Epstein-Barr Virus Transforming Protein LMP1 Plays
http://www.100md.com
病菌学杂志 2005年第6期
Department of Tumor Virology
Center for Virus Vector Development, Institute for Genetic Medicine, Hokkaido University
Laboratory of Molecular and Cellular Pathology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
ABSTRACT
The Epstein-Barr virus (EBV) latent membrane protein 1 (LMP1), which is critical for EBV-induced B-cell transformation, is also abundantly expressed during the lytic cycle of viral replication. However, the biological significance of this strong LMP1 induction remains unknown. We engineered a bacterial artificial chromosome clone containing the entire genome of Akata strain EBV to specifically disrupt the LMP1 gene. Akata cell clones harboring the episomes of LMP1-deleted EBV were established, and the effect of LMP1 loss on virus production was investigated. We found that the degree of viral DNA amplification and the expression levels of viral late gene products were unaffected by LMP1 loss, demonstrating that the LMP1-deleted EBV entered the lytic replication cycle as efficiently as the wild-type counterpart. This was confirmed by our electron microscopic observation that nucleocapsid formation inside nuclei occurred even in the absence of LMP1. By contrast, loss of LMP1 severely impaired virus release into culture supernatants, resulting in poor infection efficiency. The expression of truncated LMP1 in Akata cells harboring LMP1-deleted EBV rescued the virus release into the culture supernatant and the infectivity, and full-length LMP1 partially rescued the infectivity. These results indicate that inducible expression of LMP1 during the viral lytic cycle plays a critical role in virus production.
INTRODUCTION
Epstein-Barr virus (EBV) is a human herpesvirus that establishes a life-long persistence in the host (33). The virus infects the vast majority of the world's adult population and is well known for its association with a broad spectrum of benign and malignant diseases, including infectious mononucleosis, Burkitt's lymphoma, nasopharyngeal carcinoma, B-cell lymphoma in immunocompromised patients, and gastric carcinoma (33). In vitro EBV infection of peripheral B lymphocytes results in their immortalization and continuous proliferation (39). In immortalized lymphoblastoid cell lines, EBV expresses six nuclear proteins (EBNAs), three integral membrane proteins (LMP1, -2A, and -2B), two small nuclear RNAs (EBER1 and -2), and a transcript from the BamHI-A region (21).
Latent membrane protein 1 (LMP1) is one of the viral gene products that are essential for B-cell transformation (19). LMP1 is an integral membrane protein composed of a short cytoplasmic amino-terminal domain, six hydrophobic transmembrane domains, and a cytoplasmic carboxy-terminal domain (10). It has been demonstrated that LMP1 acts as a constitutively active receptor that mimics activated CD40, a member of the tumor necrosis factor receptor family (12, 26). The cytoplasmic carboxy terminus of LMP1 plays a critical role in EBV-induced B-cell transformation through its binding to a tumor necrosis factor receptor-associated factor (TRAF) and a tumor necrosis factor receptor-associated death domain (TRADD) protein (6, 21).
The LMP1 (BNLF1) gene contains three exons that are located within the BamHI-N region of the EBV genome (10). Two open reading frames (ORFs) have been identified based on nucleotide sequences (2, 10) and mRNA mapping (16) of B95-8 strain EBV. A transcript starting from the ED-L1 promoter, which is located upstream of the first exon (16), encodes the first ORF. This ORF encodes full-length LMP1 (386 amino acids) that is abundantly expressed in lymphoblastoid cell lines. Another transcript starting from the EDL1A promoter, which is located within the first intron of the LMP1 gene, encodes the second ORF. The translation initiation site of this second ORF is methionine-129 of full-length LMP1 (16), and the ORF thus encodes an amino-terminally truncated form of the LMP1 protein. The truncated LMP1 (258 amino acids) consists of the fifth and sixth transmembrane domains and the cytoplasmic carboxy terminus of full-length LMP1.
The B95-8 cell line, an in vitro-immortalized lymphoblastoid cell line, was initially used to characterize truncated LMP1. A number of studies demonstrated that expression of truncated LMP1 was specifically upregulated during the lytic phase of viral replication (4, 5, 16, 28, 34, 44). Therefore, truncated form of LMP1 is frequently referred to as lytic LMP1 (8, 43). In contrast to full-length LMP1, truncated LMP1 does not transform rodent cells (3, 44) or alter the phenotypes of human B lymphocytes (45). The only biological activity of truncated LMP1 identified to date is its ability to negatively regulate LMP1 signaling pathways (9), although truncated LMP1 retains the carboxy terminus of LMP1.
Recent studies revealed that Met-129 of the B95-8 LMP1 ORF is not common in many virus isolates (7, 42). For example, the LMP1 gene of Akata strain EBV encodes isoleucine at codon 129, resulting in loss of the translation initiation codon for the ORF of truncated LMP1. The Akata cell line is derived from an EBV-positive Burkitt's lymphoma from a Japanese patient (41). LMP1 expression is not detected in unstimulated Akata cells (42), as Akata cells retain the Burkitt's-type viral gene expression (36). A unique feature of Akata cells is that cross-linking of cell surface immunoglobulin with anti-immunoglobulin antibodies can efficiently induce the lytic cycle of viral replication (40). Interestingly, expression of full-length LMP1 is strongly upregulated after anti-immunoglobulin treatment of Akata cells (42). The induction of full-length LMP1 from Met-129-negative EBV was also observed in Raji and BL74 cells after tetradecanoyl phorbol acetate and butyrate treatment (5, 7). Therefore, the induction of full-length LMP1 after stimulating virus production is not peculiar to Akata cells.
These observations prompted us to investigate the biological significance of LMP1 upregulation during the lytic cycle of viral replication in Akata cells. We therefore set out to generate a recombinant Akata EBV containing the disrupted LMP1 gene. We recently reported the establishment of a bacterial artificial chromosome (BAC) system for engineering the genome of Akata strain EBV and used the system to produce high-titer recombinant viruses (18). We employed the system to disrupt the LMP1 gene of Akata strain EBV and investigated whether Akata cells harboring the episomes of LMP1-deleted EBV (d.LMP1-EBV) could produce infectious viruses. We found that the d.LMP1-EBV entered the lytic replication cycle as efficiently as the wild-type counterpart. However, we unexpectedly found that loss of LMP1 severely impaired virus release into the culture supernatant, resulting in poor infection efficiency. We expressed either truncated LMP1 or full-length LMP1 in Akata cells harboring d.LMP1-EBV to rescue the infectivity of the culture supernatants. The results revealed that loss of LMP1 is truly responsible for the impaired virus release.
MATERIALS AND METHODS
Cell culture. Akata, a human Burkitt's lymphoma-derived cell line carrying EBV as episomes (41) and EBV-negative Akata cells (36) were maintained in RPMI 1640 medium (Sigma-Aldrich Fine Chemicals, St. Louis, Mo.) containing 10% fetal bovine serum at 37°C in a 5% humidified CO2 atmosphere.
Plasmids. A plasmid, pBS246 (Life Technologies), having two loxP sites and multiple cloning sites, was used for excising a loxP cassette. Two loxP sites of pBS246, one located between BspEI and SpeI and the other between EcoRI and Acc65I, were replaced with mutated loxP sites (5171 loxP) with 43-bp synthetic oligonucleotides to construct pBS246-mloxp5171-EXA. The 5171 loxP site has the substitution of two nucleotide pairs within the asymmetric 8-nucleotide spacer region of the loxP sequence, and therefore it barely recombines with the wild-type loxP site (23).
A synthetic oligonucleotide containing triple stop codons and restriction enzyme sites (BamHI, NotI, and SalI) was inserted into the SfiI site (after codon 9 of the LMP1 gene) of pUC119-Nhet (pUC119 containing the BamHI-Nhet fragment of the Akata EBV genome) to construct pUC119-Nhet-del.LMP1.
The blunted FokI-BclI fragment of pCDNA4/HisMax (Invitrogen, Carlsbad, Calif.) containing the zeocin resistance marker gene (zeo) was cloned into the EcoRV site of pBS246-mloxp5171-EXA to make pBS246-mloxp-zeo. The NotI fragment of pBS246-mloxp-zeo containing the zeocin resistance marker flanked by two 5171 loxP sites was then cloned into the NotI site of pUC119-Nhet-del.LMP1 to construct pUC119-Nhet-del.LMP1-zeo.
To construct a plasmid expressing full-length LMP1, an LMP1 cDNA of Akata strain EBV was cloned into the pGEM-T easy vector (Promega) and verified by DNA sequencing. The LMP1 cDNA was subcloned into the pSG5 vector (Stratagene), which was then transferred to a vector equipped with a hygromycin resistance marker gene to construct pSG5-AK-LMP1-Hyg.
The sequences of oligonucleotides used for constructions are available upon request.
Electrotransformation of E. coli. Electrocompetent Escherichia coli DH10B bacterial cells were prepared as described previously (35). The Bio-Rad Gene Pulser II electroporation system (0.1-cm cuvette, 1.25 kV, 25 μF, 100 , 40 μl of cells) was used for electrotransformation.
Construction of d.LMP1-EBV. Construction of d.LMP1-EBV was performed in E. coli with Red, RecE, RecT (GET) recombination (29, 31) as described (18). Plasmid pUC119-Nhet-del.LMP1-zeo was PCR amplified with primers NA03 (5'-AGAGCAAGGCCTATGGAAGAGGAG-3') and NA04 (5'-CCTCAGTTGCCTTGCTCCTGCCAC-3') to obtain a linear targeting construct required for GET recombination. The resultant PCR product (912 bp long) had 60-bp (5' end) and 62-bp (3' end) sequences homologous to the target region of AK-BAC-GFP (18). The PCR product obtained was gel purified and DpnI digested, and 200 ng of the linear PCR product was electroporated into recombinase-induced DH10B electrocompetent E. coli (harboring AK-BAC-GFP and pGETrec). Colonies of recombinant BAC clones were identified by plating cells on Luria-Bertani (LB) plates containing chloramphenicol (12.5 μg/ml) and zeocin (20 μg/ml). The modified BAC clone (d.LMP1-EBV-zeo) was purified from pGETrec by miniprep DNA isolation and electroporation into E. coli DH10B.
To remove the zeocin resistance marker, the miniprep DNA of d.LMP1-EBV-zeo was treated by Cre recombinase (Novagen) in vitro according to the manufacturer's instructions. One microliter of the reaction mixture was then used to transform electrocompetent E. coli DH10B, followed by selection on LB plates containing chloramphenicol. Bacterial clones of d.LMP1-EBV which had the zeocin marker gene deleted were chosen by checking their sensitivity to zeocin.
BACmid preparation and DNA analysis. For minipreparations, each BAC plasmid (BACmid) was isolated from 1.5 ml of a bacterial culture with an alkaline lysis protocol (Current Protocols in Human Genetics, Unit 5.15. Construction of Bacterial Artificial Chromosome BAC/PAC libraries, found at http://murdoch.rch.unimelb.edu.au/). For maxipreparations, BACmid DNA was isolated from each 500-ml bacterial culture with the Nucleobond BAC 100 kit (Macherey-Nagel, Duren, Germany). BACmid DNAs digested with restriction enzymes were resolved by 0.8% agarose gel electrophoresis for 16 h at 40 V and visualized by ethidium bromide staining.
