Defective Infectious Particles and Rare Packaged G
http://www.100md.com
病菌学杂志 2005年第12期
CR-UK Institute for Cancer Studies, Department of Pathology, University of Birmingham, B15 2TT, Birmingham, United Kingdom
German Cancer Research Center, Department of Virus-Associated Tumours, Im Neuenheimer Feld 242, 69120 Heidelberg, Germany
ABSTRACT
The Epstein-Barr virus (EBV) lytic program includes lytic viral DNA replication and the production of a viral particle into which the replicated viral DNA is packaged. The terminal repeats (TRs) located at the end of the linear viral DNA have been identified as the packaging signals. A TR-negative (TR–) mutant therefore provides an appropriate tool to analyze the relationships between EBV DNA packaging and virus production. Here, we show that supernatants from lytically induced 293 cells carrying TR mutant EBV genomes (293/TR–) contain large amounts of viral particles devoid of viral DNA which are nevertheless able to bind to EBV target cells. This shows that viral DNA packaging is not a prerequisite for virion formation and egress. Rather surprisingly, supernatants from lytically induced 293/TR– cells also contained rare infectious viruses carrying the viral mutant DNA. This observation indicates that the TRs are important but not absolutely essential for virus encapsidation.
INTRODUCTION
Viral lytic DNA replication of herpesviruses is a complex process that results from sequential activation of different protein classes (for reviews, see references 12 and 18). Most available data on this process come from the study of herpes simplex virus type 1, the prototypical member of the alphaherpesvirus family. During lytic DNA replication, the herpes simplex virus type 1 genome is amplified several thousand times from the different origins of replication to form highly branched structures (11). These large concatemers are then resolved into unit-length linear viral genomes that will be packaged into preformed procapsids within the infected cell nucleus. Capsids containing viral DNA will undergo further conformational and structural changes and egress from the infected cell as enveloped virion particles (for reviews, see references 2 and 18).
The essential cis elements that lead to herpesviral encapsidation have been identified (2). These include the terminal repeats (TRs) located at both ends of the linear genome that are involved in the excision of individual viral genomes from the concatemers formed during viral lytic replication as well as in their packaging. Detailed analysis has shown that packaging is actually mediated by the pac1 and pac2 motifs present within the TRs of herpesvirus genomes (3, 15). In addition, packaging of the Epstein-Barr virus (EBV) genome requires the presence of the 9-bp and 11-bp cassettes adjacent to the pac1 and pac2 sequences (22).
Little is known about the mechanism of cleavage and circularization of EBV DNA during infection. The terminase subunits pUL56 and pUL89 of human cytomegalovirus play an important role in concatemer excision and packaging of the viral genome (1, 19). Two EBV homologues of these proteins (BALF3 and BdRF1) have been identified on the basis of DNA and protein sequence homology, but there is no experimental evidence that these proteins possess terminase activity (21). Earlier work done on Epstein-Barr virus in fact suggests that the production of single linear viral DNA might result from homologous recombination rather than from cleavage by a terminase (22). The cellular protein Sp1 has been identified in a protein complex that binds to the TRs and could be involved in circularization of the linear EBV genome after infection (20).
We have previously reported the construction of an EBV mutant strain in which the terminal repeats had been exchanged against the kanamycin antibiotic resistance cassette (5). A 293 cell line carrying this TR-negative (TR–) virus (293/TR–) efficiently encapsidates EBV reporter plasmids and can therefore be used as a packaging cell line. Upon induction of the lytic cycle, we found that the TR-negative virus could not be encapsidated. We now report that induction of the lytic cycle in 293/TR– cells leads to the production of a large amount of empty viral particles. In addition, we have recently shown that recombinant viruses constructed from the B95.8 genome such as our TR-negative mutant contain abnormally low amounts of the viral gp110 protein (encoded by the EBV BALF4 gene) (16). We have therefore reanalyzed the infectious potential of the helper virus under conditions that lead to the incorporation of large amounts of gp110 in the viral envelope and show that rare infectious viruses carrying the TR-negative DNA can be isolated.
MATERIALS AND METHODS
Cell lines. BJAB is an EBV-negative Burkitt's lymphoma cell line (14), Raji is an EBV-positive Burkitt's lymphoma cell line (17), and 293 is a human embryonic epithelial kidney cell line transformed by adenovirus type 5 E1a and E1b proteins (7). We also used a 293 cell clone that stably expresses the EBNA1 gene (293-E1) (Courtesy of B. Sugden, McArdle University, Madison, Wis.). Peripheral blood mononuclear cells were isolated from fresh buffy coat by density gradient centrifugation. CD19-positive B cells were isolated using M-450 CD19 (Pan B) Dynabeads (Dynal) and were detached from Dynabeads using Detachabead (Dynal). All cells were grown in RPMI 1640 medium supplemented with 10% fetal calf serum (Invitrogen).
Recombinant DNA plasmids and cell lines. Plasmid 2089 is a plasmid comprising the EBV wild-type genome, the F-factor origin of replication, the gene for hygromycin resistance in eukaryotic cells, and the gene for green fluorescent protein (GFP) (4). The EBV wild-type TRs were replaced by the kanamycin resistance gene by homologous recombination to generate the TR– mutant as described previously (5). The 293 cells stably carrying the 2089 or TR– EBV genome are designated 293/2089 and 293/TR–, respectively.
Plasmid rescue in E. coli. Circular DNA molecules were extracted from 293/TR-inf cells (see below) using a denaturation-renaturation method as described previously (9). Escherichia coli strain DH10B was transformed with the extracted viral recombinant DNA by electroporation (1,000 V, 25 μF, 100 ), and cells were plated onto agar plates containing chloramphenicol (15 μg/ml) for selection.
Virus induction and infection of target cells. For induction of the EBV lytic phase, 293/2089 and 293/TR– cells were transfected in six-well cluster plates with expression plasmids encoding the BZLF1 and BALF4 gene products (0.5 μg each/well) using lipid micelles (Lipofectamine; Invitrogen). In some experiments, transfection of the BALF4 gene was omitted. Three days after transfection, virus supernatants were harvested, filtered through a 0.8-μm filter, and kept frozen at –80°C. One million target cells were infected using 1 ml of TR– or 2089 supernatant. For selection of stably infected 293-E1 cell clones, cells were transferred to a 150-mm tissue culture dish 24 h after infection, and hygromycin was added to the culture medium (100 μg/ml). Four weeks later, outgrowing single clones were expanded and named 293/TR-inf (for 293-E1 cells infected with TR– mutant virus).
