Synchronous and sequential activation of latently infected Epstein-Barr virus genomes

Association of Epstein-Barr virus (EBV) with Sjögren's syndrome: differential EBV expression between epithelial cells and lymphocytes in salivary glands

The association of Epstein Barr virus (EBV) with Sjögren's syndrome (SS) is still in dispute. This study is aimed to investigate the existence of EBV genomes and their products in salivary glands of SS. Salivary gland samples were surgically obtained from Chinese patients. EBV DNA was detected in three of seven cases by dot blot hybridization and in four of seven cases by in situ hybridization. The EBV-encoded small RNA-1 (EBER1) was detected in two of seven cases by in situ hybridization. The immunohistochemical staining of EBV proteins showed that the EBV latent membrane protein-1 was detected in four of seven cases and that BZLF1, BALF2, and gp350/220 proteins associating with virus production were not expressed. In eight controls, no positive signal was observed by these methods. DNA in situ hybridization identified ERV on both epithelial cells and lymphocytes. On the other hand, EBER1-positive signals were exclusively localized on lymphocytes. These results indicate that two forms of EBV infection may exist in salivary glands of SS. One is EBER1-positive latency in lymphocytes, the other is EBER1-negative latency in epithelial cells. Frequent EBV detection in salivary glands of SS suggests that EBV plays a role in the genesis of SS.

Clonal propagation of Epstein-Barr virus (EBV) recombinants in EBV-negative Akata cells

We lack a host cell supporting an efficient lytic replication of Epstein-Barr virus (EBV). Recently, we isolated EBV-negative cell clones from the Akata cell line (referred as Akata- [N. Shimizu, A. Tanabe-Tochikura, Y. Kuroiwa, and K. Takada, J. Virol. 68:6069-6073, 1994). Since the parental Akata line is one of the highest EBV producers, we examined whether Akata- cells had become a good host for EBV propagation. The parental Akata cells have about 20 copies of EBV plasmid per cell. A drug resistance gene was inserted into one of them by homologous recombination. The resultant virus preparation, a mixture of wild-type and recombinant EBV, was used to infect Akata- cells. After incubation in the selective medium, drug-resistant Akata- cell clones were isolated and proved to be infected with recombinant EBV only. By treatment of the cells with antiimmunoglobulin antibodies, a large amount of recombinant EBV (i.e., more than 10 microg/1-liter culture) was produced. In contrast, three other B-lymphoma lines, BJAB, Ramos, and Louckes, were nonpermissive for virus replication. These results indicate that Akata- cells are suitable for propagation of recombinant EBV clonally, which becomes a powerful tool for determining EBV genetics and which makes it possible to use EBV as a vector for gene therapy.

Epstein-Barr viral latency is disrupted by the immediate-early BRLF1 protein through a cell-specific mechanism

Epstein-Barr virus (EBV), the causative agent of infectious mononucleosis, is a human herpesvirus associated with epithelial cell malignancies (nasopharyngeal carcinoma) as well as B-cell malignancies. Understanding how viral latency is disrupted is a central issue in herpesvirus biology. Epithelial cells are the major site of lytic EBV replication within the human host, and viral reactivation occurs in EBV-associated nasopharyngeal carcinomas. It is known that expression of a single viral immediate-early protein, BZLF1, is sufficient to initiate the switch from latent to lytic infection in B cells. Cellular regulation of BZLF1 transcription is therefore thought to play a key role in regulating the stringency of viral latency. Here we show that, unexpectedly, expression of another viral immediate-early protein, BRLF1, can disrupt viral latency in an epithelial cell-specific fashion. Therefore, the mechanisms leading to disruption of EBV latency appear to be cell-type specific.

