Previous Article | Next Article ![]()
Journal of Virology, May 2006, p. 5078-5081, Vol. 80, No. 10
0022-538X/06/$08.00+0 doi:10.1128/JVI.80.10.5078-5081.2006
Copyright © 2006, American Society for Microbiology. All Rights Reserved.
German Cancer Research Center, Department of Virus Associated Tumours, Im Neuenheimer Feld 242, 69120 Heidelberg, Germany
Received 21 October 2005/ Accepted 20 February 2006
|
|
|---|
BMRF1 EBV phenotype.
BMRF1 viruses were replication deficient, but the wild-type phenotype could be restored by BMRF1 trans-complementation. The replication-deficient phenotype included impaired lytic DNA replication and late protein expression.
BMRF1 and wild-type viruses were undistinguishable in terms of their ability to transform primary B cells. Our results provide genetic evidence that BMRF1 is essential for lytic replication of the EBV genome. |
|
|---|
DNA polymerase processivity factors form a heterogeneous group of proteins, most of which function as "sliding clamps" (19, 20). In contrast to these sliding clamps, the herpes simplex virus (HSV) processivity factor, UL42, directly binds DNA with high affinity (23). A UL42 HSV knockout genome was found to produce no infectious progeny, viral DNA, or late gene products (12), corroborating earlier results about DNA synthesis from an HSV origin (24). Even if related, lytic replication in EBV and that in HSV also show some profound differences. These include preferential establishment of latency in B lymphocytes, EBV's natural cellular host, and a comparatively low permissivity for EBV lytic replication in infected cells. At the molecular level, ClustalW alignment shows a low degree of homology (17% identity) between BMRF1 and UL42. Altogether, it is not possible to draw conclusions about BMRF1 functions from what is known about UL42. Here we report the phenotype of an EBV mutant carrying a complete deletion of the BMRF1 open reading frame.
Construction of the
BMRF1 EBV genome.
We exchanged the entire BMRF1 open reading frame for the kanamycin resistance gene using the strategy depicted in Fig. 1. This technique is based on homologous recombination between a targeting vector and the entire B95.8 EBV genome cloned as a plasmid in the Escherichia coli strain DH10B (3, 16). The cloned EBV carries a hygromycin resistance gene as well as a gene encoding a green fluorescent protein (Clontech). The construct resulting from this recombination,
BMRF1, was stably transfected into 293 cells (9) that were previously shown to be permissive for EBV replication (4). Outgrowing hygromycin-resistant 293 clones carrying the mutant EBV (293/
BMRF1) were tested for their suitability to evaluate the
BMRF1 phenotype. We first assessed the structure of the mutant viral DNA present in E. coli cells and in 293 cells by restriction analysis using several restriction enzymes (Fig. 1a and b). This analysis confirmed that these cells carried a viral genome that was structurally intact but devoid of the
BMRF1 gene. Because BaRF1 and BMRF1 are translated from the same RNA, it was important to exclude the possibility that insertion of the BMRF1 targeting vector could have altered BaRF1 transcription (15). We therefore performed a Northern blot analysis on induced 293/
BMRF1 and 293/wt-EBV with a BaRF1-specific probe. This analysis showed no difference between the two cell lines (Fig. 1c). In a further step, we induced lytic replication by transfection of the transactivator BZLF1 into 293/
BMRF1 cells (Fig. 2). Staining of the induced 293/
BMRF1 cells with two antibodies specific for BMRF1 (Chemicon) (17) was negative, confirming the mutant status of this EBV genome (Fig. 2). We then complemented the
BMRF1 mutant by cotransfection of the BZLF1 and BMRF1 (26) genes into different 293/
BMRF1 clones. Incubation of supernatants from these clones with Raji cells (a human Burkitt's lymphoma cell line) (18) resulted in infection as demonstrated by the observation of green fluorescent protein-positive Raji cells (data not shown). This assay showed that the complemented 293/
BMRF1 is fully able to replicate and is therefore suitable for the study of the
BMRF1 phenotype. It also allowed selection of the best 293/
BMRF1 virus producer cell line, with which subsequent experiments were performed.
