sequences. P is widely assumed to function as a monomer based on indirect evidence. The strongest circumstantial argument comes from the cis-preference for encapsidation of P as a ribonucleoprotein complex with the pgRNA (4Bartenschlager R. Junker-Niepmann M. Schaller H. J. Virol. 1990; 64: 5324-5332Crossref PubMed Google Scholar, 5Hirsch R.C. Lavine J.E. Chang L.J. Varmus H.E. Ganem D. Nature. 1990; 344: 552-555Crossref PubMed Scopus (258) Google Scholar) (the pgRNA is also the mRNA for P). Although the simplest mechanism would be an interaction between monomeric P and the pgRNA, the complex could contain one, two, or more copies of P bound to the pgRNA. Bartenschlager and Schaller (21Bartenschlager R. Schaller H. EMBO J. 1992; 11: 3413-3420Crossref PubMed Scopus (285) Google Scholar) performed the most direct measurement of the multimeric state of P and concluded that P probably acts as a monomer, but as discussed below, their experiment could not measure this parameter because it did not account for the effects of the covalent linkage of DNA on the detection of P. Retroviral reverse transcriptases can be monomers, homodimers, or heterodimers (reviewed in Refs. 27Goff S.P. Knipe D.M. Howley P.M. Fields Virology. Lippincott Williams & Wilkins, Philadelphia2001: 1871-1940Google Scholar and 28Katz R.A. Skalka A.M. Annu. Rev. Biochem. 1994; 63: 133-173Crossref PubMed Scopus (535) Google Scholar), and thus, it is mechanistically possible for P to function as a monomer, dimer, or higher-order multimer. Therefore, it is still unclear whether P functions as a full-length molecule and whether it works as a monomer or a multimer, despite the fundamental implications that these physical properties of P have on the mechanism of reverse transcription. We wish to understand hepadnaviral reverse transcription to promote development of novel antiviral drugs, and thus, we used a combination of immunologic and mutagenesis assays to determine whether active P is proteolytically processed and whether it functions as a monomer or a multimer. Viruses and DNA Clones—LMH cells are a chicken hepatoma line that produces DHBV upon transfection with DHBV genomic expression constructs (29Condreay L.D. Aldrich C.E. Coates L. Mason W.S. Wu T.T. J. Virol. 1990; 64: 3249-3258Crossref PubMed Google Scholar). LMH cells were stably transfected with a DHBV16 dimer to create LMH-D2 cells that constitutively secrete DHBV16 virions (30Moraleda G. Wu T.T. Jilbert A.R. Aldrich C.E. Condreay L.D. Larsen S.H. Tang J.C. Colacino J.M. Mason W.S. Antiviral Res. 1993; 20: 235-247Crossref PubMed Scopus (47) Google Scholar). pT7DPol contains DHBV3 (31Sprengel R. Kuhn C. Will H. Schaller H. J. Med. Virol. 1985; 15: 323-333Crossref PubMed Scopus (109) Google Scholar) nucleotides (nt) 170-3021 encoding P in pBluescript (Promega) with a 33-nt insertion at nt 792 encoding the influenza virus hemagglutinin epitope (HA tag) (32Kolodziej P.A. Young R.A. Methods Enzymol. 1991; 194: 508-519Crossref PubMed Scopus (436) Google Scholar) in the spacer domain. The Brome Mosaic virus internal ribosomal entry site was introduced upstream of the P open reading frame in pT7DPol to produce pT7BDPol to increase the fidelity of initiation at the first AUG of the P open reading frame. pT7BDPol-3′-HA is pT7BDPol with an insertion of the HA tag at the 3′ end of the P gene instead of at nt 792. pT7DPol(Y96F) contains a point mutation in the P gene altering Tyr-96 to Phe to prevent covalent attachment of DNA to P. pT7DPol(YMHA) contains D513H and D514A active site missense mutations that ablate DNA synthesis (14Chang L.J. Hirsch R.C. Ganem D. Varmus H.E. J. Virol. 