Functional and physical interactions between the Epstein-Barr virus (EBV) proteins BZLF1 and BMRF1: Effects on EBV transcription and lytic replication

The Epstein-Barr virus (EBV) proteins BZLF1 and BMRF1 are both essential for lytic EBV replication. BZLF1 is a transcriptional activator which binds directly to the lytic origin of replication (oriLyt) and plays a critical role in the disruption of viral latency. The BMRF1 protein is required for viral polymerase processivity. Here we demonstrate that the BMRF1 gene product functions as a transcriptional activator and has direct (as well as indirect) interactions with the BZLF1 gene product. The BMRF1 gene product activates an essential oriLyt promoter, BHLF1, but does not activate two other early EBV promoters (BMRF1 and BHRF1). Direct interaction between the BMRF1 and BZLF1 gene products requires the first 45 amino acids of BMRF1 and the bZip domain of BZLF1. The effect of the BZLF1-BMRF1 interaction on early EBV transcription is complex and is promoter specific. The oriLyt BHLF1 promoter is activated by either the BZLF1 or BMRF1 gene product alone and is further activated by the combination of the BZLF1 and BMRF1 gene products. Enhanced activation of BHLF1 transcription by the BMRF1-BZLF1 combination does not require direct interaction between these proteins. In contrast, BZLF1-induced activation of the BMRF1 promoter is inhibited in the presence of the BMRF1 gene product. A point mutation in the BZLF1 protein (amino acid 200), which prevents in vitro interaction with the BMRF1 protein but which does not reduce BZLF1 transactivator function, allows the BZLF1 protein to activate the BMRF1 promoter equally well in the presence or absence of the BMRF1 gene product. Therefore, direct interaction between the BZLF1 and BMRF1 proteins may inhibit BZLF1-induced transcription of the BMRF1 promoter. BZLF1 mutated at amino acid 200 is as efficient as wild-type BZLF1 in promoting replication of an oriLyt plasmid. However, this mutation reduces the ability of BZLF1 to induce lytic replication of the endogenous viral genome in D98/HE-R-1 cells. Our results indicate that functional and physical interactions between the BMRF1 and BZLF1 proteins may modulate the efficiency of lytic EBV infection. The BMRF1 gene product clearly has a transcriptional, as well as replicative, role during lytic EBV infection.

Transcriptional analysis of Epstein-Barr virus gene expression in EBV-positive gastric carcinoma: unique viral latency in the tumour cells

Although case-oriented evidence for an association of Epstein-Barr virus (EBV) with gastric carcinoma has been accumulating recently, the interaction(s) between EBV and gastric epithelial cells is/are largely unknown. In this study, we examined seven EBV-positive gastric carcinoma tissues for viral gene expression at the mRNA level, from which studies on the EBV oncogenicity in human epithelial cells will benefit. Reverse transcription-PCR analysis showed that all seven EBV-positive tumour tissues constitutively expressed EBV nuclear antigen (EBNA) 1 mRNA, but not EBNA2 mRNA. The EBNA transcription was initiated from one of three EBNA promoters, Qp: by contrast, both Cp and Wp were silent, thus resulting in the lack of EBNA2 mRNA. Latent membrane protein (LMP) 2A mRNA was detected in three of seven cases; however, neither LMP1 nor LMP2B mRNA was detected in any of the tumours tested. Transcripts from the BamHI-A region of the viral genome were detectable in all cases. BZLF1 mRNA and the product, an immediate-early gene for EBV replication, was not expressed in any of them, thereby suggesting that the tumour cells carried EBV genomes in a tightly latent form. These findings further extended our previous data regarding EBV latency in gastric carcinoma cells at the protein level, and have affirmed that the programme of viral gene expression in the tumour more closely resembles 'latency I' represented by Burkitt's lymphoma than 'latency II' represented by the majority of nasopharyngeal carcinomas.