Electron microscopy. Lytically induced 293/2089 or 293/TR– cells were washed three times in phosphate-buffered saline (PBS) and fixed with 2.5% glutaraldehyde in the same buffer for 20 min at 4°C. Samples were postfixed in 2% osmium tetroxide in cacodylate buffer for 1 h at 4°C, stained with uranyl acetate (0.5%) for 16 h at 25°C, washed twice in distilled water, dehydrated in ethanol, and embedded in Epon 812. Thin sections were examined by using a Zeiss electron microscope.
Gardella gel electrophoresis and Southern blot analysis. Preparation of genomic DNA and Southern blot analysis were performed as described previously (4). Gardella gel electrophoresis followed by Southern blotting was used to detect viral DNA bound to target cells or in virus supernatants (6). In this experiment, 0.5 million BJAB cells previously incubated with infectious supernatants (0.5 ml supernatant; 3 h on ice) or ultracentrifuged virus from lytically induced cells (1 ml of supernatant; 22,000 rpm, 3 h at 4°C) were lysed directly in gel slots by adding sodium dodecyl sulfate (5% final concentration). Following electrophoresis, the gel was blotted onto a Hybond N+ membrane and hybridized with a radioactively 32P-labeled plasmid encompassing EBV-specific sequences (as indicated).
EBV binding assay, blocking experiments, and DNase I digestion. B cells (0.5 million) were incubated with 0.5 ml of 2089 or TR– virus supernatant for 3 h on ice. Cells were washed three times in PBS and fixed on glass slides for 20 min using pure acetone for immunostaining. Slides were incubated for 30 min with a purified mouse monoclonal antibody directed against EBV glycoprotein gp350/220 (ATCC 72A1 hybridoma; dilution of 1:2,500 in PBS) or gp110 (dilution of 1:1,000, MAB 8184; Chemicon), washed three times in PBS, and incubated for 30 min with a secondary goat-anti-mouse antibody conjugated with the Cy-3 fluorochrome (dilution of 1:300; Dianova). After several washes in PBS, slides were embedded with 90% glycerol and immunofluorescence evaluated by using an inverted fluorescence microscope (Nikon).
To assess the specificity of EBV binding to its target cells, 500 μl viral supernatant was first incubated for 1 h at 37°C with 100 μg of purified anti-gp350/220 antibody before infection. After incubation for 3 h on ice, cells were washed three times in PBS to remove unbound virus particles, and the cells were immunostained as described above. In some experiments, infectious supernatants were pretreated with DNase I (10 units for 1 ml supernatant; Roche) for 1 h at 37°C.
RESULTS
(i) Supernatants from induced 293/TR– cells contain a large number of defective viral particles. To assess the presence of viral particles in supernatants generated by cotransfection of BZLF1 and BALF4 into 293/TR– cells, a binding assay was performed using primary B cells as a target (Fig. 1A). Supernatants from induced 293 cells containing EBV wild-type viruses (293/2089) provided the appropriate positive control. Following incubation of B cells with both types of supernatants and extensive washings, staining with an antibody specific to gp350/220 showed strong signals at the surface of the B cells (Fig. 1A, left). This result shows that both wild-type and TR– supernatants contain viral structures capable of binding to cells. We repeated this binding assay with supernatants from induced 293/TR– cells that had not been transfected with the BALF4 plasmid and obtained similar results (data not shown). The production of these viral structures is therefore not influenced by the amount of gp110 present in the mature virion. To assess the specificity of this binding, we repeated the experiment after preincubation of the viral supernatants with a neutralizing gp350 antibody. The binding assay performed with antibodies specific to gp350/220 showed a drastic reduction of the fluorescent signals (Fig. 1A, right), confirming that supernatants from both 293/2089 and 293/TR– cells contain viral structures that specifically bind to B cells. The same results were obtained by staining the cells with an antibody specific to gp110 (data not shown). Using a complementary approach, we then performed a Gardella gel analysis of the viral structures bound to target B cells. This assay detects the presence of viral nucleic acids, and in this case, we used the EBV-negative BJAB cell line as a target. Supernatants from both induced 293/2089 and 293/TR– cells were incubated with BJAB cells for 3 h at 4°C. After several washings, BJAB cells and viral structures bound to them were loaded onto an agarose gel. Cells and bound viruses were then lysed within the agarose gel slots and submitted to gel electrophoresis. Following Southern blotting, hybridization with an EBV-specific probe against BLLF1 showed the presence of bound EBV genomes in samples incubated with wild-type supernatants but not in samples incubated with TR– supernatants (Fig. 1B).
In a further experiment, 1 ml of supernatant from lytically induced 293/TR– or 293/2089 cells was then ultracentrifuged, and pellets were similarly submitted to Gardella gel electrophoresis. This experiment showed that supernatant from induced 293/TR– cells contains no detectable EBV genomes (Fig. 2A).
These results indicate that supernatants from induced 293/TR– cells contained defective or empty virions that are viral structures devoid of viral genomes. In order to prove this hypothesis, lytically induced 293/TR– and 293/2089 cells were fixed in glutaraldehyde, and ultrathin sections of these cells were examined using an electron microscope. As suggested by our previous experiments, numerous empty capsids could indeed be detected within the nucleus and cytoplasm of the 293/TR– cell line (Fig. 2B). Therefore, in the absence of EBV packaging signals, empty viral structures are produced in large amounts. These results demonstrate that capsid formation is independent from EBV viral packaging.
(ii) Supernatants from induced 293/TR– cells contain rare infectious units carrying the TR– mutant genome. In a further set of experiments, we wished to evaluate the ability of viral structures present in TR– supernatants to infect target cells. We knew from previous experiments that induction of the lytic cycle in 293/TR– cells was not associated with encapsidation of the TR– genome (5). However, these experiments were not performed with induced 293/TR– cells cotransfected with BZLF1 and a BALF4 expression plasmid. We now know that the viral strain B95.8 and all its derivates, such as the 2089 recombinant wild-type genome from which the TR– mutant is derived, express abnormally low levels of the BALF4 gene product gp110 (16). We therefore repeated these experiments with TR– viruses containing high amounts of gp110. We incubated 293 cells that stably carry an EBNA1 expression plasmid, termed here 293-E1 (courtesy of B. Sugden, McArdle Institute, Madison, Wis.), with 293/TR– supernatants overnight. Quite unexpectedly, incubation of 293-E1 cells with supernatants from 293/TR– cells cotransfected with a BZLF1 and BALF4 expression plasmid led to the infection of rare cells as attested by GFP fluorescence (Fig. 3A). Incubation of 293-E1 cells with 293/2089 wild-type supernatants resulted in 29% of the cells being infected, whereas only 0.001% GFP-positive cells were obtained after incubation with 293/TR– supernatants (Table 1).