1990; 64: 5553-5558Crossref PubMed Google Scholar). D1.5G contains 1.5 copies of DHBV3 (duplication of nt 1658–3021) in pBluescript (Stratagene) and produces wild-type DHBV following transfection into cells. D1.5G-3′-HA contains the HA tag in the 3′ end of the P gene. D1.5G-SH has the insertion of the HA tag at nt 792 encoding the spacer domain. Fig. 1 shows the structure of all P derivatives employed in this study. PCR—Template for T7 RNA polymerase-mediated transcription of 3′-HA-tagged P (P(3′-HA); 806 amino acids long) was generated by amplification of pT7BDPol-3′-HA with primers T7 (5′-TAA TAC GAC TCA CTA TAG GG-3′) and HA-R1 (5′-CCA TCG ATT TAA GCG TAG TCT GGG ACG TCG TAT GGG TAA GTT CCA CAT AGC CTA TGT G-3′). Template for P truncated at amino acid 738 (P-(1–738)) was generated by cleaving the P(3′-HA) amplicon with NcoI. In Vitro Transcription and Translation—All mRNAs for P were transcribed from plasmids and PCR amplicons employing Megascript kits (Ambion) according to the manufacturer's instructions. The mRNAs for P with HA (P(3′-HA)) or without HA truncated at amino acid 738 (P-(1–738)) were transcribed from PCR amplicons. 35S-labeled P was translated in vitro employing rabbit reticulocyte lysate (Promega) in a 10-μl volume containing 0.6 μl of [35S]methionine (>1,000 Ci/mmol; Amersham Biosciences) at 30 °C for 1 h according to the manufacturer's instructions. Immunoprecipitation—Monoclonal antibody 11 (mAb 11) specific for an epitope between amino acids 46–77 of P (26Yao E. Gong Y. Chen N. Tavis J.E. J. Virol. 2000; 74: 8648-8657Crossref PubMed Scopus (33) Google Scholar) and 3F10 specific for the HA tag (Roche Applied Science) were bound to protein A/G beads (Calbiochem), the antibody-bead complexes were incubated overnight with in vitro-translated P diluted into 0.2 ml of IPP150 (10 mm Tris (pH 7.5), 150 mm NaCl, 0.1% Nonidet P-40), the immunocomplexes were washed four times with 1 ml of IPP150, and P was released by boiling in 25 μlof1× Laemmli buffer. Following SDS-PAGE, radioactive P was detected by phosphorimaging analysis. DNA Priming—P was translated in vitro in the absence (see Fig. 2) or presence (see Fig. 3) of ϵ. Half of the reaction was used to monitor translation, and half was used for the priming assay. If ϵ was absent during translation, 1 μgof ϵ RNA was added to the priming samples. MgCl2 (to 2.4 mm) and 5 μCi of [α-32P]dGTP (>3,000 Ci/mmol; Amersham Biosciences) were added to the translation reactions, and the samples were incubated at 37 °C for 30 min. Reactions were terminated with Laemmli buffer, and the products were resolved by SDS-PAGE. The 32P priming signal was detected by phosphorimaging analysis (the 35S signal from translation was blocked by a layer of exposed x-ray film).FIGURE 3The enzymatically inactive P(Y96F) and P(YMHA) mutants do not inhibit wild-type (WT) P or complement each other during DNA priming. A, 35S-labeled wild-type P, P(Y96F) and P(YMHA) were translated in vitro in the presence of ϵ and were resolved by SDS-PAGE; mock, no mRNA was added during translation. B, priming assays were performed with 35P-labeled dGTP, and the products were resolved by SDS-PAGE.View Large Image Figure ViewerDownload Hi-res image Download (PPT) Transfection of LMH Cells and Isolation of Core Particles—LMH and LMH-D2 cells were maintained in Dulbecco's modified Eagle's medium/F12 with 10% fetal bovine serum. LMH cells were transfected with FuGENE (Roche Applied Science) according to the manufacturer's instructions. Extracellular virions and naked core particles were isolated 5 days post-transfection by clarifying the supernatant by centrifugation at 3,000 × g for 10 min and then layering it over a 30% sucrose cushion and centrifuging at 192,000 × g overnight. Intracellular core particles were purified by centrifugation through a 30% sucrose pellet as described (33Tavis J.E. Massey B. Gong Y. J. Virol. 1998; 72: 5789-5796Crossref PubMed Google Scholar). Pellets containing intracellular or extracellular core particles were dissolved in 50 μl/100-mm plate of cells of B/EDTA (10 mm HEPES (pH 7.8), 15 mm KCl, 5 mm EDTA) containing 5% sucrose. Southern Blot Analysis—Viral DNAs were isolated by proteinase K digestion followed by phenol-chloroform extraction and ethanol precipitation and were then resolved by electrophoresis on 1.2% agarose gels. Southern blotting was performed as described previously (34Staprans S. Loeb D.D. Ganem D. J. Virol. 