The extracellular domain of the Epstein-Barr virus BZLF2 protein binds the HLA-DR beta chain and inhibits antigen presentation

The Epstein-Barr virus BZLF2 gene encodes a glycoprotein that associates with gH and gL and facilitates the infection of B lymphocytes. In order to determine whether the BZLF2 protein recognizes a B-cell-specific surface antigen, a soluble protein containing the extracellular portion of the BZLF2 protein linked to the Fc portion of human immunoglobulin G1 (BZLF2.Fc) was expressed from mammalian cells. BZLF2.Fc was used in an expression cloning system and found to bind to a beta-chain allele of the HLA-DR locus of the class II major histocompatibility complex (MHC). Analysis of amino- and carboxy-terminal deletion mutants of the BZLF2.Fc protein indicated that the first 90 amino acids of BZLF2.Fc are not required for HLA-DR beta-chain recognition. Site-directed mutagenesis of an HLA-DR beta-chain cDNA and subsequent immunoprecipitation of expressed mutant beta-chain proteins using BZLF2.Fc indicated that the beta1 domain, which participates in the formation of peptide binding pockets, is required for BZLF2.Fc recognition. The addition of BZLF2.Fc to sensitized peripheral blood mononuclear cells in vitro abolished their proliferative response to antigen and inhibited cytokine-dependent cytotoxic T-cell generation in mixed lymphocyte cultures. Flow-cytometric analysis of Akata cells induced to express late Epstein-Barr virus antigens indicated that expression of BZLF2 did not result in reduced surface expression levels of MHC class II. The ability of BZLF2.Fc to bind to the HLA-DR beta chain suggests that the BZLF2 protein may interact with MHC class II on the surfaces of B cells.

Transcription of the Epstein-Barr virus nuclear antigen 1 (EBNA1) gene occurs before induction of the BCR2 (Cp) EBNA gene promoter during the initial stages of infection in B cells

The purpose of this study was to gain insights into the regulation of Epstein-Barr virus (EBV) gene transcription during the establishment of viral latency in B cells. During the early stages of EBV infection in B lymphocytes, transcription of six viral nuclear antigens (EBNAs) is initiated from an early promoter (Wp). This is followed by a switch of promoter usage to an upstream promoter, Cp, whose activity is autoregulated by both EBNA1 and EBNA2. Previously it was demonstrated that infection of primary B cells with EBNA2-negative (EBNA2-) EBNA4-mutant (EBNA4mut) virus resulted only in the expression of mutant EBNA4 protein and failure to express the other EBNA gene products (C. Rooney H. G. Howe, S. H. Speck, and G. Miller, J. Virol. 63:1531-1539, 1989). We extended this research to demonstrate that Wp-to-Cp switching did not occur upon infection of primary B cells with an EBNA2- EBNA4mut virus (M. Woisetschlaeger, X. W. Jin, C. N. Yandara, L. A. Furmanski, J. L. Strominger, and S. H. Speck, Proc. Natl. Acad. Sci. USA 88:3942-3946, 1991). Further characterization of this phenomenon led to the identification of an EBNA2-dependent enhancer upstream of Cp. On the basis of these data, a model was proposed in which initial transcription from Wp gives rise to the expression of EBNA2 and EBNA4, and then transcription is upregulated from Cp via the EBNA2- dependent enhancer (Woisetschlaeger et al., as noted above). Implicit in this model is that transcription of the EBNA1 and EBNA3a to -3c genes is dependent on the switch from Wp to Cp, since primary cells infected with EBNA2- EBNA4mut virus fail to switch and also fail to express these viral antigens. Here we critically evaluate this model and demonstrate, in contrast to the predictions of the model, that transcription of both the EBNA1 and EBNA2 genes precedes activation of Cp. Furthermore, the level of EBNA1 gene transcription was strongly reduced in primary B cells infected with EBNA2- EBNA4mut virus compared with that of cells infected with wild-type virus. Switching to Cp, as well as EBNA1 gene transcription, was observed upon infection of EBV-negative Burkitt's lymphoma (BL) cell lines with EBNA2- EBNA4mut virus, thus establishing a correlation between early EBNA1 gene transcription and upregulation of transcription initiation from Cp. However, in EBV-negative BL cell lines infected with EBNA2- EBNA4mut virus, transcription of the EBNA1 gene at early time points postinfection initiated from Qp, the EBNA1 gene promoter active in group I BL cells (B. C. Schaefer, J. L. Strominger, and S. H. Speck, Proc. Natl. Acad. Sci. USA 92:10565-10569, 1995), rather than from Wp. The data support a model in which EBNA1 plays an important role in the cascade of events leading to successful switching from Wp to Cp and subsequent immortalization of the infected B cell.