We confirmed the role of gp110 during infection by incubating 293-E1 cells with supernatants from induced 293/TR– cells not previously transfected with the gp110 expression plasmid. No infection could be detected in this case (data not shown), which is in line with our previous observations. To assess whether our results could be extended to B cells, the infection experiment was repeated with the B-cell line Raji. We found very similar results (40% GFP-positive cells with wild-type 2089 supernatants and 0.001% GFP-positive cells with the TR– supernatant [Table 1 and Fig. 3B]). Furthermore, it was important to show that these rare GFP-positive cells acquired the viral genome following viral infection and not by transfer of naked TR– DNA. To this aim, we repeated the infection experiments with 293-E1 cells as target cells, but both 293/2089 wild-type and 293/TR– supernatants were preincubated with either DNase I or a neutralizing gp350-specific antibody. Whereas DNase I pretreatment had no influence on infection efficiency with both 293/2089 and 293/TR– supernatants, preincubation with a gp350 antibody markedly reduced the efficiency of infection with both types of supernatants (Fig. 3A and Table 1).
To demonstrate that the GFP-positive cells we detected after infection with the 293/TR– supernatants actually carry TR– mutant genomes, infected 293-E1 cells were submitted to hygromycin selection (the TR– genome carries the hygromycin resistance gene). After 3 weeks, hygromycin-resistant cells (termed 293/TR-inf, for infected 293-E1 cells) grew out and were expanded for further characterization. A Southern blot analysis of these hygromycin-resistant cells was first performed using a probe specific for the EBV terminal repeats (Fig. 4A). This experiment confirmed that the 293/TR-inf cells carried the TR– mutant genome. We used the same method to confirm that the hygromycin-resistant infected clones carry the EBNA1 expression plasmid and therefore are genuine 293-E1 cells (Fig. 4B). This excludes a potential contamination of the infected population by 293/TR– cells. It was important to further show that the genome present in the 293/TR-inf population was intact. To this aim, the viral genome present in these cells was isolated and transformed into E. coli DH10B. The genome of the TR– mutant carries a prokaryotic replicon and can therefore be propagated in prokaryotic hosts. Preparation of plasmid DNA from the transformed bacterial cells showed that the viral genome present in 293 cells after infection with TR– supernatants was intact. Minor differences were observed only in the highly repetitive region of the NotI repeats and within the F plasmid as indicated (Fig. 4C).
(iii) Analysis of the BamHI W repeat numbers suggests that infectious particles are carrying TR-negative viral genomes that underwent lytic replication. At this stage, it was not clear whether the rare infectious viruses present in TR– supernatants underwent lytic replication before encapsidation. Lytically replicated DNAs can differ from their templates in that they carry a different number of repeats; e.g., the number of TRs varies within the pool of newly replicated viruses. Obviously, variation in the TR numbers cannot be used to investigate our TR– mutant virus population, and we decided to analyze the BamHI W repeats instead. Figure 4D shows the result of a Southern blot using BamHI W as a probe. The 293/TR– parental cell line contains two different populations of equal intensity, whereas the stably infected 293-E1 cell line (293/TR-inf) showed a different pattern with a unique signal visible. The episomes present in the stably infected 293-E1 cell line differed in terms of the number of BamHI W repeats from the episomes present in the 293/TR– cell line. It is therefore likely that the viral genomes in the stably infected 293-E1 cell line are derived from replicated EBV copies generated by lytic replication of 293/TR–.
DISCUSSION
In this paper, we have shown that supernatants from lytically induced 293/TR– cells contain large numbers of empty EBV particles that efficiently bind to EBV target cells but also rare infectious particles. This has several consequences for our understanding of the molecular mechanisms that lead to EBV virus production. It shows that the presence of TRs is not strictly required for packaging, as rare infected 293-E1 cells could be observed after incubation with 293/TR– supernatants. However, in the absence of TRs, the efficiency of encapsidation was markedly low (0.001% of infected cells), as most of the produced virions were devoid of viral DNA and therefore noninfectious. This confirms that the terminal repeats are essential for efficient packaging to take place. Direct analysis of the structure of the rare encapsidated TR– genomes was not possible, and we can only speculate on the mechanisms that lead to packaging of the TR– genome. First, it might be possible to excise single genomes from the concatemers formed during lytic replication in the absence of TRs and to package these viral units into infectious viruses. This hypothesis appears unlikely, as nonspecific random breakage of concatemers will very rarely produce intact single EBV genomes that contain the viral elements such as oriP or EBNA1 and the hygromycin resistance cassette that are necessary for persistence and selection of recombinant EBV episomes. Another hypothesis is that episomes can be encapsidated without going through lytic replication. This is again quite unlikely, as analysis of viral DNA rescued from 293/TR-inf cells showed that the infectious particles contained a full-length TR– genome with less BamHI W repeats than the virus present in the 293/TR– cell line. This strongly suggests but does not formally prove that the viral genome present in the 293/TR-inf clones went through one round of lytic replication in the parental 293/TR– cells. Finally, it is possible that in addition to concatemer formation, lytic replication gives rise to single copies of the linear viral genome. These could be directly incorporated into the viral capsid without requiring prior excision from the concatemers. Alternatively, pieces of concatemers carrying more than a single genome might be resolved into one viral unit following homologous recombination. Homologous recombination would give rise to an intact EBV genome.
The other conclusion from these observations is that synthesis and egress of virions are independent of the presence of TRs on the viral genome. Empty or defective particles are commonly observed in human strains of cytomegalovirus, where a high level of lytic replication is paralleled by the presence of empty viral particles (10). In this case, however, empty viral particles are mixed with fully infectious particles, whereas in contrast, induction of lytic replication in 293/TR– cells leads to the largely predominant production of empty viral particles. Selective elimination of the rare particles containing viral genomes should therefore provide a pure population of empty viral particles with B-cell tropism. Empty viral particles have been preferentially used for vaccination against several viruses such as hepatitis B and papillomavirus (8, 13). Further work to investigate the immunogenicity and safety of these empty particles as a potential multipeptidic EBV vaccine is therefore warranted.