1991; 65: 1255-1262Crossref PubMed Google Scholar), with internally 32P-labeled DHBV DNA as a probe. Western Blot Analysis—P was detected with monoclonal antibodies mAb 11 against P or mAb 3F10 against the HA tag as described (26Yao E. Gong Y. Chen N. Tavis J.E. J. Virol. 2000; 74: 8648-8657Crossref PubMed Scopus (33) Google Scholar). P within capsids was released from the covalently attached DNA prior to Western analysis by permeabilizing the capsids (35Radziwill G. Zentgraf H. Schaller H. Bosch V. Virology. 1988; 163: 123-132Crossref PubMed Scopus (50) Google Scholar) followed by treatment with micrococcal nuclease. 1 μl of glycine (300 mm, pH 2.5) was added to 5 μl of core extract and incubated at room temperature for 30 s. The reaction was neutralized by the addition of 1 μl of Tris-HCl (600 mm, pH 8.0) followed by the addition of 1 μl of CaCl2 (100 mm) and 1 μl of 5 units/μl micrococcal nuclease and incubation at 37 °C for 20 min. 1 μl of 150 mm EGTA and 2.5 μl of 5× Laemmli buffer were applied to stop the reaction. For virions, the envelope was removed by the addition of 0.5% Nonidet P-40 prior to permeabilization. Active P Translated in Vitro Does Not Form Stable Multimers—If P functions as a dimer or higher-order multimer during DNA priming, these complexes would be present at least transiently during priming assays with in vitro-translated P. Therefore, we attempted to co-immunoprecipitate complexes of active DHBV P isoforms that differed in size. We took advantage of the ability to modify the C terminus of P without ablating its ability to prime DNA synthesis and created two P derivatives carrying alterations at their C termini to permit their electrophoretic distinction from the 786 amino acid long wild-type protein. P-(1–738) is truncated at amino acid 738, and P(3′-HA) is P with an HA epitope tag added to the C-terminal end that increases its size to 806 amino acids. We first demonstrated that these particular P derivatives were enzymatically active for DNA priming. The proteins were translated in rabbit reticulocyte lysate in the absence of ϵ, ϵ was added, and a priming assay was performed. Fig. 2A demonstrates that P-(1–738) and P(3′-HA) could be electrophoretically resolved (lanes 1–3), that both proteins were active in DNA priming (lanes 4 and 5), and that translating the proteins together in the same lysate did not interfere with the priming activity of either protein (lane 6). Therefore, truncating P to amino acid 738 or inserting the HA tag at the C terminus did not interfere with the priming activity of P. Next, we attempted to co-immunoprecipitate these P derivatives to determine whether they formed stable complexes when co-translated. The P derivatives were translated either separately or together and diluted into IPP150 (a buffer that retains the active form of P and is used to measure P:ϵ binding (36Pollack J.R. Ganem D. J. Virol. 1994; 68: 5579-5587Crossref PubMed Google Scholar)), and equal aliquots of the translation mixture were immunoprecipitated with mAb 3F10 specific for the HA tag, mAb 11 specific for the terminal protein domain of P as a positive control, or anti-β-galactosidase as a negative control. All P derivatives were precipitated by mAb 11 (Fig. 2B, lanes 1–3); however, only P molecules containing the HA tag were precipitated by mAb 3F10 (Fig. 2B, lanes 4–6). The anti-β-galactosidase antibody did not precipitate any of the P proteins (Fig. 2B, lane 7). Therefore, untagged P did not co-precipitate with HA-tagged P. These results indicate that P molecules active in the DNA priming reaction do not form dimers or multimers stable enough to survive immunoprecipitation. P Molecules with Complementary Mutations Cannot Prime DNA Synthesis in trans—To determine whether transient or unstable P dimers or multimers form during DNA priming, we assessed the