Characterization of the ZI domains in the Epstein-Barr virus BZLF1 gene promoter: role in phorbol ester induction

Induction of the Epstein-Barr virus lytic cycle is mediated through the immediate-early BZLF1 gene and the coordinately regulated BRLF1 gene. The BZLF1 gene product, Zta, transactivates its own promoter, as well as the promoters of a number of lytic genes, thereby initiating a cascade of viral gene expression. Previous work identified four related elements (ZIA, ZIB, ZIC, and ZID) and a cyclic AMP response element binding-AP-1 element (ZII) that are involved in the induction of the BZLF1 promoter (Zp) by the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA) (E. Flemington and S. H. Speck, J. Virol. 64:1217-1226, 1990). Here we report a detailed characterization of TPA induction mediated by the ZI domains. Mutation of individual ZI domains within the context of the intact promoter significantly diminished TPA induction. Cloning of individual ZI domains upstream of a minimal promoter demonstrated that the ZIA, ZIC, and ZID domains, but not the ZIB domain, are TPA responsive. Furthermore, cloning of the ZII domain downstream of the ZI domains significantly augmented TPA induction. The critical regions within the ZIA and ZIC elements involved in binding of cellular factors were identified by using methylation interference and electrophoretic mobility shift analyses of ZI domain mutants. Four specific complexes were observed with the ZIA and ZID domains, all of which could be specifically competed for by either the ZIA or ZID domain. Methylation interference analyses of bound complexes revealed the presence of two overlapping binding sites for cellular factors in the ZIA domain, and functional studies provided evidence that both of these sites are involved in TPA induction. Functional analyses of the ZIC domain revealed that the 5' region of this domain is largely responsible for mediating TPA induction. Binding data correlated well with functional activity and revealed that the ZIC domain binds only a subset of the cellular factors that bind to the ZIA and ZID domains. Analysis of factor binding to the ZIB domain revealed only a single shifted complex, which correlated with the most slowly migrating complex observed with the ZIA and ZID domains. These data provide a direct demonstration of TPA induction mediated by the ZIA, ZIC, and ZID domains and also provide the first evidence that the ZI domains exhibit distinct functional characteristics.

Comparing transcriptional activation and autostimulation by ZEBRA and ZEBRA/c-Fos chimeras

The lytic cycle of Epstein-Barr virus (EBV) can be activated by transfection of the gene for ZEBRA, a viral basic-zipper (bZip) transcriptional activator. ZEBRA and cellular AP-1 bZip activators, such as c-Fos, have homologous DNA-binding domains, and their DNA-binding specificities overlap. Moreover, EBV latency can also be disrupted by phorbol esters, which act, in part, through AP-1 activators. It is not known whether ZEBRA and AP-1 factors play equivalent roles in the initial stages of reactivation. Here the contribution of ZEBRA's basic DNA recognition domain to disruption of latency was analyzed by comparing ZEBRA with chimeric mutants in which the DNA recognition domain of ZEBRA was replaced with the analogous domain of c-Fos. Chimeric ZEBRA/c-Fos proteins overexpressed in Escherichia coli bound DNA with the specificity of c-Fos; they bound a heptamer AP-1 site and an octamer TPA response element (TRE). ZEBRA bound the AP-1 site and an array of ZEBRA response elements (ZREs). In assays with reporter genes, both ZEBRA and ZEBRA/c-Fos chimeric mutants activated transcription from Zp, a promoter of the ZEBRA gene (BZLF1) that contains the TRE and multiple ZREs. However, despite their capacity to activate reporters bearing Zp, neither ZEBRA nor the c-Fos chimeras activated transcription from Zp in the context of the intact latent viral genome. In contrast, ZEBRA but not ZEBRA/c-Fos chimeras activated Rp, a second viral promoter that controls ZEBRA expression. Hence, transcriptional autostimulation by transfected ZEBRA occurred preferentially at Rp. Both ZEBRA and the ZEBRA/c-Fos chimeras activated transcription from reporters with multimerized AP-1 sites. However, in the context of the virus, only ZEBRA activated the promoters of two early lytic cycle genes, BMRF1 and BMLF1, that contain an AP-1 site. Thus, overexpression of an activator that recognized AP-1 and TRE sites was not sufficient to activate EBV early lytic cycle genes.