The results of this work should be considered for the use of EBV-derived vectors for gene replacement purposes. The 293/TR– cell line can be used as a helper cell line to encapsidate EBV-derived vectors. To avoid potential harmful consequences of infection with a transforming virus, one of the properties expected from such a cell line is that the helper virus does not get encapsidated. In this regard, therefore, and even if the number of infectious TR-negative viruses is very low, security concerns are legitimate. Removing all immortalizing EBV sequences from the helper virus would overcome the potentially harmful effects of infection. The EBV immortalizing genes are not required for lytic replication, and an EBV genome in which the latent genes have been deleted should keep its ability to package EBV reporter genes.
Our results contradict previous observations from our group in which encapsidated TR– genomes could not be detected. These discrepancies can be explained by the use of recombinant EBV containing variable amounts of gp110. Recombinant B95.8 viruses contain abnormally low amounts of gp110, and it is therefore necessary to artificially increase gp110 incorporation to reach levels found in common laboratory strains such as Akata or M-ABA (16). The much higher infectious potential of viruses containing high amounts of gp110 is likely to explain our observation of rare infectious TR– genomes. The production of empty viral particles, in contrast, does not seem to be influenced by the amount of gp110 present in the mature virion.
ACKNOWLEDGMENTS
This work was supported by institutional grants.
We thank B. Hub for expert technical assistance with the electron microscope.
REFERENCES
Bogner, E., K. Radsak, and M. F. Stinski. 1998. The gene product of human cytomegalovirus open reading frame UL56 binds the pac motif and has specific nuclease activity. J. Virol. 72:2259-2264.
Brown, J. C., M. A. McVoy, and F. L. Homa. 2002. Packaging DNA into herpesvirus capsids, p. 111-153. In A. Holzenburg and E. Bogner (ed.), Structure-function relationships of human pathogenic viruses. Kluwer Academic/Plenum, London, United Kingdom.
Deiss, L. P., and N. Frenkel. 1986. Herpes simplex virus amplicon: cleavage of concatemeric DNA is linked to packaging and involves amplification of the terminally reiterated a sequence. J. Virol. 57:933-941.
Delecluse, H. J., T. Hilsendegen, D. Pich, R. Zeidler, and W. Hammerschmidt. 1998. Propagation and recovery of intact, infectious Epstein-Barr virus from prokaryotic to human cells. Proc. Natl. Acad. Sci. USA 95:8245-8250.
Delecluse, H. J., D. Pich, T. Hilsendegen, C. Baum, and W. Hammerschmidt. 1999. A first-generation packaging cell line for Epstein-Barr virus-derived vectors. Proc. Natl. Acad. Sci. USA 96:5188-5193.
Gardella, T., P. Medveczky, T. Sairenji, and C. Mulder. 1984. Detection of circular and linear herpesvirus DNA molecules in mammalian cells by gel electrophoresis. J. Virol. 50:248-254.
Graham, F. L., J. Smiley, W. C. Russell, and R. Nairn. 1977. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J. Gen. Virol. 36:59-74.
Greenstone, H. L., J. D. Nieland, K. E. de Visser, M. L. De Bruijn, R. Kirnbauer, R. B. Roden, D. R. Lowy, W. M. Kast, and J. T. Schiller. 1998. Chimeric papillomavirus virus-like particles elicit antitumor immunity against the E7 oncoprotein in an HPV16 tumor model. Proc. Natl. Acad. Sci. USA 95:1800-1805.
Griffin, B. E., E. Bjorck, G. Bjursell, and T. Lindahl. 1981. Sequence complexity of circular Epstein-Bar virus DNA in transformed cells. J. Virol. 40:11-19.
Irmiere, A., and W. Gibson. 1983. Isolation and characterization of a noninfectious virion-like particle released from cells infected with human strains of cytomegalovirus. Virology 130:118-133.
Jackson, S. A., and N. A. DeLuca. 2003. Relationship of herpes simplex virus genome configuration to productive and persistent infections. Proc. Natl. Acad. Sci. USA 100:7871-7876.
Kieff, E., and A. B. Rickinson. 2001. Epstein-Barr virus and its replication, p. 2511-2573. 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., vol. 2. Lippincott Williams & Wilkins, Philadelphia, Pa.
Mandic, A., and T. Vujkov. 2004. Human papillomavirus vaccine as a new way of preventing cervical cancer: a dream or the future? Ann. Oncol. 15:197-200.
Menezes, J., W. Leibold, G. Klein, and G. Clements. 1975. Establishment and characterization of an Epstein-Barr virus (EBC)-negative lymphoblastoid B cell line (BJA-B) from an exceptional, EBV-genome-negative African Burkitt's lymphoma. Biomedicine 22:276-284.
Nasseri, M., and E. S. Mocarski. 1988. The cleavage recognition signal is contained within sequences surrounding an a-a junction in herpes simplex virus DNA. Virology 167:25-30.
Neuhierl, B., R. Feederle, W. Hammerschmidt, and H. J. Delecluse. 2002. Glycoprotein gp110 of Epstein-Barr virus determines viral tropism and efficiency of infection. Proc. Natl. Acad. Sci. USA 99:15036-15041.
Pulvertaft, R. J. V. 1964. Cytology of Burkitt's lymphoma (African lymphoma). Lancet i:238-240.
Roizman, B., and D. M. Knipe. 2001. Herpes simplex viruses and their replication, p. 2399-2459. 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., vol. 2. Lippincott Williams & Wilkins, Philadelphia, Pa.
Scheffczik, H., C. G. Savva, A. Holzenburg, L. Kolesnikova, and E. Bogner. 2002. The terminase subunits pUL56 and pUL89 of human cytomegalovirus are DNA-metabolizing proteins with toroidal structure. Nucleic Acids Res. 30:1695-1703.
Sun, R., T. A. Spain, S. F. Lin, and G. Miller. 1997. Sp1 binds to the precise locus of end processing within the terminal repeats of Epstein-Barr virus DNA. J. Virol. 71:6136-6143.
van Grunsven, W. M., E. C. van Heerde, H. J. de Haard, W. J. Spaan, and J. M. Middeldorp. 1993. Gene mapping and expression of two immunodominant Epstein-Barr virus capsid proteins. J. Virol. 67:3908-3916.