ability of P molecules with complementary defects in either the essential YMDD motif of the reverse transcriptase active site (P(YMHA)) or the tyrosine 96 (P(Y96F)) that primes DNA synthesis to complement each other and prime DNA synthesis in trans or to act as dominant-negative inhibitors of priming by the wild-type enzyme. Wild-type P, P(YMHA), and P(Y96F) were translated in vitro either individually or in combination in the presence of ϵ (Fig. 3A), and a DNA priming assay was performed (Fig. 3B). Wild-type P was active (Fig. 3B, lane 2), and the Y96F and YMHA mutants were inactive in DNA priming when the proteins were translated individually (Fig. 3B, lanes 3 and 4). Wild-type P retained full activity when co-translated with the inactive Y96F and YMHA mutants (Fig. 3B, lanes 5 and 6), and the two inactive mutants failed to complement each other when they were co-translated (Fig. 3B, lane 7). These results with DHBV strain 3 are the same as those reported previously for DHBV strain 16 (7Zoulim F. Seeger C. J. Virol. 1994; 68: 6-13Crossref PubMed Google Scholar). These data indicate that P mutants with lesions that inactivate the reverse transcriptase active site or ablate the hydroxyl on Tyr-96 that forms the covalent linkage with DNA cannot act as dominant negative inhibitors of the wild-type enzyme, nor can they complement each other and prime DNA synthesis in trans. Therefore, transient P dimers or higher multimers are very unlikely to exist during DNA priming in vitro. All P Molecules in Mature Virions Are Covalently Bound to Large DNAs—The preceding experiments were performed in vitro with synthetically produced P, and thus, the conclusions may be limited to this system. Therefore, we extended our studies to mature virions produced in cell culture and derived from infected duck serum. Our analytical strategy relies on the chimeric nature of mature hepadnaviral DNA in virions, in which the 5′ end of the minus polarity DNA is covalently attached to tyrosine 96 of P as a result of protein priming. P in these chimeric molecules cannot be detected by Western analysis because the mass of the DNA (∼2 MDa) greatly exceeds the mass of P (89 kDa) and traps the P-DNA chimeras in the wells during SDS-PAGE. The few P molecules that enter the gel migrate as a high molecular mass smear and do not transfer well from gels during blotting. Therefore, to detect P by Western analysis after synthesis of an appreciable amount of minus polarity DNA, the covalently attached DNA must be removed prior to electrophoresis. The use of mature cores from virions is important in this experiment because core particles must synthesize significant amounts of DNA to be secreted as virions (3Summers J. Mason W.S. Cell. 1982; 29: 403-415Abstract Full Text PDF PubMed Scopus (1138) Google Scholar, 37Gerelsaikhan T. Tavis J.E. Bruss V. J. Virol. 1996; 70: 4269-4274Crossref PubMed Google Scholar, 38Perlman D. Hu H.M. J. Virol. 2003; 77: 2287-2294Crossref PubMed Scopus (44) Google Scholar, 39Wei Y. Tavis J.E. Ganem D. J. Virol. 1996; 70: 6455-6458Crossref PubMed Google Scholar), and analyzing mature cores excludes P molecules that may have initiated DNA synthesis but have not made enough DNA to prevent detection by Western analysis. We employed two sources of mature virions, LMH-D2 cells and infected duck serum. LMH-D2 cells are LMH cells carrying a stably integrated DHBV dimer that constitutively secrete DHBV virions (30Moraleda G. Wu T.T. Jilbert A.R. Aldrich C.E. Condreay L.D. Larsen S.H. Tang J.C. Colacino J.M. Mason W.S. Antiviral Res. 1993; 20: 235-247Crossref PubMed Scopus (47) Google Scholar). Virions were collected from LMH-D2 culture supernatant or duck serum by centrifugation through a 30% sucrose cushion. The virions were treated with 0.5% Nonidet P-40 to remove the viral envelope, and the subviral capsid particles were permeabilized by brief treatment at pH 2.5 (35Radziwill G. Zentgraf H. Schaller H. Bosch V. Virology. 