Antisense oligodeoxynucleotides against the BZLF1 transcript inhibit induction of productive Epstein-Barr virus replication

Expression of the Epstein-Barr virus (EBV) BZLF1 gene product, ZEBRA, in latently infected cells is sufficient to induce the viral lytic cycle. The use of oligodeoxynucleotides complementary to the BZLF1 transcript was studied to inhibit this induction of productive viral replication. For this purpose, we employed oligodeoxynucleotides complementary to the translation initiation codons and their flanking sequences. Incubation of Akata cells with the 25-mer phosphodiester (PO)- or phosphorothioate (PS)-antisense oligodeoxynucleotides for 3 h before stimulation with anti-immunoglobulin G antibodies (anti-IgG) partially inhibited the anti-IgG-mediated induction of ZEBRA synthesis. Both the PO- and PS-antisense oligodeoxynucleotide treatments also suppressed the productive EBV replication (as measured by linear DNA production) in a dose-dependent manner, with much greater efficiency than did PO and PS-oligodeoxynucleotides with sense, reverse or random sequences of the same length. Another 20-mer antisense oligodeoxynucleotide complementary to sequences downstream of the translation initiation codons showed a similar inhibitory effect on EBV replication. However, the inhibition was considerably lower when the cells were treated with oligodeoxynucleotides complementary to sequences upstream of the start codons. These results indicate that BZLF1 antisense oligodeoxynucleotides inhibit the viral activation in a sequence-specific fashion. In the virus-producer cell line P3HR-1, the same PS-antisense oligodeoxynucleotides also partially suppressed the spontaneous viral replication after 6-10 days, substantially more than the PS-random oligodeoxynucleotides. Inhibition of BZLF1 appears to be sufficient to suppress the induction of EBV replication.

The Epstein-Barr virus gene BHRF1, a homologue of the cellular oncogene Bcl-2, inhibits apoptosis induced by gamma radiation and chemotherapeutic drugs

Analysis of apoptosis, active and controllable cell death, has demonstrated that the size of a cell population can be regulated by changes in the cell death rate as well as in the rates of proliferation and differentiation. Factors which alter the rate of cell death, such as expression of the proto-oncogene bcl-2, can therefore directly affect the number of cells within a population. Bcl-2 has been shown to suppress apoptosis in response to a variety of stimuli and to act as a complementary survival signal for the random acquisition of other oncogenic mutations, such as deregulated c-myc. The Epstein Barr virus (EBV) gene BHRF1 was the first of a family of bcl-2 homologues now being identified. BHRF1 and bcl-2 share 25% primary amino acid sequence homology. Here we show that gamma radiation and several cytotoxic anticancer agents induce apoptosis in Burkitt's lymphoma (BL) cell lines, as has been found in several other systems. Using gene transfection studies we have also shown that expression of either BHRF1 or bcl-2 in BL cell lines significantly suppresses apoptosis in response to a variety of anticancer treatment. This has confirmed that BHRF1 is functionally homologous to bcl-2 in B-cells and suggests that BHRF1 may act to prevent apoptosis during EBV infection, maximising virus particle production, as has been suggested for other human and insect viral genes. Suppression of chemotherapeutic drug induced cell death by bcl-2 and BHRF1 as demonstrated in this cell system, results in resistance to a variety of different agents and may represent an alternative mechanism by which multidrug resistance arises during chemotherapy.