Zimmermann, J., and W. Hammerschmidt. 1995. Structure and role of the terminal repeats of Epstein-Barr virus in processing and packaging of virion DNA. J. Virol. 69:3147-3155.(R. Feederle, C. Shannon-L)
German Cancer Research Center, Department of Virus-Associated Tumours, Im Neuenheimer Feld 242, 69120 Heidelberg, Germany
ABSTRACT
The Epstein-Barr virus (EBV) lytic program includes lytic viral DNA replication and the production of a viral particle into which the replicated viral DNA is packaged. The terminal repeats (TRs) located at the end of the linear viral DNA have been identified as the packaging signals. A TR-negative (TR–) mutant therefore provides an appropriate tool to analyze the relationships between EBV DNA packaging and virus production. Here, we show that supernatants from lytically induced 293 cells carrying TR mutant EBV genomes (293/TR–) contain large amounts of viral particles devoid of viral DNA which are nevertheless able to bind to EBV target cells. This shows that viral DNA packaging is not a prerequisite for virion formation and egress. Rather surprisingly, supernatants from lytically induced 293/TR– cells also contained rare infectious viruses carrying the viral mutant DNA. This observation indicates that the TRs are important but not absolutely essential for virus encapsidation.
INTRODUCTION
Viral lytic DNA replication of herpesviruses is a complex process that results from sequential activation of different protein classes (for reviews, see references 12 and 18). Most available data on this process come from the study of herpes simplex virus type 1, the prototypical member of the alphaherpesvirus family. During lytic DNA replication, the herpes simplex virus type 1 genome is amplified several thousand times from the different origins of replication to form highly branched structures (11). These large concatemers are then resolved into unit-length linear viral genomes that will be packaged into preformed procapsids within the infected cell nucleus. Capsids containing viral DNA will undergo further conformational and structural changes and egress from the infected cell as enveloped virion particles (for reviews, see references 2 and 18).
The essential cis elements that lead to herpesviral encapsidation have been identified (2). These include the terminal repeats (TRs) located at both ends of the linear genome that are involved in the excision of individual viral genomes from the concatemers formed during viral lytic replication as well as in their packaging. Detailed analysis has shown that packaging is actually mediated by the pac1 and pac2 motifs present within the TRs of herpesvirus genomes (3, 15). In addition, packaging of the Epstein-Barr virus (EBV) genome requires the presence of the 9-bp and 11-bp cassettes adjacent to the pac1 and pac2 sequences (22).
Little is known about the mechanism of cleavage and circularization of EBV DNA during infection. The terminase subunits pUL56 and pUL89 of human cytomegalovirus play an important role in concatemer excision and packaging of the viral genome (1, 19). Two EBV homologues of these proteins (BALF3 and BdRF1) have been identified on the basis of DNA and protein sequence homology, but there is no experimental evidence that these proteins possess terminase activity (21). Earlier work done on Epstein-Barr virus in fact suggests that the production of single linear viral DNA might result from homologous recombination rather than from cleavage by a terminase (22). The cellular protein Sp1 has been identified in a protein complex that binds to the TRs and could be involved in circularization of the linear EBV genome after infection (20).
We have previously reported the construction of an EBV mutant strain in which the terminal repeats had been exchanged against the kanamycin antibiotic resistance cassette (5). A 293 cell line carrying this TR-negative (TR–) virus (293/TR–) efficiently encapsidates EBV reporter plasmids and can therefore be used as a packaging cell line. Upon induction of the lytic cycle, we found that the TR-negative virus could not be encapsidated. We now report that induction of the lytic cycle in 293/TR– cells leads to the production of a large amount of empty viral particles. In addition, we have recently shown that recombinant viruses constructed from the B95.8 genome such as our TR-negative mutant contain abnormally low amounts of the viral gp110 protein (encoded by the EBV BALF4 gene) (16). We have therefore reanalyzed the infectious potential of the helper virus under conditions that lead to the incorporation of large amounts of gp110 in the viral envelope and show that rare infectious viruses carrying the TR-negative DNA can be isolated.
MATERIALS AND METHODS
Cell lines. BJAB is an EBV-negative Burkitt's lymphoma cell line (14), Raji is an EBV-positive Burkitt's lymphoma cell line (17), and 293 is a human embryonic epithelial kidney cell line transformed by adenovirus type 5 E1a and E1b proteins (7). We also used a 293 cell clone that stably expresses the EBNA1 gene (293-E1) (Courtesy of B. Sugden, McArdle University, Madison, Wis.). Peripheral blood mononuclear cells were isolated from fresh buffy coat by density gradient centrifugation. CD19-positive B cells were isolated using M-450 CD19 (Pan B) Dynabeads (Dynal) and were detached from Dynabeads using Detachabead (Dynal). All cells were grown in RPMI 1640 medium supplemented with 10% fetal calf serum (Invitrogen).
Recombinant DNA plasmids and cell lines. Plasmid 2089 is a plasmid comprising the EBV wild-type genome, the F-factor origin of replication, the gene for hygromycin resistance in eukaryotic cells, and the gene for green fluorescent protein (GFP) (4). The EBV wild-type TRs were replaced by the kanamycin resistance gene by homologous recombination to generate the TR– mutant as described previously (5). The 293 cells stably carrying the 2089 or TR– EBV genome are designated 293/2089 and 293/TR–, respectively.
Plasmid rescue in E. coli. Circular DNA molecules were extracted from 293/TR-inf cells (see below) using a denaturation-renaturation method as described previously (9). Escherichia coli strain DH10B was transformed with the extracted viral recombinant DNA by electroporation (1,000 V, 25 μF, 100 ), and cells were plated onto agar plates containing chloramphenicol (15 μg/ml) for selection.
Virus induction and infection of target cells. For induction of the EBV lytic phase, 293/2089 and 293/TR– cells were transfected in six-well cluster plates with expression plasmids encoding the BZLF1 and BALF4 gene products (0.5 μg each/well) using lipid micelles (Lipofectamine; Invitrogen). In some experiments, transfection of the BALF4 gene was omitted. Three days after transfection, virus supernatants were harvested, filtered through a 0.8-μm filter, and kept frozen at –80°C. One million target cells were infected using 1 ml of TR– or 2089 supernatant. For selection of stably infected 293-E1 cell clones, cells were transferred to a 150-mm tissue culture dish 24 h after infection, and hygromycin was added to the culture medium (100 μg/ml). Four weeks later, outgrowing single clones were expanded and named 293/TR-inf (for 293-E1 cells infected with TR– mutant virus).