1988; 163: 123-132Crossref PubMed Scopus (50) Google Scholar). One-half of each sample was treated with micrococcal nuclease to remove covalently attached DNA, and the other half was mock-digested. If P acts as a monomer, all of the enzyme would be covalently attached to DNA because each P molecule would have initiated DNA synthesis via protein priming, and so P would only be detectable in the nuclease-treated sample. If P acts as a dimer in which only one molecule is attached to DNA, the two monomers would dissociate during preparation for denaturing SDS-PAGE on reducing gels, so the amount of P detected in the nuclease-treated sample would be twice that detected in the mock-treated sample. If P acts as a trimer with one monomer attached to DNA, the ratio of P in the treated relative to the untreated samples would be 3:2, and the amount of P in the treated versus untreated samples would approach parity with higher-order multimers. When samples derived from LMH-D2 cells were subjected to Western analysis with mAb 11 specific for the terminal protein domain of P, intracellular non-encapsidated P was found at its anticipated sizes, including the 89-kDa primary translation product, an ∼8-kDa larger form, and N-terminal breakdown products of ∼55–60 kDa (Fig. 4A, lane 1) (26Yao E. Gong Y. Chen N. Tavis J.E. J. Virol. 2000; 74: 8648-8657Crossref PubMed Scopus (33) Google Scholar). In the extracellular virion-derived core samples, P was detected at 89 kDa in the nuclease-treated sample (lane 5), but it was not detected in the mock-treated sample (lane 4). As expected, P was not detected in control supernatants from untransfected LMH cells (lanes 2 and 3). Identical results were obtained with mature DHBV virions derived from infected duck serum (Fig. 4B, lanes 4 and 5). An extra set of bands was observed in both nuclease-treated and mock-treated serum-derived samples at about 60 kDa (Fig. 4B, lanes 4 and 5), but these were from nonspecific cross-reactions with host proteins because they were also observed at lower levels in control samples from uninfected serum (Fig. 4B, lanes 2 and 3). Therefore, essentially all P molecules in mature, extracellular virions detectable by mAb 11 were covalently attached to large DNA molecules. Because all viral DNA in capsid particles is covalently attached to P and there is a single site of linkage between the DNA and P (Tyr-96 in DHBV (6Weber M. Bronsema V. Bartos H. Bosserhoff A. Bartenschlager R. Schaller H. J. Virol. 1994; 68: 2994-2999Crossref PubMed Google Scholar, 7Zoulim F. Seeger C. J. Virol. 1994; 68: 6-13Crossref PubMed Google Scholar)), we conclude that there is a 1:1 ratio between the viral DNA molecules and P detectable by mAb 11. P in Mature Virions Is Full Length—The experiments described above ruled out the possibility that P acts as homodimer with one monomer being bound to the DNA because there was a 1:1 ratio of P:DNA in virions. However, they did not eliminate the possibility that P acts as a heterodimer in which one of the monomers lacks the N-terminal portion of P because mAb 11 recognizes an epitope between amino acids 46–77, and thus, the experiment in Fig. 4 is blind to forms of P lacking the terminal protein domain. Repeated attempts to raise sensitive monoclonal antibodies against the C-terminal 75% of DHBV P were unsuccessful, and thus, we resorted to epitope tagging to permit the detection of the C-terminal domains of P. The HA tag was inserted into the spacer domain after amino acid 264 and at the C terminus in the context of the DHBV pgRNA expression vector D1.5G (D1.5G-SH and D1.5G-3′-HA, respectively). To detect potential P isoforms lacking the terminal protein domain, D1.5G-SH and D1.5G-3′-HA were transfected into LMH cells, and core particles were isolated from the cytoplasm and supernatant of the transfected cultures. Core particles were permeabilized and then treated with micrococcal nuclease or mock-treated, and Western blots of these samples were perf
Referência(s)