The ecology and pathology of Epstein-Barr virus

Epstein-Barr virus achieves its ubiquitous and uniform epidemiological distribution by a dual strategy of latency to guarantee lifelong persistence and intermittent replication to guarantee transmission. These two functions appear to dictate residence in different cell types: latency in B lymphocytes and replication in epithelial cells. Both of these cell compartments are potential sites for EBV-associated malignancies.

Identification and characterization of ZIIBC, a complex formed by cellular factors and the ZII site of the Epstein-Barr virus BZLF1 promoter

The transition from latency to lytic Epstein-Barr virus replication is dependent on the Epstein-Barr virus BZLF1 gene product. Genetic and biochemical attempts to link cellular second-messenger signaling pathways that trigger this transition with the subsequent viral gene cascade have identified functional elements within the BZLF1 promoter (Zp) that appear to bind undefined cellular transcription factors. One of these previously identified sites, ZII, has homology to consensus AP-1 and CREB binding sites, implying a role for these factors in the inductive process. We have identified and characterized ZIIBC, a ZII site binding complex that is distinct from the factors previously proposed to bind this site. Active ZIIBC was found to be present in both uninduced and chemically induced cell extracts at approximately equivalent concentrations. Analysis of the DNA sequence requirements for the binding of ZIIBC to the ZII site shows that sequences homologous to AP-1 and CREB consensus sites are necessary but not sufficient for complex formation. Although the components of ZIIBC that directly contact DNA were found to be of the same molecular masses (26 and 36 kDa) in both uninduced and chemically induced cell extracts, a slight mobility difference between DNA-protein complexes formed by these two types of extracts is observable and indicates that ZIIBC is directly affected by chemical induction. The effects of ZIIBC binding to the ZII site on expression from Zp were evaluated, and they suggest that ZIIBC plays a critical role in the regulation of Zp expression.

DNA damage in human B cells can induce apoptosis, proceeding from G1/S when p53 is transactivation competent and G2/M when it is transactivation defective

Cisplatin treatment of Epstein-Barr virus-immortalized human B lymphoblastoid cell lines (LCLs) results in p53-mediated apoptosis which occurs largely in a population of cells at the G1/S boundary of the cell cycle. Cell cycle progression appears to be required for this apoptosis because arresting cells earlier in G1 inhibited apoptosis despite the accumulation of p53. Overexpression of wild-type p53 also induces apoptosis in an LCL. Therefore six mutant genes derived from Burkitt's lymphoma (BL) cells were assayed for their ability to induce apoptosis when similarly overexpressed. The same genes were analysed in transient transfection assays for their ability to transactivate appropriate reporter plasmids. A correlation between the ability of p53 to transactivate and induce apoptosis was revealed. The only mutant capable of transactivation also induced apoptosis. Further analysis of the BL lines in which p53 had been characterized showed that whereas some lines were essentially resistant to cisplatin, three were rapidly induced to undergo apoptosis. All three have a single p53 allele encoding a mutant which is incapable of transactivation or (for two tested) mediating apoptosis when expressed in an LCL. Cell cycle analysis revealed that this apparently p53-independent apoptosis did not follow G1 arrest but in fact occurred largely in cells distributed in the G2/M phase of the cell cycle. These data suggest the existence of a second checkpoint in the G2 or M phase which, in the absence of a functional p53, is the primary point of entry into the apoptosis programme following DNA damage.