Transfection of BACmids into Akata cells. Akata cells (5.0 x 106) were transfected with 10 μg of d.LMP1-EBV DNA (isolated by maxipreparation) via electroporation (Bio-Rad Gene Pulser II; 0.190 kV, 950 μF). Transfected cells were resuspended in 5 ml of culture medium and plated into six-well dishes. At 2 days posttransfection, cells were plated at 104 cells per well in 96-well tissue culture plates in medium containing 375 μg of G418 (Sigma) per ml. Half of the culture medium was replaced with fresh G418-containing medium every 5 days until G418-resistant cell clones emerged (3 to 4 weeks after plating). Episomal DNA fractions were prepared from G418-resistant cell clones by an alkaline lysis procedure as described previously (18, 38). The purified episomal DNAs were used to transform electrocompetent E. coli DH10B to rescue episomally maintained BACmids as bacterial clones. Out of 39 G418-resistant cell clones, 21 cell clones contained episomally maintained BACmids, and two cell clones contained BACmids identical to the transfected d.LMP1-EBV.
Southern blot analysis. Genomic DNAs were extracted by the standard proteinase K-sodium dodecyl sulfate method, followed by phenol-chloroform extraction and ethanol precipitation. Southern blotting was performed as previously described (27). The BamHI X and EcoRI K fragments of the Akata strain EBV genome were used as probes for Southern blotting.
Immunofluorescence. Cells were smeared on glass slides and fixed with acetone for 2 min. Indirect immunofluorescence was performed with either monoclonal antibody Cl.50-1 (specific to gp110) or monoclonal antibody C-1 (specific to gp350) as primary antibodies and a Cy3-conjugated anti-mouse immunoglobulin G (IgG) (Jackson ImmunoResearch) as a secondary antibody.
Immunoblots. Cells (1.0 x 106) were harvested before and after inducing virus production (48 h after starting anti-IgG treatment) and suspended in 200 μl of low-salt buffer. The cell suspensions were sonicated and centrifuged at 10,000 rpm at 4°C for 5 min, and the supernatant fractions were collected. Ten microliters of each protein sample was resolved in sodium dodecyl sulfate-8.0% polyacrylamide gels and blotted onto nitrocellulose membranes. The expression of LMP1 was detected with monoclonal antibody S12 (specific to LMP1) (25, 45) as a primary antibody and horseradish peroxidase-conjugated anti-mouse IgG as a secondary antibody. Signals were detected by an enhanced chemiluminescence method (Amersham) according to the manufacturer's protocol.
Rescue experiments via LMP1 expression. For transient expression, Akata cells harboring d.LMP1-EBV episomes (d.LMP1-EBV cells) (5.0 x 106 cells) were transfected with 50 μg of plasmid pSG5-XLMP1 with a BTX electroporator (BTX, San Diego, Calif.) at 1,000 μF and 200 V. To get cell clones with constitutive expression of truncated LMP1, cells were cotransfected with 50 μg of pSG5-XLMP1 (24) and 5 μg of the hygromycin B phosphotransferase expression vector. For constitutively expressing full-length LMP1, cells were transfected with 30 μg of plasmid pSG5-AK-LMP1-Hyg by electroporation. At 2 days posttransfection, transfected cells were plated at 104 cells per well in 96-well tissue culture plates in 200 μl of complete medium containing 250 μg of hygromycin B (Calbiochem) and 700 μg of G418 per ml. Cells were fed with fresh selective medium every 5 days until hygromycin B-resistant cell clones emerged (3 to 4 weeks after plating).
Virus production, infection, and preparation of viral DNA. Akata cells harboring EBV episomes (2.0 x 106) were resuspended in 1 ml of fresh medium containing 0.5% rabbit anti-human IgG (DakoCytomation, Carpinteria, Calif.) and incubated for 8 to 12 h. The culture medium was replaced with fresh medium, and 48 h later, the culture supernatant was harvested. The culture supernatant was filtered through a 0.45-μm-pore-size membrane and used as a virus solution. For infection, EBV-negative Akata cells (1.0 x 106) were suspended in 1 ml of filtered virus solution and incubated at 37°C for 90 min with continuous gentle mixing.
For quantification of virions in the culture supernatants, 8 ml of virus solution was prepared as described above and ultracentrifuged at 15,000 rpm with an SW41 rotor (L8-80 M ultracentrifuge, Beckman) at 4°C for 1.5 h. Pellets were resuspended in 400 μl of phosphate-buffered saline, and 200 μl of lysis buffer (3.0% sodium dodecyl sulfate, 75 mM Tris-HCl [pH 8.0], 25 mM EDTA) and 1 μl of proteinase K (10 mg/ml) were added. After incubating the lysates at 37°C for 1 h, they were extracted twice with phenol and once with a chloroform-isoamyl alcohol (24:1) mixture. The viral DNAs were then ethanol precipitated, dissolved in Tris-EDTA, digested with EcoRI, and used for Southern blot analysis.
Flow cytometry analysis. Green fluorescent protein (GFP) expression in infected EBV-negative Akata cells was analyzed by fluorescence-activated cell sorting (FACS) (20,000 events) with a FACScalibur flow cytometer (Becton Dickinson Co., San Jose, Calif.). Statistical differences in infection efficiency between groups were determined by Welch's t test.
Electron microscopy. Cells were pelleted 48 h after starting anti-IgG treatment, and the pellets were washed with washing buffer (0.1 M phosphate buffer [pH 7.4], 7% sucrose) three times. The pellets were then immersed in 2.5% glutaraldehyde (TAAB) in 0.1 M phosphate buffer (pH 7.4) for 24 h at room temperature. The samples were then washed in the same washing buffer, postfixed in 1% osmium tetroxide (Merck) in 0.1 M phosphate buffer for 1.5 h at room temperature, dehydrated in graded acetones, and embedded in Epon (TAAB). Ultrathin sections (0.1 μm) were stained with 3% uranyl acetate for 20 min and 0.1% lead citrate for 15 min, and they were examined under a Hitachi H-7100 electron microscope.
RESULTS
Construction of d.LMP1-EBV via homologous recombination in E. coli. We recently cloned the genome of Akata strain EBV as a BAC clone (designated AK-BAC) and used derivatives of AK-BAC to generate recombinant EBVs (18). One of the derivatives of AK-BAC, designated AK-BAC-GFP, that had a GFP gene and neomycin and kanamycin resistance genes inserted into the region deleted in B95-8 strain EBV produced progeny virus in the absence of wild-type EBV. The recombinant virus AK-BAC-GFP exhibited high B-cell-transforming ability and expressed all of the latently expressed viral genes in the established lymphoblastoid cell lines (18). Therefore, we used AK-BAC-GFP as the starting material for mutagenesis.
A homologous recombination method called GET recombination (29, 31) was employed in order to specifically disrupt the LMP1 gene of AK-BAC-GFP. First, a nonsense linker (encoding triple stop codons) and a zeocin resistance gene were introduced after codon 9 of the LMP1 gene via GET recombination to generate d.LMP1-EBV-Zeor (see Materials and Methods and Fig. 1). Subsequently, the zeocin resistance gene, which was flanked by a pair of 5171 loxP sites (23), was removed by an in vitro Cre recombinase reaction. The resultant BAC clone, designated d.LMP1-EBV, had triple stop codons and one residual 5171 loxP site (with short flanking sequences) inserted after codon 9 of the LMP1 gene. The insertion of triple stop codons after codon 9 of the LMP1 gene is expected to completely abolish LMP1 expression, as only full-length LMP1 protein is expressed from the LMP1 gene of Akata strain EBV, which lacks the Met-129 in the LMP1 gene of B95-8 strain EBV (7, 42). The modified genomic locus of d.LMP1-EBV was PCR amplified, and the modification was verified by DNA sequencing (data not shown). Restriction enzyme digestion analyses revealed the expected changes in band sizes corresponding to the modified LMP1 locus, while the bands of unmodified loci remained unchanged (data not shown).
Generation of Akata cells harboring d.LMP1-EBV. We introduced the DNA of d.LMP1-EBV into Akata cells in order to establish cell lines producing the LMP1-deleted recombinant viruses. We previously showed that transfected AK-BAC-GFP efficiently formed episomes in EBV-positive Akata cells but not in EBV-negative Akata cells (18). Accordingly, EBV-positive Akata cells were transfected with d.LMP1-EBV, and the transfected cells were selected with G418. Screening of 39 G418-resistant cell clones resulted in obtaining two cell clones (clones 28 and 31) that harbored intact d.LMP1-EBV episomes (data not shown).
Cell clone 28 was subjected to further experiments. Virus production was induced from clone 28, and a mixture of wild-type and d.LMP1-EBV was produced. The mixture was then used to infect EBV-negative Akata cells (37) to establish cell lines that contained only d.LMP1-EBV episomes. G418-resistant cells that appeared in individual wells of culture plates were treated with anti-IgG, and they were examined for the induction of viral glycoprotein gp110, one of the EBV glycoproteins that is abundantly expressed in the late lytic phase (13). Three cell clones exhibiting high induction levels of gp110 were chosen, and they were further subcloned by a limiting-dilution protocol. As a result, we obtained three independent cell clones (designated clones 28-1, 28-2, and 28-3) which exhibited high levels of gp110 induction in response to anti-IgG treatment. The levels of gp110 induction observed in cell clones 28-1, 28-2, and 28-3 were found to be comparable to that obtained with AK-BAC-GFP (Fig. 2A and data not shown).
We stimulated virus production by treating the cell clones with anti-IgG and examined whether these cell clones contained only the LMP1-deleted episomes but no wild-type genomes. Genomic DNA of clone 28 (harboring wild-type and d.LMP1-EBV) and cell clones harboring the LMP1-deleted episomes (28-1, 28-2, and 28-3) were prepared before and after anti-IgG treatment, and they were subjected to Southern blot analysis. The result revealed that clone 28 exhibited bands derived from both wild-type and d.LMP1-EBV episomes, while clones 28-1, 28-2, and 28-3 exhibited only the d.LMP1-EBV bands (Fig. 2B). The result also demonstrated that, after stimulating virus production, the genomes of d.LMP1-EBV amplified significantly in the absence of wild-type episomes. The degrees of viral DNA amplification of the LMP1-deleted episomes were comparable to that of its wild-type counterpart (AK-BAC-GFP) (Fig. 2B).
We next examined whether insertion of the triple stop codons into the LMP1 ORF abolished the expression of LMP1 during the lytic cycle of viral replication. The expression of LMP1 was examined before and after stimulating virus production by Western blot analysis. Akata cells retain the Burkitt's-type viral gene expression, and LMP1 expression is undetectable unless virus production is stimulated (42). EBV-positive Akata cells and the cell clones with LMP1-deleted episomes (28-1, 28-2, and 28-3) were then treated with anti-IgG, and the expression of LMP1 protein was examined. Strong inductions of full-length LMP1 protein (55 kDa in sodium dodecyl sulfate-polyacrylamide gel electrophoresis) and its associated faster-migrating products were observed in Akata cells harboring either wild-type EBV or AK-BAC-GFP. By contrast, no LMP1 expression was detected in the cell clones harboring d.LMP1-EBV (Fig. 2C). Taken together, these data demonstrate the successful knockout of LMP1 protein expression as well as the establishment of cell clones harboring only LMP1-deleted episomes.
Nucleocapsid formation occurring inside cell nuclei was not affected by LMP1 loss. We then performed electron microscopic observations to obtain supportive evidence that d.LMP1-EBV was undergoing productive replication in the absence of LMP1 protein. The cell clone harboring AK-BAC-GFP (having the wild-type LMP1 gene) and cell clone 28-1, harboring d.LMP1-EBV, were treated with anti-IgG and embedded in Epon for electron microscopic observation. Intact nucleocapsids, having an electron-dense core surrounded by a sharply bordered electron-dense rim, were abundant in the nucleus of anti-IgG-treated Akata cells harboring AK-BAC-GFP (wild-type LMP1) (Fig. 3A). Similarly, abundant intact nucleocapsids were observed in the nucleus of anti-IgG-treated cell clone 28-1 (Fig. 3B), which completely lacked LMP1 expression. Thus, LMP1 is apparently dispensable for intranuclear nucleocapsid formation. This result is compatible with our observation that d.LMP1-EBV is undergoing lytic replication in the absence of LMP1.