Electron microscopy. Lytically induced 293/2089 or 293/TR– cells were washed three times in phosphate-buffered saline (PBS) and fixed with 2.5% glutaraldehyde in the same buffer for 20 min at 4°C. Samples were postfixed in 2% osmium tetroxide in cacodylate buffer for 1 h at 4°C, stained with uranyl acetate (0.5%) for 16 h at 25°C, washed twice in distilled water, dehydrated in ethanol, and embedded in Epon 812. Thin sections were examined by using a Zeiss electron microscope.
Gardella gel electrophoresis and Southern blot analysis. Preparation of genomic DNA and Southern blot analysis were performed as described previously (4). Gardella gel electrophoresis followed by Southern blotting was used to detect viral DNA bound to target cells or in virus supernatants (6). In this experiment, 0.5 million BJAB cells previously incubated with infectious supernatants (0.5 ml supernatant; 3 h on ice) or ultracentrifuged virus from lytically induced cells (1 ml of supernatant; 22,000 rpm, 3 h at 4°C) were lysed directly in gel slots by adding sodium dodecyl sulfate (5% final concentration). Following electrophoresis, the gel was blotted onto a Hybond N+ membrane and hybridized with a radioactively 32P-labeled plasmid encompassing EBV-specific sequences (as indicated).
EBV binding assay, blocking experiments, and DNase I digestion. B cells (0.5 million) were incubated with 0.5 ml of 2089 or TR– virus supernatant for 3 h on ice. Cells were washed three times in PBS and fixed on glass slides for 20 min using pure acetone for immunostaining. Slides were incubated for 30 min with a purified mouse monoclonal antibody directed against EBV glycoprotein gp350/220 (ATCC 72A1 hybridoma; dilution of 1:2,500 in PBS) or gp110 (dilution of 1:1,000, MAB 8184; Chemicon), washed three times in PBS, and incubated for 30 min with a secondary goat-anti-mouse antibody conjugated with the Cy-3 fluorochrome (dilution of 1:300; Dianova). After several washes in PBS, slides were embedded with 90% glycerol and immunofluorescence evaluated by using an inverted fluorescence microscope (Nikon).
To assess the specificity of EBV binding to its target cells, 500 μl viral supernatant was first incubated for 1 h at 37°C with 100 μg of purified anti-gp350/220 antibody before infection. After incubation for 3 h on ice, cells were washed three times in PBS to remove unbound virus particles, and the cells were immunostained as described above. In some experiments, infectious supernatants were pretreated with DNase I (10 units for 1 ml supernatant; Roche) for 1 h at 37°C.
RESULTS
(i) Supernatants from induced 293/TR– cells contain a large number of defective viral particles. To assess the presence of viral particles in supernatants generated by cotransfection of BZLF1 and BALF4 into 293/TR– cells, a binding assay was performed using primary B cells as a target (Fig. 1A). Supernatants from induced 293 cells containing EBV wild-type viruses (293/2089) provided the appropriate positive control. Following incubation of B cells with both types of supernatants and extensive washings, staining with an antibody specific to gp350/220 showed strong signals at the surface of the B cells (Fig. 1A, left). This result shows that both wild-type and TR– supernatants contain viral structures capable of binding to cells. We repeated this binding assay with supernatants from induced 293/TR– cells that had not been transfected with the BALF4 plasmid and obtained similar results (data not shown). The production of these viral structures is therefore not influenced by the amount of gp110 present in the mature virion. To assess the specificity of this binding, we repeated the experiment after preincubation of the viral supernatants with a neutralizing gp350 antibody. The binding assay performed with antibodies specific to gp350/220 showed a drastic reduction of the fluorescent signals (Fig. 1A, right), confirming that supernatants from both 293/2089 and 293/TR– cells contain viral structures that specifically bind to B cells. The same results were obtained by staining the cells with an antibody specific to gp110 (data not shown). Using a complementary approach, we then performed a Gardella gel analysis of the viral structures bound to target B cells. This assay detects the presence of viral nucleic acids, and in this case, we used the EBV-negative BJAB cell line as a target. Supernatants from both induced 293/2089 and 293/TR– cells were incubated with BJAB cells for 3 h at 4°C. After several washings, BJAB cells and viral structures bound to them were loaded onto an agarose gel. Cells and bound viruses were then lysed within the agarose gel slots and submitted to gel electrophoresis. Following Southern blotting, hybridization with an EBV-specific probe against BLLF1 showed the presence of bound EBV genomes in samples incubated with wild-type supernatants but not in samples incubated with TR– supernatants (Fig. 1B).
In a further experiment, 1 ml of supernatant from lytically induced 293/TR– or 293/2089 cells was then ultracentrifuged, and pellets were similarly submitted to Gardella gel electrophoresis. This experiment showed that supernatant from induced 293/TR– cells contains no detectable EBV genomes (Fig. 2A).
These results indicate that supernatants from induced 293/TR– cells contained defective or empty virions that are viral structures devoid of viral genomes. In order to prove this hypothesis, lytically induced 293/TR– and 293/2089 cells were fixed in glutaraldehyde, and ultrathin sections of these cells were examined using an electron microscope. As suggested by our previous experiments, numerous empty capsids could indeed be detected within the nucleus and cytoplasm of the 293/TR– cell line (Fig. 2B). Therefore, in the absence of EBV packaging signals, empty viral structures are produced in large amounts. These results demonstrate that capsid formation is independent from EBV viral packaging.
(ii) Supernatants from induced 293/TR– cells contain rare infectious units carrying the TR– mutant genome. In a further set of experiments, we wished to evaluate the ability of viral structures present in TR– supernatants to infect target cells. We knew from previous experiments that induction of the lytic cycle in 293/TR– cells was not associated with encapsidation of the TR– genome (5). However, these experiments were not performed with induced 293/TR– cells cotransfected with BZLF1 and a BALF4 expression plasmid. We now know that the viral strain B95.8 and all its derivates, such as the 2089 recombinant wild-type genome from which the TR– mutant is derived, express abnormally low levels of the BALF4 gene product gp110 (16). We therefore repeated these experiments with TR– viruses containing high amounts of gp110. We incubated 293 cells that stably carry an EBNA1 expression plasmid, termed here 293-E1 (courtesy of B. Sugden, McArdle Institute, Madison, Wis.), with 293/TR– supernatants overnight. Quite unexpectedly, incubation of 293-E1 cells with supernatants from 293/TR– cells cotransfected with a BZLF1 and BALF4 expression plasmid led to the infection of rare cells as attested by GFP fluorescence (Fig. 3A). Incubation of 293-E1 cells with 293/2089 wild-type supernatants resulted in 29% of the cells being infected, whereas only 0.001% GFP-positive cells were obtained after incubation with 293/TR– supernatants (Table 1).