![]() View larger version (39K): [in a new window] |
FIG. 1. Schematic representation of the strategy used to construct a BMRF1 EBV. (a) To disrupt the BMRF1 gene, the kanamycin resistance gene was amplified by PCR from pCP15 using primers that introduce 40-bp stretches of homology to sequences immediately upstream and downstream of the BMRF1 gene (EBV coordinates 79859 to 79899 and 81111 to 81150, respectively). Recombination of this linear DNA fragment with the wild-type EBV genome in E. coli leads to an exchange between the kanamycin cassette and the BMRF1 gene region, leading to an increase in size of the BamHI-M fragment from 4,696 bp to 5,027 bp. (b) Restriction analysis of the BMRF1 mutant. Following recombination, the integrity of the recombinant BMRF1 genome was analyzed at different steps of the mutant's construction. The mutant DNA was subjected to restriction analysis both before its transfer to the EBV-permissive 293 cell line and after its rescue from one of the obtained 293/ BMRF1 clones into E. coli cells. The restriction pattern obtained was compared to the one obtained after digestion of the wt EBV genome. Except for the wt 4,696-bp fragment that is exchanged for a 5,027-bp fragment in BMRF1 (see white arrowheads), the overall structure of the mutant appears unchanged. This confirms the integrity and the mutant character of the 293/ BMRF1 cells. (c) Northern blot analysis of the transcription of the upstream BaRF1 gene of BMRF1. RNA from lytically induced cells was electrophoresed on a formaldehyde-1.2% agarose gel, blotted onto a Hybond N+ membrane, and hybridized with a radioactively labeled probe specific for the EBV BaRF1 gene. The intensities of the bands from two wild-type (B95.8) and noncomplemented or complemented BMRF1 clones are very similar, proving that there is no effect of the BMRF1 deletion on the expression of the BaRF1 gene. The transcript in the BMRF1 mutant is 330 bp longer since it spans the site of insertion of the kanamycin resistance gene.
|
![]() View larger version (49K): [in a new window] |
FIG. 2. BMRF1 cells produce neither the BMRF1 protein nor lytic cycle late proteins. 293/ BMRF1 and 293/wt-EBV cells were transfected with an expression plasmid for the BZLF1 gene to induce the lytic cycle and stained with antibodies directed against the BMRF1 protein (the EA-D early antigen). Lytically induced 293/ BMRF1 cells do not produce the antigen, whereas mutant cells complemented with BMRF1 and wild-type cells do, confirming that the BMRF1 genome lacks the BMRF1 gene. When the same cells were stained with antibodies directed against the gp350 or gp110 late antigen, 293/ BMRF1 cells were found to produce neither gp110 nor gp350. In contrast both complemented 293/ BMRF1 cells and wild-type cells stained positive for these antigens.
|
BMRF1 cells with antibodies directed against the gp110 and gp350 EBV late proteins that proved to be negative (Fig. 2). In contrast, both proteins could be readily detected both in induced cells carrying wild-type (wt) EBV (293/wt-EBV) and in BMRF1-complemented 293/
BMRF1 cells. The ability of
BMRF1 to replicate was assessed by Southern blotting with an EBV-specific probe (BamHI-W repeats) (Fig. 3a). Induced 293/
BMRF1 cells were lysed in the slots of an agarose gel and electrophoresed to separate the linear and circular forms of the EBV genome (5, 8). BMRF1-complemented 293/
BMRF1 cells and 293/wt-EBV cells were included as controls. Linear EBV genomes that result from lytic replication could be detected in BMRF1-complemented 293/
BMRF1 cells and 293/wt-EBV cells but not in induced 293/
BMRF1 cells. These results were confirmed by direct observation of induced cells with electron microscopy. No mature or immature viral structures were visible in more than 100 induced 293/
BMRF1 cells (Fig. 3b). In contrast, mature and immature virions were readily detected in the nucleus of BMRF1-complemented 293/
BMRF1 cells (Fig. 3c). All together, these results provide solid evidence that BMRF1 is absolutely required for both DNA lytic replication and lytic protein synthesis.