Culture supernatants of d.LMP1-EBV exhibited poor infection efficiency. We next tested whether the cell clones with the LMP1-deleted EBV could efficiently produce infectious viruses. Culture supernatants were prepared from cell clone 28-1 after anti-IgG treatment and used to infect EBV-negative Akata cells. The culture supernatant containing AK-BAC-GFP virus (18) was also used for another infection as a control. Infection was detected by observing GFP-positive cells under the fluorescent microscope at 2 days postinfection (Fig. 4A), and the efficiency of infection was determined by flow cytometry (Table 1). We found that approximately 0.4% of the recipient cells became GFP positive after the infection of the LMP1-deleted virus, while more than 16% of the recipient cells became GFP positive after the infection of AK-BAC-GFP virus. This was unexpected; because we had observed that loss of LMP1 did not affect late-lytic protein (gp110) expression (Fig. 2A), viral genome amplification (Fig. 2B), and intranuclear nucleocapsid formation (Fig. 3B).
These observations prompted us to examine the expression of the most predominant glycoprotein, gp350/220 (a BLLF1 gene product), which is required for efficient EBV infection of B cells (17, 27, 30). We found that the cell clone harboring the LMP1-deleted EBV expressed gp350/220 as efficiently as the cell clones harboring AK-BAC-GFP (Fig. 4B). This result excluded the possibility that the poor infectivity of the LMP1-deleted virus was due to poor expression of gp350/220.
Expression of truncated LMP1 restored infection efficiency. The impaired infection efficiency of the culture supernatants of LMP1-deleted EBV could have been due to accidental genetic alterations acquired elsewhere on the EBV genome during the course of LMP1 mutagenesis in E. coli. To rule out such a possibility, we tried to rescue the infection efficiency by supplying LMP1 protein in trans. For the following reasons, truncated LMP1 was chosen for the initial experiments. First, previous studies have demonstrated that making stable cell clones expressing high levels of the full-length LMP1 is difficult due to its cytotoxic effect (14), whereas it is relatively easy with truncated LMP1. Second, it is well known that truncated LMP1, not the full-length LMP1, is strongly induced upon the induction of virus production in B95-8 cells (4, 5, 16, 28, 34, 44). Therefore, truncated LMP1 is apparently worth being tested for the rescue experiment.
First, Akata cells harboring the LMP1-deleted EBV (clone 28-1), referred to d.LMP1-EBV cells hereafter, were transiently transfected with the expression vector encoding truncated LMP1 (24). Transfected cells were then treated with anti-IgG, and the culture supernatant was used to infect EBV-negative Akata cells. We found that supplying truncated LMP1 prior to virus production slightly rescued the infectivity of the culture supernatant of d.LMP1-EBV cells, while transfecting the control vector did not (Fig. 4C). Similar results were obtained when two other d.LMP1-EBV cell clones (28-2 and 28-3) were subjected to the same experiment (data not shown).
These results prompted us to examine whether the constitutive expression of truncated LMP1 could further rescue the infectivity of the culture supernatant of d.LMP1-EBV cells. For this purpose, d.LMP1-EBV cells (clone 28-1) were stably transfected with the expression vector encoding truncated LMP1. We examined the expression of truncated LMP1 by Western blot analysis and found that four out of five cell clones examined expressed truncated LMP1 (45 kDa) (Fig. 5A). Three independent cell clones (clones 1, 3, and 9) exhibited good gp110 induction levels comparable to the level observed in anti-IgG-treated Akata cells harboring AK-BAC-GFP. These cell clones were then treated with anti-IgG, and the culture supernatants were used to infect EBV-negative Akata cells. We used FACS analyses to quantitatively determine the infection efficiency. We found that the infection efficiency obtained with these culture supernatants (7 to 12%) was comparable to that obtained with AK-BAC-GFP virus (more than 16%), and they were significantly higher (P < 0.05) than the infection efficiency obtained with the parental d.LMP1-EBV cells (Fig. 5B and Table 1). Thus, the phenotype of poor infectivity observed for the LMP1-deleted virus was successfully restored by constitutive expression of truncated LMP1.
Loss of LMP1 caused marked reduction of virus release into culture supernatants, and constitutive expression of truncated LMP1 restored it. Two possibilities remained regarding the cause of the poor infection efficiency of the LMP1-deleted viruses: either the infectivity was impaired or the amount of viruses in the culture supernatants was reduced by LMP1 loss. We therefore checked the amounts of viruses in the culture supernatants after anti-IgG treatment of various Akata-derived cell clones. The amounts of virus in the culture supernatants were examined by centrifugal pelleting of viruses, DNA extraction, and Southern blot analysis. We found that anti-IgG-treated d.LMP1-EBV cells released significantly fewer viruses into the culture supernatants compared to the cell clone harboring AK-BAC-GFP (Fig. 5C, compare lanes 1 and 3). Interestingly, when truncated LMP1 was constitutively expressed in d.LMP1-EBV cells, the amount of virus in the culture supernatant after anti-IgG treatment increased significantly (Fig. 5C, lane 2). Thus, loss of LMP1 impaired virus release into the culture supernatant, and the constitutive expression of truncated LMP1 significantly restored it. This result indicates that the impaired infection efficiency observed for the LMP1-deleted EBV is most likely due to the reduction of virus release into the culture supernatant rather than to impaired infectivity of the viruses.
Effect of full-length LMP1 expression on virus production. Akata strain EBV exhibits strong expression of full-length LMP1 upon induction of the virus productive cycle. Therefore, our next question was whether full-length LMP1 could rescue the phenotype caused by LMP1 knockout. To answer this question, d.LMP1-EBV cells were stably transfected with the expression vector encoding the Akata strain (full-length) LMP1 and hygromycin B phosphotransferase (pSG5-AK-LMP1-Hyg). The resultant hygromycin-resistant cell clones were examined for the expression of full-length LMP1. Western blot analyses revealed that, out of 40 hygromycin-resistant cell clones examined, only a few expressed full-length LMP1, the size of which was identical to that of the LMP1 expressed in lymphoblastoid cell lines (Fig. 6A and data not shown). Cell clone 27 was identified as the highest LMP1 expresser. However, the expression level of full-length LMP1 in cell clone 27 was only about 20% of that of a lymphoblastoid cell line, which was far less than that obtained with truncated LMP1 (compare Fig. 5A and Fig. 6A).
Cell clones 16, 27, and 35, which exhibited low to moderate levels of LMP1 expression, were treated with anti-IgG, and the induction levels of the late lytic protein gp110 were examined by immunofluorescence. This result revealed that the gp110 induction levels of these cell clones were comparable to that of anti-IgG-treated Akata cells harboring AK-BAC-GFP (data not shown). The resultant culture supernatants were then used to infect EBV-negative Akata cells. The culture supernatant of cell clone 27 exhibited the highest infection efficiency (3%) (Fig. 6B and Table 1), which was significantly higher (P < 0.05) than the result with the parental d.LMP1-EBV cells (clone 28-1) (Table 1). On the other hand, the results with cell clones 16 and 35 were indistinguishable from that with the parental cell clone (data not shown).
In summary, we found that not only truncated LMP1 but also full-length LMP1 could rescue virus release in Akata cells harboring LMP1-deleted EBV.
DISCUSSION
It is well known that the expression of LMP1 is strongly upregulated upon the induction of viral lytic replication (4, 5, 7, 8, 16, 28, 34, 42, 44). However, the biological significance of this strong LMP1 induction during lytic replication has been unclear. Here we provide the first direct evidence that LMP1 protein plays a critical role during the lytic phase of viral replication.
We specifically disrupted the LMP1 gene of Akata strain EBV with a BAC system and established Akata cell clones infected with the LMP1-deleted EBV. By treating d.LMP1-EBV cells with anti-IgG, LMP1-deleted EBV entered lytic replication as efficiently as the wild-type counterpart even in the absence of LMP1 induction. However, the culture supernatant of the d.LMP1-EBV cells exhibited severely impaired infection efficiency. We found that the impaired infection efficiency was due to the reduction of virus release into the culture supernatant. Importantly, we were able to rescue virus release and recover the infectivity of the culture supernatant by expressing truncated LMP1 in d.LMP1-EBV cells. Furthermore, the expression of full-length LMP1 could partially rescue the infectivity of the culture supernatant. These results demonstrate that LMP1 gene products are truly responsible for efficient virus release in Akata cells.
We admit that the degree of rescued infectivity observed for full-length LMP1 was less than that observed for truncated LMP1 (Table 1). A possible reason for this inefficient rescue is that the steady-state expression level of full-length LMP1 in d.LMP1-EBV cells was far less than that observed in lymphoblastoid cell lines (Fig. 6A) or in anti-IgG-treated EBV-positive Akata cells (Fig. 2C). Full-length LMP1 could have rescued the infectivity more efficiently if we had established higher LMP1 expressers. However, it is well known that making stable cell clones expressing high levels of full-length LMP1 is difficult due to its cytotoxic effect (14).
Our model is that strong induction of LMP1 expression is critical for efficient virus release, but previous studies have proposed an apparently conflicting model in which LMP1 can inhibit lytic cycle progression (1, 32). One study demonstrated that constitutive LMP1 expression in lymphoblastoid cell lines (while EBNA2 expression was shut off) could inhibit EBV reactivation (1), and another study showed that transient overexpression of LMP1 could inhibit the virus-productive cycle in a derivative of the P3HR1 cell line (32). However, in our rescue experiments, we saw no significant reduction in late lytic protein expression (gp110 and gp350) after constitutively expressing low to moderate levels of full-length LMP1 in d.LMP1-EBV cells (data not shown). The discrepancy between our study and the previous studies could be due to the different experimental systems and different expression levels of LMP1 proteins.
We still do not know where and how LMP1 can augment the process of virus release. A working model of intracellular nucleocapsid transport involves a deenvelopment process from nucleus to cytoplasm and a reenvelopment process either at post-Golgi-derived cytoplasmic vesicles or at the plasma membrane (13). It is possible that the LMP1 enhances the deenvelopment and reenvelopment processes either at the nuclear membrane or at the cytoplasm during the lytic phase. Full-length LMP1 is known to localize to lipid microdomains, designated lipid rafts, located on the plasma membrane (15, 20). However, recent studies reinvestigated the intracellular localization of LMP1 protein and found that a substantial fraction of LMP1 protein localized to intracellular compartments (11, 22). Therefore, it is possible that full-length LMP1 expressed shortly after virus induction localizes to intracellular compartments and augments virus production.
Another open question is how full-length LMP1 induction in stimulated Akata cells can be reconciled with truncated LMP1 induction in stimulated B95-8 cells (43). We found that introducing triple stop codons after codon 9 of the LMP1 gene of Akata strain EBV abolished the induction of not only full-length LMP1 but also smaller LMP1-immunoreactive proteins (Fig. 2C). This result clearly indicates that proteolytic cleavage of full-length LMP1 generates these smaller products. As the epitope of the monoclonal antibody used for the Western analyses resides in the C terminus of LMP1 protein (25), they are apparently N-terminally truncated. Therefore, the possibility remains that these N-terminally truncated LMP1 products actually play a role in augmenting virus release in Akata cells. Whether these N-terminally truncated LMP1 products correspond to the lytic LMP1 in stimulated B95-8 cells remains to be clarified.