We confirmed the role of gp110 during infection by incubating 293-E1 cells with supernatants from induced 293/TR– cells not previously transfected with the gp110 expression plasmid. No infection could be detected in this case (data not shown), which is in line with our previous observations. To assess whether our results could be extended to B cells, the infection experiment was repeated with the B-cell line Raji. We found very similar results (40% GFP-positive cells with wild-type 2089 supernatants and 0.001% GFP-positive cells with the TR– supernatant [Table 1 and Fig. 3B]). Furthermore, it was important to show that these rare GFP-positive cells acquired the viral genome following viral infection and not by transfer of naked TR– DNA. To this aim, we repeated the infection experiments with 293-E1 cells as target cells, but both 293/2089 wild-type and 293/TR– supernatants were preincubated with either DNase I or a neutralizing gp350-specific antibody. Whereas DNase I pretreatment had no influence on infection efficiency with both 293/2089 and 293/TR– supernatants, preincubation with a gp350 antibody markedly reduced the efficiency of infection with both types of supernatants (Fig. 3A and Table 1).
To demonstrate that the GFP-positive cells we detected after infection with the 293/TR– supernatants actually carry TR– mutant genomes, infected 293-E1 cells were submitted to hygromycin selection (the TR– genome carries the hygromycin resistance gene). After 3 weeks, hygromycin-resistant cells (termed 293/TR-inf, for infected 293-E1 cells) grew out and were expanded for further characterization. A Southern blot analysis of these hygromycin-resistant cells was first performed using a probe specific for the EBV terminal repeats (Fig. 4A). This experiment confirmed that the 293/TR-inf cells carried the TR– mutant genome. We used the same method to confirm that the hygromycin-resistant infected clones carry the EBNA1 expression plasmid and therefore are genuine 293-E1 cells (Fig. 4B). This excludes a potential contamination of the infected population by 293/TR– cells. It was important to further show that the genome present in the 293/TR-inf population was intact. To this aim, the viral genome present in these cells was isolated and transformed into E. coli DH10B. The genome of the TR– mutant carries a prokaryotic replicon and can therefore be propagated in prokaryotic hosts. Preparation of plasmid DNA from the transformed bacterial cells showed that the viral genome present in 293 cells after infection with TR– supernatants was intact. Minor differences were observed only in the highly repetitive region of the NotI repeats and within the F plasmid as indicated (Fig. 4C).
(iii) Analysis of the BamHI W repeat numbers suggests that infectious particles are carrying TR-negative viral genomes that underwent lytic replication. At this stage, it was not clear whether the rare infectious viruses present in TR– supernatants underwent lytic replication before encapsidation. Lytically replicated DNAs can differ from their templates in that they carry a different number of repeats; e.g., the number of TRs varies within the pool of newly replicated viruses. Obviously, variation in the TR numbers cannot be used to investigate our TR– mutant virus population, and we decided to analyze the BamHI W repeats instead. Figure 4D shows the result of a Southern blot using BamHI W as a probe. The 293/TR– parental cell line contains two different populations of equal intensity, whereas the stably infected 293-E1 cell line (293/TR-inf) showed a different pattern with a unique signal visible. The episomes present in the stably infected 293-E1 cell line differed in terms of the number of BamHI W repeats from the episomes present in the 293/TR– cell line. It is therefore likely that the viral genomes in the stably infected 293-E1 cell line are derived from replicated EBV copies generated by lytic replication of 293/TR–.
DISCUSSION
In this paper, we have shown that supernatants from lytically induced 293/TR– cells contain large numbers of empty EBV particles that efficiently bind to EBV target cells but also rare infectious particles. This has several consequences for our understanding of the molecular mechanisms that lead to EBV virus production. It shows that the presence of TRs is not strictly required for packaging, as rare infected 293-E1 cells could be observed after incubation with 293/TR– supernatants. However, in the absence of TRs, the efficiency of encapsidation was markedly low (0.001% of infected cells), as most of the produced virions were devoid of viral DNA and therefore noninfectious. This confirms that the terminal repeats are essential for efficient packaging to take place. Direct analysis of the structure of the rare encapsidated TR– genomes was not possible, and we can only speculate on the mechanisms that lead to packaging of the TR– genome. First, it might be possible to excise single genomes from the concatemers formed during lytic replication in the absence of TRs and to package these viral units into infectious viruses. This hypothesis appears unlikely, as nonspecific random breakage of concatemers will very rarely produce intact single EBV genomes that contain the viral elements such as oriP or EBNA1 and the hygromycin resistance cassette that are necessary for persistence and selection of recombinant EBV episomes. Another hypothesis is that episomes can be encapsidated without going through lytic replication. This is again quite unlikely, as analysis of viral DNA rescued from 293/TR-inf cells showed that the infectious particles contained a full-length TR– genome with less BamHI W repeats than the virus present in the 293/TR– cell line. This strongly suggests but does not formally prove that the viral genome present in the 293/TR-inf clones went through one round of lytic replication in the parental 293/TR– cells. Finally, it is possible that in addition to concatemer formation, lytic replication gives rise to single copies of the linear viral genome. These could be directly incorporated into the viral capsid without requiring prior excision from the concatemers. Alternatively, pieces of concatemers carrying more than a single genome might be resolved into one viral unit following homologous recombination. Homologous recombination would give rise to an intact EBV genome.
The other conclusion from these observations is that synthesis and egress of virions are independent of the presence of TRs on the viral genome. Empty or defective particles are commonly observed in human strains of cytomegalovirus, where a high level of lytic replication is paralleled by the presence of empty viral particles (10). In this case, however, empty viral particles are mixed with fully infectious particles, whereas in contrast, induction of lytic replication in 293/TR– cells leads to the largely predominant production of empty viral particles. Selective elimination of the rare particles containing viral genomes should therefore provide a pure population of empty viral particles with B-cell tropism. Empty viral particles have been preferentially used for vaccination against several viruses such as hepatitis B and papillomavirus (8, 13). Further work to investigate the immunogenicity and safety of these empty particles as a potential multipeptidic EBV vaccine is therefore warranted.