Immune regulation in Epstein-Barr virus-associated diseases

Epstein-Barr virus (EBV) is a member of the human herpesvirus family and, like many other herpesviruses, maintains a lifelong latent association with B lymphocytes and a permissive association with stratified epithelium in the oropharynx. Clinical manifestations of primary EBV infection range from acute infectious mononucleosis to an asymptomatic persistent infection. EBV is also associated with a number of malignancies in humans. This review discusses features of the biology of the virus, both in cell culture systems and in the natural host, before turning to the role of the immune system in controlling EBV infection in healthy individuals and in individuals with EBV-associated diseases. Cytotoxic T cells that recognize virally determined epitopes on infected cells make up the major effector arm and control the persistent infection. In contrast, the options for immune control of EBV-associated malignancies are more restricted. Not only is antigen expression restricted to a single nuclear antigen, EBNA1, but also these tumor cells are unable to process EBV latent antigens, presumably because of a transcriptional defect in antigen-processing genes (such as TAP1 and TAP2). The likelihood of producing a vaccine capable of controlling the acute viral infection and EBV-associated malignancies is also discussed.

The Epstein-Barr virus BamHI F promoter is an early lytic promoter: lack of correlation with EBNA 1 gene transcription in group 1 Burkitt's lymphoma cell lines

The Epstein-Barr virus BamHI F promoter (Fp) was previously identified as the putative EBNA 1 gene promoter in group 1 Burkitt's lymphoma (BL) cell lines. Fp has also been shown to be activated in Epstein-Barr virus-positive B-cell lines following induction of the viral productive cycle (A. L. Lear, M. Rowe, M. G. Kurilla, S. Lee, S. Henderson, E. Kieff, and A. B. Rickinson, J. Virol. 66:7461-7468, 1992). Here we demonstrate that Fp is exclusively a lytic promoter which was incorrectly identified as the EBNA 1 gene promoter in group 1 BL cell lines. It is shown that while Fp activity was observed in two group 1 BL cell lines, it could not be detected in a third group 1 BL cell line. Furthermore, the level of Fp activity detected in both group 1 and group 3 cell lines appeared to correlate only with the level of spontaneous lytic activity. Induction of the lytic cycle in group 1 or group 3 BL cell lines resulted in a dramatic increase in Fp-initiated transcripts but no detectable increase in EBNA 1 transcripts. Anti-immunoglobulin induction of the lytic cycle in the Akata group 1 BL cell line revealed that induction of Fp activity was detectable by 2 to 4 h after induction of the lytic cycle and was dependent on de novo protein synthesis. In addition, Fp reporter constructs transiently transfected into group 1 BL cell lines exhibited activity which was independent of the Fp initiation site, TATAA box, or other upstream sequences. The sequences required for efficient reporter gene activity mapped to a region ca. 210 bp downstream of the Fp cap site. Furthermore, Northern (RNA) blot analyses indicated that there are two Fp-initiated lytic transcripts between 9 and 15 kb in size, neither of which correspond to the known EBNA 1 transcripts present in group 1 BL cell lines.

The Epstein-Barr virus 3.5-kilobase latent membrane protein 1 mRNA initiates from a TATA-Less promoter within the first terminal repeat

The complete 5' sequence of the Epstein-Barr virus 3.5-kb latent membrane protein 1 (LMP1) mRNA, expressed in nasopharyngeal carcinoma, has been determined. The transcript initiates from heterogeneous start sites within the first terminal repeat (TR) of the viral genome. This region is TATA-less, consistent with heterogeneous starting, but contains multiple GC-rich elements which potentially interact with the Sp1 transcription factor. Expression of the 3.5-kb mRNA was consistently detected in nasopharyngeal carcinoma samples and in additional cell types, including a Burkitt's lymphoma. This is the first identification of an Epstein-Barr virus mRNA containing TR sequence and the first report of the ability of the TR to function as a transcriptional promoter.