In summary, we propose a model in which EBV utilizes differentially expressed LMP1 gene products for dual purposes during its life cycle. First, in latently infected B lymphocytes, steady-state expression of full-length LMP1 is critical for maintaining transformation status. Second, at the onset of the virus production cycle, the strong induction of LMP1 enables efficient virus release from cells. In B95-8 cells, it is probably a truncated LMP1 that plays a critical role for virus release, fitting well with our finding that truncated LMP1 can restore the defect of LMP1-deleted virus. In Akata cells, it is full-length LMP1 or its associated N-terminally truncated LMP1 that plays a critical role during virus production.
LMP1 is apparently one of the most important viral proteins absolutely required for EBV's life cycle. Choosing such an important protein for augmenting virus release may be a matter of importance for virus evolution.
ACKNOWLEDGMENTS
We thank P. Ioannou (Murdoch Institute, Melbourne, Australia) for the generous gift of plasmid pGETrec, M. Sato for expertise with electron microscopy, and S. Maruo for FACS analysis. We also thank J. M. Martin (Boulder University), D. Iwakiri, and M. Tanaka for helpful suggestions.
This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (T.K. and K.T.), Akiyama Memorial Foundation (T.K.), and Uehara Memorial Foundation (K.T.).
REFERENCES
Adler, B., E. Schaadt, B. Kempkes, U. Zimber-Strobl, B. Baier, and G. W. Bornkamm. 2002. Control of Epstein-Barr virus reactivation by activated CD40 and viral latent membrane protein 1. Proc. Natl. Acad. Sci. USA 99:437-442.
Baer, R., A. T. Bankier, M. D. Biggin, P. L. Deininger, P. J. Farrell, T. J. Gibson, G. Hatfull, G. S. Hudson, S. C. Satchwell, C. Seguin, et al. 1984. DNA sequence and expression of the B95-8 Epstein-Barr virus genome. Nature 310:207-211.
Baichwal, V. R., and B. Sugden. 1989. The multiple membrane-spanning segments of the BNLF-1 oncogene from Epstein-Barr virus are required for transformation. Oncogene 4:67-74.
Baichwal, V. R., and B. Sugden. 1987. Posttranslational processing of an Epstein-Barr virus-encoded membrane protein expressed in cells transformed by Epstein-Barr virus. J. Virol. 61:866-875.
Boos, H., R. Berger, C. Kuklik-Roos, T. Iftner, and N. Mueller-Lantzsch. 1987. Enhancement of Epstein-Barr virus membrane protein (LMP) expression by serum, TPA, or n-butyrate in latently infected Raji cells. Virology 159:161-165.
Cahir McFarland, E. D., K. M. Izumi, and G. Mosialos. 1999. Epstein-barr virus transformation: involvement of latent membrane protein 1-mediated activation of NF-kappaB. Oncogene 18:6959-6964.
Erickson, K. D., C. Berger, W. F. Coffin, 3rd, E. Schiff, D. M. Walling, and J. M. Martin. 2003. Unexpected absence of the Epstein-Barr virus (EBV) lyLMP-1 open reading frame in tumor virus isolates: lack of correlation between Met129 status and EBV strain identity. J. Virol. 77:4415-4422.
Erickson, K. D., and J. M. Martin. 1997. Early detection of the lytic LMP-1 protein in EBV-infected B-cells suggests its presence in the virion. Virology 234:1-13.
Erickson, K. D., and J. M. Martin. 2000. The late lytic LMP-1 protein of Epstein-Barr virus can negatively regulate LMP-1 signaling. J. Virol. 74:1057-1060.
Fennewald, S., V. van Santen, and E. Kieff. 1984. Nucleotide sequence of an mRNA transcribed in latent growth-transforming virus infection indicates that it may encode a membrane protein. J. Virol. 51:411-419.
Flanagan, J., J. Middeldorp, and T. Sculley. 2003. Localization of the Epstein-Barr virus protein LMP 1 to exosomes. J. Gen. Virol. 84:1871-1879.
Gires, O., U. Zimber-Strobl, R. Gonnella, M. Ueffing, G. Marschall, R. Zeidler, D. Pich, and W. Hammerschmidt. 1997. Latent membrane protein 1 of Epstein-Barr virus mimics a constitutively active receptor molecule. EMBO J. 16:6131-6140.
Gong, M., and E. Kieff. 1990. Intracellular trafficking of two major Epstein-Barr virus glycoproteins, gp350/220 and gp110. J. Virol. 64:1507-1516.
Hammerschmidt, W., B. Sugden, and V. R. Baichwal. 1989. The transforming domain alone of the latent membrane protein of Epstein-Barr virus is toxic to cells when expressed at high levels. J. Virol. 63:2469-2475.
Higuchi, M., K. M. Izumi, and E. Kieff. 2001. Epstein-Barr virus latent-infection membrane proteins are palmitoylated and raft-associated: protein 1 binds to the cytoskeleton through TNF receptor cytoplasmic factors. Proc. Natl. Acad. Sci. USA 98:4675-4680.
Hudson, G. S., P. J. Farrell, and B. G. Barrell. 1985. Two related but differentially expressed potential membrane proteins encoded by the EcoRI Dhet region of Epstein-Barr virus B95-8. J. Virol. 53:528-535.
Janz, A., M. Oezel, C. Kurzeder, J. Mautner, D. Pich, M. Kost, W. Hammerschmidt, and H. J. Delecluse. 2000. Infectious Epstein-Barr virus lacking major glycoprotein BLLF1 (gp350/220) demonstrates the existence of additional viral ligands. J. Virol. 74:10142-10152.
Kanda, T., M. Yajima, N. Ahsan, M. Tanaka, and K. Takada. 2004. Production of high-titer Epstein-Barr virus recombinants derived from Akata cells by using a bacterial artificial chromosome system. J. Virol. 78:7004-7015.
Kaye, K. M., K. M. Izumi, and E. Kieff. 1993. Epstein-Barr virus latent membrane protein 1 is essential for B-lymphocyte growth transformation. Proc. Natl. Acad. Sci. USA 90:9150-9154.
Kaykas, A., K. Worringer, and B. Sugden. 2001. CD40 and LMP-1 both signal from lipid rafts but LMP-1 assembles a distinct, more efficient signaling complex. EMBO J. 20:2641-2654.
Kieff, E., and A. B. Rikinson. 2001. Epstein-Barr virus and its replication, p. 2511-2573. In M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Lam, N., and B. Sugden. 2003. LMP1, a viral relative of the TNF receptor family, signals principally from intracellular compartments. EMBO J. 22:3027-3038.
Lee, G., and I. Saito. 1998. Role of nucleotide sequences of loxP spacer region in Cre-mediated recombination. Gene 216:55-65.
Liebowitz, D., J. Mannick, K. Takada, and E. Kieff. 1992. Phenotypes of Epstein-Barr virus LMP1 deletion mutants indicate transmembrane and amino-terminal cytoplasmic domains necessary for effects in B-lymphoma cells. J. Virol. 66:4612-4616.
Mann, K. P., D. Staunton, and D. A. Thorley-Lawson. 1985. Epstein-Barr virus-encoded protein found in plasma membranes of transformed cells. J. Virol. 55:710-720.
Martin, J., and B. Sugden. 1991. The latent membrane protein oncoprotein resembles growth factor receptors in the properties of its turnover. Cell Growth Differ. 2:653-700.
Maruo, S., L. Yang, and K. Takada. 2001. Roles of Epstein-Barr virus glycoproteins gp350 and gp25 in the infection of human epithelial cells. J. Gen. Virol. 82:2373-2383.
Modrow, S., and H. Wolf. 1986. Characterization of two related Epstein-Barr virus-encoded membrane proteins that are differentially expressed in Burkitt lymphoma and in vitro-transformed cell lines. Proc. Natl. Acad. Sci. USA 83:5703-5707.
Narayanan, K., R. Williamson, Y. Zhang, A. F. Stewart, and P. A. Ioannou. 1999. Efficient and precise engineering of a 200 kb beta-globin human/bacterial artificial chromosome in E. coli DH10B using an inducible homologous recombination system. Gene Ther. 6:442-447.
Nemerow, G. R., R. A. Houghten, M. D. Moore, and N. R. Cooper. 1989. Identification of an epitope in the major envelope protein of Epstein-Barr virus that mediates viral binding to the B lymphocyte EBV receptor (CR2). Cell 56:369-377.
Orford, M., M. Nefedov, J. Vadolas, F. Zaibak, R. Williamson, and P. A. Ioannou. 2000. Engineering EGFP reporter constructs into a 200 kb human beta-globin BAC clone using GET Recombination. Nucleic Acids Res. 28:E84.
Prince, S., S. Keating, C. Fielding, P. Brennan, E. Floettmann, and M. Rowe. 2003. Latent membrane protein 1 inhibits Epstein-Barr virus lytic cycle induction and progress via different mechanisms. J. Virol. 77:5000-5007.
Rickinson, A. B., and E. Kieff. 2001. Epstein-Barr virus, p. 2575-2627. In M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Rowe, M., H. S. Evans, L. S. Young, K. Hennessy, E. Kieff, and A. B. Rickinson. 1987. Monoclonal antibodies to the latent membrane protein of Epstein-Barr virus reveal heterogeneity of the protein and inducible expression in virus-transformed cells. J. Gen. Virol. 68:1575-1586.
Sharma, R. C., and R. T. Schimke. 1996. Preparation of electrocompetent E. coli using salt-free growth medium. BioTechniques 20:42-44.
Shimizu, N., A. Tanabe-Tochikura, Y. Kuroiwa, and K. Takada. 1994. Isolation of Epstein-Barr virus (EBV)-negative cell clones from the EBV-positive Burkitt's lymphoma (BL) line Akata: malignant phenotypes of BL cells are dependent on EBV. J. Virol. 68:6069-6073.
Shimizu, N., H. Yoshiyama, and K. Takada. 1996. Clonal propagation of Epstein-Barr virus (EBV) recombinants in EBV-negative Akata cells. J. Virol. 70:7260-7263.
Simpson, K., and C. Huxley. 1996. A shuttle system for transfer of YACs between yeast and mammalian cells. Nucleic Acids Res. 24:4693-4699.
Sugden, B. 1989. An intricate route to immortality. Cell 57:5-7.
Takada, K. 1984. Cross-linking of cell surface immunoglobulins induces Epstein-Barr virus in Burkitt lymphoma lines. Int. J. Cancer 33:27-32.
Takada, K., K. Horinouchi, Y. Ono, T. Aya, T. Osato, M. Takahashi, and S. Hayasaka. 1991. An Epstein-Barr virus-producer line Akata: establishment of the cell line and analysis of viral DNA. Virus Genes 5:147-156.
Torii, T., K. Konishi, J. Sample, and K. Takada. 1998. The truncated form of the Epstein-Barr virus LMP-1 is dispensable or complementable by full-length form in virus infection and replication. Virology 251:273-278.
Vazirabadi, G., T. R. Geiger, W. F. Coffin, 3rd, and J. M. Martin. 2003. Epstein-Barr virus latent membrane protein-1 (LMP-1) and lytic LMP-1 localization in plasma membrane-derived extracellular vesicles and intracellular virions. J. Gen. Virol. 84:1997-2008.
Wang, D., D. Liebowitz, and E. Kieff. 1988. The truncated form of the Epstein-Barr virus latent-infection membrane protein expressed in virus replication does not transform rodent fibroblasts. J. Virol. 62:2337-2346.