![]() View larger version (69K): [in a new window] |
FIG. 3. BMRF1 cells do not lytically replicate. (a) BMRF1 viral lytic DNA replication cannot be detected in Gardella gel electrophoresis. DNA from lytically induced cells was electrophoresed on an 0.8% agarose gel, blotted onto a Hybond N+ membrane, and hybridized with a radioactively labeled probe specific for the EBV BamHI-W repeats of EBV. The mutant EBV genome DNA is not replicated. In contrast, strong signals from linear EBV genomes were detected in the 293/wt-EBV and in the 293/ BMRF1 cells complemented with a BMRF1 expression plasmid. (b and c) Lytically induced 293/ BMRF1 cells do not produce EBV virions. 293/ BMRF1 cells were lytically induced and fixed in glutaraldehyde. Ultrathin sections were analyzed by electron microscopy (b). The same experiment was performed in parallel with 293/ BMRF1 cells cotransfected with an expression plasmid carrying the BMRF1 gene (c). In the 293/ BMRF1 cells no viral nucleocapsids could be observed. In contrast, these were readily visible in 293/ BMRF1 cells complemented with BMRF1. Arrowheads point to empty capsids (open arrowheads) or capsids filled with DNA (filled arrowheads). nuc, nucleus; cyt, cytoplasm.
|
BMRF1 or wt EBV virus were incubated with peripheral blood B cells purified with CD19 Dynabeads (Dynal). Infections were performed at multiplicities of infection (MOIs) ranging from 0 to 100 particles per B cell. One thousand infected B cells were plated per well on 96-well cluster plates containing irradiated human fibroblasts. The ability of the viruses to immortalize B cells was evaluated by assessing the proportion of EBNA2-positive cells 3 days after infection and by counting cell outgrowth after 4 weeks. EBNA2 is essential for B-cell transformation, and its expression is a good marker of this process. No significant difference was observed between the two tested viral strains (5% versus 6% EBNA2-positive cells at an MOI of 1, 14% versus 15% at an MOI of 10, and 81% versus 65% at an MOI of 100 after infection with wt EBV and complemented
BMRF1, respectively). Cell outgrowth was also very similar (6% versus 8% wells containing immortalized colonies at an MOI of 1, 54% versus 58% at an MOI of 10, and 100% versus 100% at an MOI of 100 after infection with wt EBV and complemented
BMRF1, respectively). No outgrowth was observed in uninfected B cells or in cells infected with noncomplemented
BMRF1 virus. To confirm that the EBV-immortalized B-cell lines carried the viral strains with which they were infected, we analyzed the structure of the BMRF1 locus by Southern blotting (Fig. 4). As expected, B cells immortalized with
BMRF1 carried this viral mutant only. Our results show that BMRF1 is not required for maintenance of B-cell immortalization. Since BMRF1 cannot be detected in the mature virion (11), it is unlikely to play a role in the early stages of immortalization either.
![]() View larger version (66K): [in a new window] |
FIG. 4. Lymphoblastoid cell line (LCL) generated by infection of primary B cells with BMRF1 virus contains the mutant genome only. Genomic DNA extracted from LCL generated by infection of B cells with either wt EBV or complemented BMRF1 virus ( BMRF1-LCL) was cleaved with the BamHI restriction enzyme and subjected to agarose gel electrophoresis, and Southern blotting analysis was performed with a probe for the BMRF2 gene. Blots were hybridized with a BMRF2-specific probe. The BMRF2 gene is located on the BamHI-M fragment of the EBV genome and allows distinction between the wild-type BamHI-M fragment, which is 4.7 kb in size, and the BamHI-M fragment of the BMRF1 mutant, which is 5.0 kb (Fig. 1). This analysis confirms that B cells immortalized with the BMRF1 virus carry the mutant gene while B cells infected with wt EBV carry an intact BamHI-M fragment.
|
BMRF1 EBV would represent a suitable basis for the generation of "one-shot" EBV-based vectors that immortalize B cells but cannot spread. In addition, the EBV
BMRF1 provides an essential tool to study the phenotype of BMRF1 mutants.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»