The results of this work should be considered for the use of EBV-derived vectors for gene replacement purposes. The 293/TR– cell line can be used as a helper cell line to encapsidate EBV-derived vectors. To avoid potential harmful consequences of infection with a transforming virus, one of the properties expected from such a cell line is that the helper virus does not get encapsidated. In this regard, therefore, and even if the number of infectious TR-negative viruses is very low, security concerns are legitimate. Removing all immortalizing EBV sequences from the helper virus would overcome the potentially harmful effects of infection. The EBV immortalizing genes are not required for lytic replication, and an EBV genome in which the latent genes have been deleted should keep its ability to package EBV reporter genes.
Our results contradict previous observations from our group in which encapsidated TR– genomes could not be detected. These discrepancies can be explained by the use of recombinant EBV containing variable amounts of gp110. Recombinant B95.8 viruses contain abnormally low amounts of gp110, and it is therefore necessary to artificially increase gp110 incorporation to reach levels found in common laboratory strains such as Akata or M-ABA (16). The much higher infectious potential of viruses containing high amounts of gp110 is likely to explain our observation of rare infectious TR– genomes. The production of empty viral particles, in contrast, does not seem to be influenced by the amount of gp110 present in the mature virion.
ACKNOWLEDGMENTS
This work was supported by institutional grants.
We thank B. Hub for expert technical assistance with the electron microscope.
REFERENCES
Bogner, E., K. Radsak, and M. F. Stinski. 1998. The gene product of human cytomegalovirus open reading frame UL56 binds the pac motif and has specific nuclease activity. J. Virol. 72:2259-2264.
Brown, J. C., M. A. McVoy, and F. L. Homa. 2002. Packaging DNA into herpesvirus capsids, p. 111-153. In A. Holzenburg and E. Bogner (ed.), Structure-function relationships of human pathogenic viruses. Kluwer Academic/Plenum, London, United Kingdom.
Deiss, L. P., and N. Frenkel. 1986. Herpes simplex virus amplicon: cleavage of concatemeric DNA is linked to packaging and involves amplification of the terminally reiterated a sequence. J. Virol. 57:933-941.
Delecluse, H. J., T. Hilsendegen, D. Pich, R. Zeidler, and W. Hammerschmidt. 1998. Propagation and recovery of intact, infectious Epstein-Barr virus from prokaryotic to human cells. Proc. Natl. Acad. Sci. USA 95:8245-8250.
Delecluse, H. J., D. Pich, T. Hilsendegen, C. Baum, and W. Hammerschmidt. 1999. A first-generation packaging cell line for Epstein-Barr virus-derived vectors. Proc. Natl. Acad. Sci. USA 96:5188-5193.
Gardella, T., P. Medveczky, T. Sairenji, and C. Mulder. 1984. Detection of circular and linear herpesvirus DNA molecules in mammalian cells by gel electrophoresis. J. Virol. 50:248-254.
Graham, F. L., J. Smiley, W. C. Russell, and R. Nairn. 1977. Characteristics of a human cell line transformed by DNA from human adenovirus type 5. J. Gen. Virol. 36:59-74.
Greenstone, H. L., J. D. Nieland, K. E. de Visser, M. L. De Bruijn, R. Kirnbauer, R. B. Roden, D. R. Lowy, W. M. Kast, and J. T. Schiller. 1998. Chimeric papillomavirus virus-like particles elicit antitumor immunity against the E7 oncoprotein in an HPV16 tumor model. Proc. Natl. Acad. Sci. USA 95:1800-1805.
Griffin, B. E., E. Bjorck, G. Bjursell, and T. Lindahl. 1981. Sequence complexity of circular Epstein-Bar virus DNA in transformed cells. J. Virol. 40:11-19.
Irmiere, A., and W. Gibson. 1983. Isolation and characterization of a noninfectious virion-like particle released from cells infected with human strains of cytomegalovirus. Virology 130:118-133.
Jackson, S. A., and N. A. DeLuca. 2003. Relationship of herpes simplex virus genome configuration to productive and persistent infections. Proc. Natl. Acad. Sci. USA 100:7871-7876.
Kieff, E., and A. B. Rickinson. 2001. Epstein-Barr virus and its replication, p. 2511-2573. 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., vol. 2. Lippincott Williams & Wilkins, Philadelphia, Pa.
Mandic, A., and T. Vujkov. 2004. Human papillomavirus vaccine as a new way of preventing cervical cancer: a dream or the future? Ann. Oncol. 15:197-200.
Menezes, J., W. Leibold, G. Klein, and G. Clements. 1975. Establishment and characterization of an Epstein-Barr virus (EBC)-negative lymphoblastoid B cell line (BJA-B) from an exceptional, EBV-genome-negative African Burkitt's lymphoma. Biomedicine 22:276-284.
Nasseri, M., and E. S. Mocarski. 1988. The cleavage recognition signal is contained within sequences surrounding an a-a junction in herpes simplex virus DNA. Virology 167:25-30.
Neuhierl, B., R. Feederle, W. Hammerschmidt, and H. J. Delecluse. 2002. Glycoprotein gp110 of Epstein-Barr virus determines viral tropism and efficiency of infection. Proc. Natl. Acad. Sci. USA 99:15036-15041.
Pulvertaft, R. J. V. 1964. Cytology of Burkitt's lymphoma (African lymphoma). Lancet i:238-240.
Roizman, B., and D. M. Knipe. 2001. Herpes simplex viruses and their replication, p. 2399-2459. 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., vol. 2. Lippincott Williams & Wilkins, Philadelphia, Pa.
Scheffczik, H., C. G. Savva, A. Holzenburg, L. Kolesnikova, and E. Bogner. 2002. The terminase subunits pUL56 and pUL89 of human cytomegalovirus are DNA-metabolizing proteins with toroidal structure. Nucleic Acids Res. 30:1695-1703.
Sun, R., T. A. Spain, S. F. Lin, and G. Miller. 1997. Sp1 binds to the precise locus of end processing within the terminal repeats of Epstein-Barr virus DNA. J. Virol. 71:6136-6143.
van Grunsven, W. M., E. C. van Heerde, H. J. de Haard, W. J. Spaan, and J. M. Middeldorp. 1993. Gene mapping and expression of two immunodominant Epstein-Barr virus capsid proteins. J. Virol. 67:3908-3916.
Zimmermann, J., and W. Hammerschmidt. 1995. Structure and role of the terminal repeats of Epstein-Barr virus in processing and packaging of virion DNA. J. Virol. 69:3147-3155.(R. Feederle, C. Shannon-L)