Wang, D., D. Liebowitz, F. Wang, C. Gregory, A. Rickinson, R. Larson, T. Springer, and E. Kieff. 1988. Epstein-Barr virus latent infection membrane protein alters the human B-lymphocyte phenotype: deletion of the amino terminus abolishes activity. J. Virol. 62:4173-4184.(Nazmul Ahsan, Teru Kanda,)
Center for Virus Vector Development, Institute for Genetic Medicine, Hokkaido University
Laboratory of Molecular and Cellular Pathology, Hokkaido University Graduate School of Medicine, Sapporo, Japan
ABSTRACT
The Epstein-Barr virus (EBV) latent membrane protein 1 (LMP1), which is critical for EBV-induced B-cell transformation, is also abundantly expressed during the lytic cycle of viral replication. However, the biological significance of this strong LMP1 induction remains unknown. We engineered a bacterial artificial chromosome clone containing the entire genome of Akata strain EBV to specifically disrupt the LMP1 gene. Akata cell clones harboring the episomes of LMP1-deleted EBV were established, and the effect of LMP1 loss on virus production was investigated. We found that the degree of viral DNA amplification and the expression levels of viral late gene products were unaffected by LMP1 loss, demonstrating that the LMP1-deleted EBV entered the lytic replication cycle as efficiently as the wild-type counterpart. This was confirmed by our electron microscopic observation that nucleocapsid formation inside nuclei occurred even in the absence of LMP1. By contrast, loss of LMP1 severely impaired virus release into culture supernatants, resulting in poor infection efficiency. The expression of truncated LMP1 in Akata cells harboring LMP1-deleted EBV rescued the virus release into the culture supernatant and the infectivity, and full-length LMP1 partially rescued the infectivity. These results indicate that inducible expression of LMP1 during the viral lytic cycle plays a critical role in virus production.
INTRODUCTION
Epstein-Barr virus (EBV) is a human herpesvirus that establishes a life-long persistence in the host (33). The virus infects the vast majority of the world's adult population and is well known for its association with a broad spectrum of benign and malignant diseases, including infectious mononucleosis, Burkitt's lymphoma, nasopharyngeal carcinoma, B-cell lymphoma in immunocompromised patients, and gastric carcinoma (33). In vitro EBV infection of peripheral B lymphocytes results in their immortalization and continuous proliferation (39). In immortalized lymphoblastoid cell lines, EBV expresses six nuclear proteins (EBNAs), three integral membrane proteins (LMP1, -2A, and -2B), two small nuclear RNAs (EBER1 and -2), and a transcript from the BamHI-A region (21).
Latent membrane protein 1 (LMP1) is one of the viral gene products that are essential for B-cell transformation (19). LMP1 is an integral membrane protein composed of a short cytoplasmic amino-terminal domain, six hydrophobic transmembrane domains, and a cytoplasmic carboxy-terminal domain (10). It has been demonstrated that LMP1 acts as a constitutively active receptor that mimics activated CD40, a member of the tumor necrosis factor receptor family (12, 26). The cytoplasmic carboxy terminus of LMP1 plays a critical role in EBV-induced B-cell transformation through its binding to a tumor necrosis factor receptor-associated factor (TRAF) and a tumor necrosis factor receptor-associated death domain (TRADD) protein (6, 21).
The LMP1 (BNLF1) gene contains three exons that are located within the BamHI-N region of the EBV genome (10). Two open reading frames (ORFs) have been identified based on nucleotide sequences (2, 10) and mRNA mapping (16) of B95-8 strain EBV. A transcript starting from the ED-L1 promoter, which is located upstream of the first exon (16), encodes the first ORF. This ORF encodes full-length LMP1 (386 amino acids) that is abundantly expressed in lymphoblastoid cell lines. Another transcript starting from the EDL1A promoter, which is located within the first intron of the LMP1 gene, encodes the second ORF. The translation initiation site of this second ORF is methionine-129 of full-length LMP1 (16), and the ORF thus encodes an amino-terminally truncated form of the LMP1 protein. The truncated LMP1 (258 amino acids) consists of the fifth and sixth transmembrane domains and the cytoplasmic carboxy terminus of full-length LMP1.
The B95-8 cell line, an in vitro-immortalized lymphoblastoid cell line, was initially used to characterize truncated LMP1. A number of studies demonstrated that expression of truncated LMP1 was specifically upregulated during the lytic phase of viral replication (4, 5, 16, 28, 34, 44). Therefore, truncated form of LMP1 is frequently referred to as lytic LMP1 (8, 43). In contrast to full-length LMP1, truncated LMP1 does not transform rodent cells (3, 44) or alter the phenotypes of human B lymphocytes (45). The only biological activity of truncated LMP1 identified to date is its ability to negatively regulate LMP1 signaling pathways (9), although truncated LMP1 retains the carboxy terminus of LMP1.
Recent studies revealed that Met-129 of the B95-8 LMP1 ORF is not common in many virus isolates (7, 42). For example, the LMP1 gene of Akata strain EBV encodes isoleucine at codon 129, resulting in loss of the translation initiation codon for the ORF of truncated LMP1. The Akata cell line is derived from an EBV-positive Burkitt's lymphoma from a Japanese patient (41). LMP1 expression is not detected in unstimulated Akata cells (42), as Akata cells retain the Burkitt's-type viral gene expression (36). A unique feature of Akata cells is that cross-linking of cell surface immunoglobulin with anti-immunoglobulin antibodies can efficiently induce the lytic cycle of viral replication (40). Interestingly, expression of full-length LMP1 is strongly upregulated after anti-immunoglobulin treatment of Akata cells (42). The induction of full-length LMP1 from Met-129-negative EBV was also observed in Raji and BL74 cells after tetradecanoyl phorbol acetate and butyrate treatment (5, 7). Therefore, the induction of full-length LMP1 after stimulating virus production is not peculiar to Akata cells.
These observations prompted us to investigate the biological significance of LMP1 upregulation during the lytic cycle of viral replication in Akata cells. We therefore set out to generate a recombinant Akata EBV containing the disrupted LMP1 gene. We recently reported the establishment of a bacterial artificial chromosome (BAC) system for engineering the genome of Akata strain EBV and used the system to produce high-titer recombinant viruses (18). We employed the system to disrupt the LMP1 gene of Akata strain EBV and investigated whether Akata cells harboring the episomes of LMP1-deleted EBV (d.LMP1-EBV) could produce infectious viruses. We found that the d.LMP1-EBV entered the lytic replication cycle as efficiently as the wild-type counterpart. However, we unexpectedly found that loss of LMP1 severely impaired virus release into the culture supernatant, resulting in poor infection efficiency. We expressed either truncated LMP1 or full-length LMP1 in Akata cells harboring d.LMP1-EBV to rescue the infectivity of the culture supernatants. The results revealed that loss of LMP1 is truly responsible for the impaired virus release.
MATERIALS AND METHODS
Cell culture. Akata, a human Burkitt's lymphoma-derived cell line carrying EBV as episomes (41) and EBV-negative Akata cells (36) were maintained in RPMI 1640 medium (Sigma-Aldrich Fine Chemicals, St. Louis, Mo.) containing 10% fetal bovine serum at 37°C in a 5% humidified CO2 atmosphere.
Plasmids. A plasmid, pBS246 (Life Technologies), having two loxP sites and multiple cloning sites, was used for excising a loxP cassette. Two loxP sites of pBS246, one located between BspEI and SpeI and the other between EcoRI and Acc65I, were replaced with mutated loxP sites (5171 loxP) with 43-bp synthetic oligonucleotides to construct pBS246-mloxp5171-EXA. The 5171 loxP site has the substitution of two nucleotide pairs within the asymmetric 8-nucleotide spacer region of the loxP sequence, and therefore it barely recombines with the wild-type loxP site (23).
A synthetic oligonucleotide containing triple stop codons and restriction enzyme sites (BamHI, NotI, and SalI) was inserted into the SfiI site (after codon 9 of the LMP1 gene) of pUC119-Nhet (pUC119 containing the BamHI-Nhet fragment of the Akata EBV genome) to construct pUC119-Nhet-del.LMP1.
The blunted FokI-BclI fragment of pCDNA4/HisMax (Invitrogen, Carlsbad, Calif.) containing the zeocin resistance marker gene (zeo) was cloned into the EcoRV site of pBS246-mloxp5171-EXA to make pBS246-mloxp-zeo. The NotI fragment of pBS246-mloxp-zeo containing the zeocin resistance marker flanked by two 5171 loxP sites was then cloned into the NotI site of pUC119-Nhet-del.LMP1 to construct pUC119-Nhet-del.LMP1-zeo.
To construct a plasmid expressing full-length LMP1, an LMP1 cDNA of Akata strain EBV was cloned into the pGEM-T easy vector (Promega) and verified by DNA sequencing. The LMP1 cDNA was subcloned into the pSG5 vector (Stratagene), which was then transferred to a vector equipped with a hygromycin resistance marker gene to construct pSG5-AK-LMP1-Hyg.
The sequences of oligonucleotides used for constructions are available upon request.
Electrotransformation of E. coli. Electrocompetent Escherichia coli DH10B bacterial cells were prepared as described previously (35). The Bio-Rad Gene Pulser II electroporation system (0.1-cm cuvette, 1.25 kV, 25 μF, 100 , 40 μl of cells) was used for electrotransformation.
Construction of d.LMP1-EBV. Construction of d.LMP1-EBV was performed in E. coli with Red, RecE, RecT (GET) recombination (29, 31) as described (18). Plasmid pUC119-Nhet-del.LMP1-zeo was PCR amplified with primers NA03 (5'-AGAGCAAGGCCTATGGAAGAGGAG-3') and NA04 (5'-CCTCAGTTGCCTTGCTCCTGCCAC-3') to obtain a linear targeting construct required for GET recombination. The resultant PCR product (912 bp long) had 60-bp (5' end) and 62-bp (3' end) sequences homologous to the target region of AK-BAC-GFP (18). The PCR product obtained was gel purified and DpnI digested, and 200 ng of the linear PCR product was electroporated into recombinase-induced DH10B electrocompetent E. coli (harboring AK-BAC-GFP and pGETrec). Colonies of recombinant BAC clones were identified by plating cells on Luria-Bertani (LB) plates containing chloramphenicol (12.5 μg/ml) and zeocin (20 μg/ml). The modified BAC clone (d.LMP1-EBV-zeo) was purified from pGETrec by miniprep DNA isolation and electroporation into E. coli DH10B.
To remove the zeocin resistance marker, the miniprep DNA of d.LMP1-EBV-zeo was treated by Cre recombinase (Novagen) in vitro according to the manufacturer's instructions. One microliter of the reaction mixture was then used to transform electrocompetent E. coli DH10B, followed by selection on LB plates containing chloramphenicol. Bacterial clones of d.LMP1-EBV which had the zeocin marker gene deleted were chosen by checking their sensitivity to zeocin.
BACmid preparation and DNA analysis. For minipreparations, each BAC plasmid (BACmid) was isolated from 1.5 ml of a bacterial culture with an alkaline lysis protocol (Current Protocols in Human Genetics, Unit 5.15. Construction of Bacterial Artificial Chromosome BAC/PAC libraries, found at http://murdoch.rch.unimelb.edu.au/). For maxipreparations, BACmid DNA was isolated from each 500-ml bacterial culture with the Nucleobond BAC 100 kit (Macherey-Nagel, Duren, Germany). BACmid DNAs digested with restriction enzymes were resolved by 0.8% agarose gel electrophoresis for 16 h at 40 V and visualized by ethidium bromide staining.
Transfection of BACmids into Akata cells. Akata cells (5.0 x 106) were transfected with 10 μg of d.LMP1-EBV DNA (isolated by maxipreparation) via electroporation (Bio-Rad Gene Pulser II; 0.190 kV, 950 μF). Transfected cells were resuspended in 5 ml of culture medium and plated into six-well dishes. At 2 days posttransfection, cells were plated at 104 cells per well in 96-well tissue culture plates in medium containing 375 μg of G418 (Sigma) per ml. Half of the culture medium was replaced with fresh G418-containing medium every 5 days until G418-resistant cell clones emerged (3 to 4 weeks after plating). Episomal DNA fractions were prepared from G418-resistant cell clones by an alkaline lysis procedure as described previously (18, 38). The purified episomal DNAs were used to transform electrocompetent E. coli DH10B to rescue episomally maintained BACmids as bacterial clones. Out of 39 G418-resistant cell clones, 21 cell clones contained episomally maintained BACmids, and two cell clones contained BACmids identical to the transfected d.LMP1-EBV.
Southern blot analysis. Genomic DNAs were extracted by the standard proteinase K-sodium dodecyl sulfate method, followed by phenol-chloroform extraction and ethanol precipitation. Southern blotting was performed as previously described (27). The BamHI X and EcoRI K fragments of the Akata strain EBV genome were used as probes for Southern blotting.
Immunofluorescence. Cells were smeared on glass slides and fixed with acetone for 2 min. Indirect immunofluorescence was performed with either monoclonal antibody Cl.50-1 (specific to gp110) or monoclonal antibody C-1 (specific to gp350) as primary antibodies and a Cy3-conjugated anti-mouse immunoglobulin G (IgG) (Jackson ImmunoResearch) as a secondary antibody.
Immunoblots. Cells (1.0 x 106) were harvested before and after inducing virus production (48 h after starting anti-IgG treatment) and suspended in 200 μl of low-salt buffer. The cell suspensions were sonicated and centrifuged at 10,000 rpm at 4°C for 5 min, and the supernatant fractions were collected. Ten microliters of each protein sample was resolved in sodium dodecyl sulfate-8.0% polyacrylamide gels and blotted onto nitrocellulose membranes. The expression of LMP1 was detected with monoclonal antibody S12 (specific to LMP1) (25, 45) as a primary antibody and horseradish peroxidase-conjugated anti-mouse IgG as a secondary antibody. Signals were detected by an enhanced chemiluminescence method (Amersham) according to the manufacturer's protocol.
Rescue experiments via LMP1 expression. For transient expression, Akata cells harboring d.LMP1-EBV episomes (d.LMP1-EBV cells) (5.0 x 106 cells) were transfected with 50 μg of plasmid pSG5-XLMP1 with a BTX electroporator (BTX, San Diego, Calif.) at 1,000 μF and 200 V. To get cell clones with constitutive expression of truncated LMP1, cells were cotransfected with 50 μg of pSG5-XLMP1 (24) and 5 μg of the hygromycin B phosphotransferase expression vector. For constitutively expressing full-length LMP1, cells were transfected with 30 μg of plasmid pSG5-AK-LMP1-Hyg by electroporation. At 2 days posttransfection, transfected cells were plated at 104 cells per well in 96-well tissue culture plates in 200 μl of complete medium containing 250 μg of hygromycin B (Calbiochem) and 700 μg of G418 per ml. Cells were fed with fresh selective medium every 5 days until hygromycin B-resistant cell clones emerged (3 to 4 weeks after plating).
Virus production, infection, and preparation of viral DNA. Akata cells harboring EBV episomes (2.0 x 106) were resuspended in 1 ml of fresh medium containing 0.5% rabbit anti-human IgG (DakoCytomation, Carpinteria, Calif.) and incubated for 8 to 12 h. The culture medium was replaced with fresh medium, and 48 h later, the culture supernatant was harvested. The culture supernatant was filtered through a 0.45-μm-pore-size membrane and used as a virus solution. For infection, EBV-negative Akata cells (1.0 x 106) were suspended in 1 ml of filtered virus solution and incubated at 37°C for 90 min with continuous gentle mixing.
For quantification of virions in the culture supernatants, 8 ml of virus solution was prepared as described above and ultracentrifuged at 15,000 rpm with an SW41 rotor (L8-80 M ultracentrifuge, Beckman) at 4°C for 1.5 h. Pellets were resuspended in 400 μl of phosphate-buffered saline, and 200 μl of lysis buffer (3.0% sodium dodecyl sulfate, 75 mM Tris-HCl [pH 8.0], 25 mM EDTA) and 1 μl of proteinase K (10 mg/ml) were added. After incubating the lysates at 37°C for 1 h, they were extracted twice with phenol and once with a chloroform-isoamyl alcohol (24:1) mixture. The viral DNAs were then ethanol precipitated, dissolved in Tris-EDTA, digested with EcoRI, and used for Southern blot analysis.
Flow cytometry analysis. Green fluorescent protein (GFP) expression in infected EBV-negative Akata cells was analyzed by fluorescence-activated cell sorting (FACS) (20,000 events) with a FACScalibur flow cytometer (Becton Dickinson Co., San Jose, Calif.). Statistical differences in infection efficiency between groups were determined by Welch's t test.
Electron microscopy. Cells were pelleted 48 h after starting anti-IgG treatment, and the pellets were washed with washing buffer (0.1 M phosphate buffer [pH 7.4], 7% sucrose) three times. The pellets were then immersed in 2.5% glutaraldehyde (TAAB) in 0.1 M phosphate buffer (pH 7.4) for 24 h at room temperature. The samples were then washed in the same washing buffer, postfixed in 1% osmium tetroxide (Merck) in 0.1 M phosphate buffer for 1.5 h at room temperature, dehydrated in graded acetones, and embedded in Epon (TAAB). Ultrathin sections (0.1 μm) were stained with 3% uranyl acetate for 20 min and 0.1% lead citrate for 15 min, and they were examined under a Hitachi H-7100 electron microscope.
RESULTS
Construction of d.LMP1-EBV via homologous recombination in E. coli. We recently cloned the genome of Akata strain EBV as a BAC clone (designated AK-BAC) and used derivatives of AK-BAC to generate recombinant EBVs (18). One of the derivatives of AK-BAC, designated AK-BAC-GFP, that had a GFP gene and neomycin and kanamycin resistance genes inserted into the region deleted in B95-8 strain EBV produced progeny virus in the absence of wild-type EBV. The recombinant virus AK-BAC-GFP exhibited high B-cell-transforming ability and expressed all of the latently expressed viral genes in the established lymphoblastoid cell lines (18). Therefore, we used AK-BAC-GFP as the starting material for mutagenesis.
A homologous recombination method called GET recombination (29, 31) was employed in order to specifically disrupt the LMP1 gene of AK-BAC-GFP. First, a nonsense linker (encoding triple stop codons) and a zeocin resistance gene were introduced after codon 9 of the LMP1 gene via GET recombination to generate d.LMP1-EBV-Zeor (see Materials and Methods and Fig. 1). Subsequently, the zeocin resistance gene, which was flanked by a pair of 5171 loxP sites (23), was removed by an in vitro Cre recombinase reaction. The resultant BAC clone, designated d.LMP1-EBV, had triple stop codons and one residual 5171 loxP site (with short flanking sequences) inserted after codon 9 of the LMP1 gene. The insertion of triple stop codons after codon 9 of the LMP1 gene is expected to completely abolish LMP1 expression, as only full-length LMP1 protein is expressed from the LMP1 gene of Akata strain EBV, which lacks the Met-129 in the LMP1 gene of B95-8 strain EBV (7, 42). The modified genomic locus of d.LMP1-EBV was PCR amplified, and the modification was verified by DNA sequencing (data not shown). Restriction enzyme digestion analyses revealed the expected changes in band sizes corresponding to the modified LMP1 locus, while the bands of unmodified loci remained unchanged (data not shown).
Generation of Akata cells harboring d.LMP1-EBV. We introduced the DNA of d.LMP1-EBV into Akata cells in order to establish cell lines producing the LMP1-deleted recombinant viruses. We previously showed that transfected AK-BAC-GFP efficiently formed episomes in EBV-positive Akata cells but not in EBV-negative Akata cells (18). Accordingly, EBV-positive Akata cells were transfected with d.LMP1-EBV, and the transfected cells were selected with G418. Screening of 39 G418-resistant cell clones resulted in obtaining two cell clones (clones 28 and 31) that harbored intact d.LMP1-EBV episomes (data not shown).
Cell clone 28 was subjected to further experiments. Virus production was induced from clone 28, and a mixture of wild-type and d.LMP1-EBV was produced. The mixture was then used to infect EBV-negative Akata cells (37) to establish cell lines that contained only d.LMP1-EBV episomes. G418-resistant cells that appeared in individual wells of culture plates were treated with anti-IgG, and they were examined for the induction of viral glycoprotein gp110, one of the EBV glycoproteins that is abundantly expressed in the late lytic phase (13). Three cell clones exhibiting high induction levels of gp110 were chosen, and they were further subcloned by a limiting-dilution protocol. As a result, we obtained three independent cell clones (designated clones 28-1, 28-2, and 28-3) which exhibited high levels of gp110 induction in response to anti-IgG treatment. The levels of gp110 induction observed in cell clones 28-1, 28-2, and 28-3 were found to be comparable to that obtained with AK-BAC-GFP (Fig. 2A and data not shown).
We stimulated virus production by treating the cell clones with anti-IgG and examined whether these cell clones contained only the LMP1-deleted episomes but no wild-type genomes. Genomic DNA of clone 28 (harboring wild-type and d.LMP1-EBV) and cell clones harboring the LMP1-deleted episomes (28-1, 28-2, and 28-3) were prepared before and after anti-IgG treatment, and they were subjected to Southern blot analysis. The result revealed that clone 28 exhibited bands derived from both wild-type and d.LMP1-EBV episomes, while clones 28-1, 28-2, and 28-3 exhibited only the d.LMP1-EBV bands (Fig. 2B). The result also demonstrated that, after stimulating virus production, the genomes of d.LMP1-EBV amplified significantly in the absence of wild-type episomes. The degrees of viral DNA amplification of the LMP1-deleted episomes were comparable to that of its wild-type counterpart (AK-BAC-GFP) (Fig. 2B).
We next examined whether insertion of the triple stop codons into the LMP1 ORF abolished the expression of LMP1 during the lytic cycle of viral replication. The expression of LMP1 was examined before and after stimulating virus production by Western blot analysis. Akata cells retain the Burkitt's-type viral gene expression, and LMP1 expression is undetectable unless virus production is stimulated (42). EBV-positive Akata cells and the cell clones with LMP1-deleted episomes (28-1, 28-2, and 28-3) were then treated with anti-IgG, and the expression of LMP1 protein was examined. Strong inductions of full-length LMP1 protein (55 kDa in sodium dodecyl sulfate-polyacrylamide gel electrophoresis) and its associated faster-migrating products were observed in Akata cells harboring either wild-type EBV or AK-BAC-GFP. By contrast, no LMP1 expression was detected in the cell clones harboring d.LMP1-EBV (Fig. 2C). Taken together, these data demonstrate the successful knockout of LMP1 protein expression as well as the establishment of cell clones harboring only LMP1-deleted episomes.
Nucleocapsid formation occurring inside cell nuclei was not affected by LMP1 loss. We then performed electron microscopic observations to obtain supportive evidence that d.LMP1-EBV was undergoing productive replication in the absence of LMP1 protein. The cell clone harboring AK-BAC-GFP (having the wild-type LMP1 gene) and cell clone 28-1, harboring d.LMP1-EBV, were treated with anti-IgG and embedded in Epon for electron microscopic observation. Intact nucleocapsids, having an electron-dense core surrounded by a sharply bordered electron-dense rim, were abundant in the nucleus of anti-IgG-treated Akata cells harboring AK-BAC-GFP (wild-type LMP1) (Fig. 3A). Similarly, abundant intact nucleocapsids were observed in the nucleus of anti-IgG-treated cell clone 28-1 (Fig. 3B), which completely lacked LMP1 expression. Thus, LMP1 is apparently dispensable for intranuclear nucleocapsid formation. This result is compatible with our observation that d.LMP1-EBV is undergoing lytic replication in the absence of LMP1.
Culture supernatants of d.LMP1-EBV exhibited poor infection efficiency. We next tested whether the cell clones with the LMP1-deleted EBV could efficiently produce infectious viruses. Culture supernatants were prepared from cell clone 28-1 after anti-IgG treatment and used to infect EBV-negative Akata cells. The culture supernatant containing AK-BAC-GFP virus (18) was also used for another infection as a control. Infection was detected by observing GFP-positive cells under the fluorescent microscope at 2 days postinfection (Fig. 4A), and the efficiency of infection was determined by flow cytometry (Table 1). We found that approximately 0.4% of the recipient cells became GFP positive after the infection of the LMP1-deleted virus, while more than 16% of the recipient cells became GFP positive after the infection of AK-BAC-GFP virus. This was unexpected; because we had observed that loss of LMP1 did not affect late-lytic protein (gp110) expression (Fig. 2A), viral genome amplification (Fig. 2B), and intranuclear nucleocapsid formation (Fig. 3B).
These observations prompted us to examine the expression of the most predominant glycoprotein, gp350/220 (a BLLF1 gene product), which is required for efficient EBV infection of B cells (17, 27, 30). We found that the cell clone harboring the LMP1-deleted EBV expressed gp350/220 as efficiently as the cell clones harboring AK-BAC-GFP (Fig. 4B). This result excluded the possibility that the poor infectivity of the LMP1-deleted virus was due to poor expression of gp350/220.
Expression of truncated LMP1 restored infection efficiency. The impaired infection efficiency of the culture supernatants of LMP1-deleted EBV could have been due to accidental genetic alterations acquired elsewhere on the EBV genome during the course of LMP1 mutagenesis in E. coli. To rule out such a possibility, we tried to rescue the infection efficiency by supplying LMP1 protein in trans. For the following reasons, truncated LMP1 was chosen for the initial experiments. First, previous studies have demonstrated that making stable cell clones expressing high levels of the full-length LMP1 is difficult due to its cytotoxic effect (14), whereas it is relatively easy with truncated LMP1. Second, it is well known that truncated LMP1, not the full-length LMP1, is strongly induced upon the induction of virus production in B95-8 cells (4, 5, 16, 28, 34, 44). Therefore, truncated LMP1 is apparently worth being tested for the rescue experiment.
First, Akata cells harboring the LMP1-deleted EBV (clone 28-1), referred to d.LMP1-EBV cells hereafter, were transiently transfected with the expression vector encoding truncated LMP1 (24). Transfected cells were then treated with anti-IgG, and the culture supernatant was used to infect EBV-negative Akata cells. We found that supplying truncated LMP1 prior to virus production slightly rescued the infectivity of the culture supernatant of d.LMP1-EBV cells, while transfecting the control vector did not (Fig. 4C). Similar results were obtained when two other d.LMP1-EBV cell clones (28-2 and 28-3) were subjected to the same experiment (data not shown).
These results prompted us to examine whether the constitutive expression of truncated LMP1 could further rescue the infectivity of the culture supernatant of d.LMP1-EBV cells. For this purpose, d.LMP1-EBV cells (clone 28-1) were stably transfected with the expression vector encoding truncated LMP1. We examined the expression of truncated LMP1 by Western blot analysis and found that four out of five cell clones examined expressed truncated LMP1 (45 kDa) (Fig. 5A). Three independent cell clones (clones 1, 3, and 9) exhibited good gp110 induction levels comparable to the level observed in anti-IgG-treated Akata cells harboring AK-BAC-GFP. These cell clones were then treated with anti-IgG, and the culture supernatants were used to infect EBV-negative Akata cells. We used FACS analyses to quantitatively determine the infection efficiency. We found that the infection efficiency obtained with these culture supernatants (7 to 12%) was comparable to that obtained with AK-BAC-GFP virus (more than 16%), and they were significantly higher (P < 0.05) than the infection efficiency obtained with the parental d.LMP1-EBV cells (Fig. 5B and Table 1). Thus, the phenotype of poor infectivity observed for the LMP1-deleted virus was successfully restored by constitutive expression of truncated LMP1.
Loss of LMP1 caused marked reduction of virus release into culture supernatants, and constitutive expression of truncated LMP1 restored it. Two possibilities remained regarding the cause of the poor infection efficiency of the LMP1-deleted viruses: either the infectivity was impaired or the amount of viruses in the culture supernatants was reduced by LMP1 loss. We therefore checked the amounts of viruses in the culture supernatants after anti-IgG treatment of various Akata-derived cell clones. The amounts of virus in the culture supernatants were examined by centrifugal pelleting of viruses, DNA extraction, and Southern blot analysis. We found that anti-IgG-treated d.LMP1-EBV cells released significantly fewer viruses into the culture supernatants compared to the cell clone harboring AK-BAC-GFP (Fig. 5C, compare lanes 1 and 3). Interestingly, when truncated LMP1 was constitutively expressed in d.LMP1-EBV cells, the amount of virus in the culture supernatant after anti-IgG treatment increased significantly (Fig. 5C, lane 2). Thus, loss of LMP1 impaired virus release into the culture supernatant, and the constitutive expression of truncated LMP1 significantly restored it. This result indicates that the impaired infection efficiency observed for the LMP1-deleted EBV is most likely due to the reduction of virus release into the culture supernatant rather than to impaired infectivity of the viruses.
Effect of full-length LMP1 expression on virus production. Akata strain EBV exhibits strong expression of full-length LMP1 upon induction of the virus productive cycle. Therefore, our next question was whether full-length LMP1 could rescue the phenotype caused by LMP1 knockout. To answer this question, d.LMP1-EBV cells were stably transfected with the expression vector encoding the Akata strain (full-length) LMP1 and hygromycin B phosphotransferase (pSG5-AK-LMP1-Hyg). The resultant hygromycin-resistant cell clones were examined for the expression of full-length LMP1. Western blot analyses revealed that, out of 40 hygromycin-resistant cell clones examined, only a few expressed full-length LMP1, the size of which was identical to that of the LMP1 expressed in lymphoblastoid cell lines (Fig. 6A and data not shown). Cell clone 27 was identified as the highest LMP1 expresser. However, the expression level of full-length LMP1 in cell clone 27 was only about 20% of that of a lymphoblastoid cell line, which was far less than that obtained with truncated LMP1 (compare Fig. 5A and Fig. 6A).
Cell clones 16, 27, and 35, which exhibited low to moderate levels of LMP1 expression, were treated with anti-IgG, and the induction levels of the late lytic protein gp110 were examined by immunofluorescence. This result revealed that the gp110 induction levels of these cell clones were comparable to that of anti-IgG-treated Akata cells harboring AK-BAC-GFP (data not shown). The resultant culture supernatants were then used to infect EBV-negative Akata cells. The culture supernatant of cell clone 27 exhibited the highest infection efficiency (3%) (Fig. 6B and Table 1), which was significantly higher (P < 0.05) than the result with the parental d.LMP1-EBV cells (clone 28-1) (Table 1). On the other hand, the results with cell clones 16 and 35 were indistinguishable from that with the parental cell clone (data not shown).
In summary, we found that not only truncated LMP1 but also full-length LMP1 could rescue virus release in Akata cells harboring LMP1-deleted EBV.
DISCUSSION
It is well known that the expression of LMP1 is strongly upregulated upon the induction of viral lytic replication (4, 5, 7, 8, 16, 28, 34, 42, 44). However, the biological significance of this strong LMP1 induction during lytic replication has been unclear. Here we provide the first direct evidence that LMP1 protein plays a critical role during the lytic phase of viral replication.
We specifically disrupted the LMP1 gene of Akata strain EBV with a BAC system and established Akata cell clones infected with the LMP1-deleted EBV. By treating d.LMP1-EBV cells with anti-IgG, LMP1-deleted EBV entered lytic replication as efficiently as the wild-type counterpart even in the absence of LMP1 induction. However, the culture supernatant of the d.LMP1-EBV cells exhibited severely impaired infection efficiency. We found that the impaired infection efficiency was due to the reduction of virus release into the culture supernatant. Importantly, we were able to rescue virus release and recover the infectivity of the culture supernatant by expressing truncated LMP1 in d.LMP1-EBV cells. Furthermore, the expression of full-length LMP1 could partially rescue the infectivity of the culture supernatant. These results demonstrate that LMP1 gene products are truly responsible for efficient virus release in Akata cells.
We admit that the degree of rescued infectivity observed for full-length LMP1 was less than that observed for truncated LMP1 (Table 1). A possible reason for this inefficient rescue is that the steady-state expression level of full-length LMP1 in d.LMP1-EBV cells was far less than that observed in lymphoblastoid cell lines (Fig. 6A) or in anti-IgG-treated EBV-positive Akata cells (Fig. 2C). Full-length LMP1 could have rescued the infectivity more efficiently if we had established higher LMP1 expressers. However, it is well known that making stable cell clones expressing high levels of full-length LMP1 is difficult due to its cytotoxic effect (14).
Our model is that strong induction of LMP1 expression is critical for efficient virus release, but previous studies have proposed an apparently conflicting model in which LMP1 can inhibit lytic cycle progression (1, 32). One study demonstrated that constitutive LMP1 expression in lymphoblastoid cell lines (while EBNA2 expression was shut off) could inhibit EBV reactivation (1), and another study showed that transient overexpression of LMP1 could inhibit the virus-productive cycle in a derivative of the P3HR1 cell line (32). However, in our rescue experiments, we saw no significant reduction in late lytic protein expression (gp110 and gp350) after constitutively expressing low to moderate levels of full-length LMP1 in d.LMP1-EBV cells (data not shown). The discrepancy between our study and the previous studies could be due to the different experimental systems and different expression levels of LMP1 proteins.
We still do not know where and how LMP1 can augment the process of virus release. A working model of intracellular nucleocapsid transport involves a deenvelopment process from nucleus to cytoplasm and a reenvelopment process either at post-Golgi-derived cytoplasmic vesicles or at the plasma membrane (13). It is possible that the LMP1 enhances the deenvelopment and reenvelopment processes either at the nuclear membrane or at the cytoplasm during the lytic phase. Full-length LMP1 is known to localize to lipid microdomains, designated lipid rafts, located on the plasma membrane (15, 20). However, recent studies reinvestigated the intracellular localization of LMP1 protein and found that a substantial fraction of LMP1 protein localized to intracellular compartments (11, 22). Therefore, it is possible that full-length LMP1 expressed shortly after virus induction localizes to intracellular compartments and augments virus production.
Another open question is how full-length LMP1 induction in stimulated Akata cells can be reconciled with truncated LMP1 induction in stimulated B95-8 cells (43). We found that introducing triple stop codons after codon 9 of the LMP1 gene of Akata strain EBV abolished the induction of not only full-length LMP1 but also smaller LMP1-immunoreactive proteins (Fig. 2C). This result clearly indicates that proteolytic cleavage of full-length LMP1 generates these smaller products. As the epitope of the monoclonal antibody used for the Western analyses resides in the C terminus of LMP1 protein (25), they are apparently N-terminally truncated. Therefore, the possibility remains that these N-terminally truncated LMP1 products actually play a role in augmenting virus release in Akata cells. Whether these N-terminally truncated LMP1 products correspond to the lytic LMP1 in stimulated B95-8 cells remains to be clarified.
In summary, we propose a model in which EBV utilizes differentially expressed LMP1 gene products for dual purposes during its life cycle. First, in latently infected B lymphocytes, steady-state expression of full-length LMP1 is critical for maintaining transformation status. Second, at the onset of the virus production cycle, the strong induction of LMP1 enables efficient virus release from cells. In B95-8 cells, it is probably a truncated LMP1 that plays a critical role for virus release, fitting well with our finding that truncated LMP1 can restore the defect of LMP1-deleted virus. In Akata cells, it is full-length LMP1 or its associated N-terminally truncated LMP1 that plays a critical role during virus production.
LMP1 is apparently one of the most important viral proteins absolutely required for EBV's life cycle. Choosing such an important protein for augmenting virus release may be a matter of importance for virus evolution.
ACKNOWLEDGMENTS
We thank P. Ioannou (Murdoch Institute, Melbourne, Australia) for the generous gift of plasmid pGETrec, M. Sato for expertise with electron microscopy, and S. Maruo for FACS analysis. We also thank J. M. Martin (Boulder University), D. Iwakiri, and M. Tanaka for helpful suggestions.
This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan (T.K. and K.T.), Akiyama Memorial Foundation (T.K.), and Uehara Memorial Foundation (K.T.).
REFERENCES
Adler, B., E. Schaadt, B. Kempkes, U. Zimber-Strobl, B. Baier, and G. W. Bornkamm. 2002. Control of Epstein-Barr virus reactivation by activated CD40 and viral latent membrane protein 1. Proc. Natl. Acad. Sci. USA 99:437-442.
Baer, R., A. T. Bankier, M. D. Biggin, P. L. Deininger, P. J. Farrell, T. J. Gibson, G. Hatfull, G. S. Hudson, S. C. Satchwell, C. Seguin, et al. 1984. DNA sequence and expression of the B95-8 Epstein-Barr virus genome. Nature 310:207-211.
Baichwal, V. R., and B. Sugden. 1989. The multiple membrane-spanning segments of the BNLF-1 oncogene from Epstein-Barr virus are required for transformation. Oncogene 4:67-74.
Baichwal, V. R., and B. Sugden. 1987. Posttranslational processing of an Epstein-Barr virus-encoded membrane protein expressed in cells transformed by Epstein-Barr virus. J. Virol. 61:866-875.
Boos, H., R. Berger, C. Kuklik-Roos, T. Iftner, and N. Mueller-Lantzsch. 1987. Enhancement of Epstein-Barr virus membrane protein (LMP) expression by serum, TPA, or n-butyrate in latently infected Raji cells. Virology 159:161-165.
Cahir McFarland, E. D., K. M. Izumi, and G. Mosialos. 1999. Epstein-barr virus transformation: involvement of latent membrane protein 1-mediated activation of NF-kappaB. Oncogene 18:6959-6964.
Erickson, K. D., C. Berger, W. F. Coffin, 3rd, E. Schiff, D. M. Walling, and J. M. Martin. 2003. Unexpected absence of the Epstein-Barr virus (EBV) lyLMP-1 open reading frame in tumor virus isolates: lack of correlation between Met129 status and EBV strain identity. J. Virol. 77:4415-4422.
Erickson, K. D., and J. M. Martin. 1997. Early detection of the lytic LMP-1 protein in EBV-infected B-cells suggests its presence in the virion. Virology 234:1-13.
Erickson, K. D., and J. M. Martin. 2000. The late lytic LMP-1 protein of Epstein-Barr virus can negatively regulate LMP-1 signaling. J. Virol. 74:1057-1060.
Fennewald, S., V. van Santen, and E. Kieff. 1984. Nucleotide sequence of an mRNA transcribed in latent growth-transforming virus infection indicates that it may encode a membrane protein. J. Virol. 51:411-419.
Flanagan, J., J. Middeldorp, and T. Sculley. 2003. Localization of the Epstein-Barr virus protein LMP 1 to exosomes. J. Gen. Virol. 84:1871-1879.
Gires, O., U. Zimber-Strobl, R. Gonnella, M. Ueffing, G. Marschall, R. Zeidler, D. Pich, and W. Hammerschmidt. 1997. Latent membrane protein 1 of Epstein-Barr virus mimics a constitutively active receptor molecule. EMBO J. 16:6131-6140.
Gong, M., and E. Kieff. 1990. Intracellular trafficking of two major Epstein-Barr virus glycoproteins, gp350/220 and gp110. J. Virol. 64:1507-1516.
Hammerschmidt, W., B. Sugden, and V. R. Baichwal. 1989. The transforming domain alone of the latent membrane protein of Epstein-Barr virus is toxic to cells when expressed at high levels. J. Virol. 63:2469-2475.
Higuchi, M., K. M. Izumi, and E. Kieff. 2001. Epstein-Barr virus latent-infection membrane proteins are palmitoylated and raft-associated: protein 1 binds to the cytoskeleton through TNF receptor cytoplasmic factors. Proc. Natl. Acad. Sci. USA 98:4675-4680.
Hudson, G. S., P. J. Farrell, and B. G. Barrell. 1985. Two related but differentially expressed potential membrane proteins encoded by the EcoRI Dhet region of Epstein-Barr virus B95-8. J. Virol. 53:528-535.
Janz, A., M. Oezel, C. Kurzeder, J. Mautner, D. Pich, M. Kost, W. Hammerschmidt, and H. J. Delecluse. 2000. Infectious Epstein-Barr virus lacking major glycoprotein BLLF1 (gp350/220) demonstrates the existence of additional viral ligands. J. Virol. 74:10142-10152.
Kanda, T., M. Yajima, N. Ahsan, M. Tanaka, and K. Takada. 2004. Production of high-titer Epstein-Barr virus recombinants derived from Akata cells by using a bacterial artificial chromosome system. J. Virol. 78:7004-7015.
Kaye, K. M., K. M. Izumi, and E. Kieff. 1993. Epstein-Barr virus latent membrane protein 1 is essential for B-lymphocyte growth transformation. Proc. Natl. Acad. Sci. USA 90:9150-9154.
Kaykas, A., K. Worringer, and B. Sugden. 2001. CD40 and LMP-1 both signal from lipid rafts but LMP-1 assembles a distinct, more efficient signaling complex. EMBO J. 20:2641-2654.
Kieff, E., and A. B. Rikinson. 2001. Epstein-Barr virus and its replication, p. 2511-2573. In M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Lam, N., and B. Sugden. 2003. LMP1, a viral relative of the TNF receptor family, signals principally from intracellular compartments. EMBO J. 22:3027-3038.
Lee, G., and I. Saito. 1998. Role of nucleotide sequences of loxP spacer region in Cre-mediated recombination. Gene 216:55-65.
Liebowitz, D., J. Mannick, K. Takada, and E. Kieff. 1992. Phenotypes of Epstein-Barr virus LMP1 deletion mutants indicate transmembrane and amino-terminal cytoplasmic domains necessary for effects in B-lymphoma cells. J. Virol. 66:4612-4616.
Mann, K. P., D. Staunton, and D. A. Thorley-Lawson. 1985. Epstein-Barr virus-encoded protein found in plasma membranes of transformed cells. J. Virol. 55:710-720.
Martin, J., and B. Sugden. 1991. The latent membrane protein oncoprotein resembles growth factor receptors in the properties of its turnover. Cell Growth Differ. 2:653-700.
Maruo, S., L. Yang, and K. Takada. 2001. Roles of Epstein-Barr virus glycoproteins gp350 and gp25 in the infection of human epithelial cells. J. Gen. Virol. 82:2373-2383.
Modrow, S., and H. Wolf. 1986. Characterization of two related Epstein-Barr virus-encoded membrane proteins that are differentially expressed in Burkitt lymphoma and in vitro-transformed cell lines. Proc. Natl. Acad. Sci. USA 83:5703-5707.
Narayanan, K., R. Williamson, Y. Zhang, A. F. Stewart, and P. A. Ioannou. 1999. Efficient and precise engineering of a 200 kb beta-globin human/bacterial artificial chromosome in E. coli DH10B using an inducible homologous recombination system. Gene Ther. 6:442-447.
Nemerow, G. R., R. A. Houghten, M. D. Moore, and N. R. Cooper. 1989. Identification of an epitope in the major envelope protein of Epstein-Barr virus that mediates viral binding to the B lymphocyte EBV receptor (CR2). Cell 56:369-377.
Orford, M., M. Nefedov, J. Vadolas, F. Zaibak, R. Williamson, and P. A. Ioannou. 2000. Engineering EGFP reporter constructs into a 200 kb human beta-globin BAC clone using GET Recombination. Nucleic Acids Res. 28:E84.
Prince, S., S. Keating, C. Fielding, P. Brennan, E. Floettmann, and M. Rowe. 2003. Latent membrane protein 1 inhibits Epstein-Barr virus lytic cycle induction and progress via different mechanisms. J. Virol. 77:5000-5007.
Rickinson, A. B., and E. Kieff. 2001. Epstein-Barr virus, p. 2575-2627. In M. Knipe and P. M. Howley (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Rowe, M., H. S. Evans, L. S. Young, K. Hennessy, E. Kieff, and A. B. Rickinson. 1987. Monoclonal antibodies to the latent membrane protein of Epstein-Barr virus reveal heterogeneity of the protein and inducible expression in virus-transformed cells. J. Gen. Virol. 68:1575-1586.
Sharma, R. C., and R. T. Schimke. 1996. Preparation of electrocompetent E. coli using salt-free growth medium. BioTechniques 20:42-44.
Shimizu, N., A. Tanabe-Tochikura, Y. Kuroiwa, and K. Takada. 1994. Isolation of Epstein-Barr virus (EBV)-negative cell clones from the EBV-positive Burkitt's lymphoma (BL) line Akata: malignant phenotypes of BL cells are dependent on EBV. J. Virol. 68:6069-6073.
Shimizu, N., H. Yoshiyama, and K. Takada. 1996. Clonal propagation of Epstein-Barr virus (EBV) recombinants in EBV-negative Akata cells. J. Virol. 70:7260-7263.
Simpson, K., and C. Huxley. 1996. A shuttle system for transfer of YACs between yeast and mammalian cells. Nucleic Acids Res. 24:4693-4699.
Sugden, B. 1989. An intricate route to immortality. Cell 57:5-7.
Takada, K. 1984. Cross-linking of cell surface immunoglobulins induces Epstein-Barr virus in Burkitt lymphoma lines. Int. J. Cancer 33:27-32.
Takada, K., K. Horinouchi, Y. Ono, T. Aya, T. Osato, M. Takahashi, and S. Hayasaka. 1991. An Epstein-Barr virus-producer line Akata: establishment of the cell line and analysis of viral DNA. Virus Genes 5:147-156.
Torii, T., K. Konishi, J. Sample, and K. Takada. 1998. The truncated form of the Epstein-Barr virus LMP-1 is dispensable or complementable by full-length form in virus infection and replication. Virology 251:273-278.
Vazirabadi, G., T. R. Geiger, W. F. Coffin, 3rd, and J. M. Martin. 2003. Epstein-Barr virus latent membrane protein-1 (LMP-1) and lytic LMP-1 localization in plasma membrane-derived extracellular vesicles and intracellular virions. J. Gen. Virol. 84:1997-2008.
Wang, D., D. Liebowitz, and E. Kieff. 1988. The truncated form of the Epstein-Barr virus latent-infection membrane protein expressed in virus replication does not transform rodent fibroblasts. J. Virol. 62:2337-2346.
Wang, D., D. Liebowitz, F. Wang, C. Gregory, A. Rickinson, R. Larson, T. Springer, and E. Kieff. 1988. Epstein-Barr virus latent infection membrane protein alters the human B-lymphocyte phenotype: deletion of the amino terminus abolishes activity. J. Virol. 62:4173-4184.(Nazmul Ahsan, Teru Kanda,)