YY1 binds to and regulates cis-acting negative elements in the Epstein-Barr virus BZLF1 promoter

A 48-bp cis-acting negative element in the Epstein-Barr virus BZLF1 gene P1 promoter has been described previously. By DNase I footprinting experiments, two regions were identified as the protein-binding sites (previously designated site I and site II). In this report, the cellular transcription factor YY1 has been identified as a protein which binds to both of these elements, now designated ZIVA and ZIVB. Both ZIVA and ZIVB conferred cis-acting negative regulation on an enhancerless simian virus 40 promoter. In cotransfection experiments, overexpression of YY1 caused further repression of the enhancerless simian virus 40 promoter containing either the ZIVA or ZIVB element. Cotransfection of a plasmid expressing antisense to YY1 increased the expression of the heterologous promoter containing ZIVA but not ZIVB. In similar experiments carried out with the P1 promoter, overexpression of YY1 caused downregulation of P1 whereas antisense RNA to YY1 caused a slight increase in expression. Analyses of various P1 mutant constructions revealed additional YY1 sites downstream of ZIVB. Overexpression of YY1 also caused downregulation of a P1 mutant with no apparent YY1-binding sites. TPA treatment of Raji cells caused a temporal loss of YY1-binding activity but had no effect on the intracellular levels of YY1 protein. Serum induction of quiescent B cells also caused loss of YY1 binding to the ZIVB site, which was found to be a weak serum response element. In contrast, anti-immunoglobulin G treatment of Akata cells had no effect on either the YY1-binding activity or protein levels. The binding of YY1 to the cis-acting negative elements in infected B cells may play a pivotal role in the maintenance of Epstein-Barr virus latency.

The Zif268 cellular transcription factor activates expression of the Epstein-Barr virus immediate-early BRLF1 promoter

The Epstein-Barr virus immediate-early protein BZLF1 mediates the switch from latent to lytic infection. BZLF1 transcription can be derived from either the BZLF1 promoter or the BRLF1 promoter (Rp). Productive viral infection of EBV-infected B cells can be induced by 12-O-tetradecanoylphorbol-13-acetate (TPA) treatment, as well as cross-linking of surface immunoglobulin with antiimmunoglobulin antibody. Both TPA and antiimmunoglobulin antibody are known to activate expression of the cellular transcription factor Zif268 in B cells. In this study, we have examined the regulation of BZLF1 transcription by Zif268. We show that Rp (but not the BZLF1 promoter) is activated by Zif268. Bacterially synthesized Zif268 binds strongly to an upstream sequence in the Rp promoter (located from -131 to -123 relative to the start site) and more weakly to a proximal sequence (-49 to -40). Zif268 activation of Rp requires these two Zif268 binding sites. TPA treatment of B cells induces the expression of Zif268 protein, which binds to Rp. Furthermore, TPA activation of Rp requires the upstream Zif268 site. These findings indicate that Zif268 can activate a critical Epstein-Barr virus immediate-early promoter and, therefore, may play a key role in the regulation of viral latency.

Replication of Epstein-Barr virus oriLyt: lack of a dedicated virally encoded origin-binding protein and dependence on Zta in cotransfection assays

Using a transient replication assay in which cosmid DNAs were cotransfected into Vero cells, we had previously demonstrated that oriLyt replication required six Epstein-Barr virus (EBV)-encoded replication genes. No oriLyt origin-binding protein was identified in this study, but oriLyt replication in the cotransfection assay was also dependent on the three lytic cycle transactivators Zta, Rta, and Mta and an activity encoded by the EBV Sal/I F fragment. We have now used expression plasmids for the six known replication proteins to further examine the question of the requirement for an oriLyt origin-binding protein. The activity in Sal/I-F was shown to be encoded by BKRF3. The predicted product of this open reading frame is an enzyme, uracyl DNA glycosylase, not an origin-binding protein, and is dispensable for replication in assays using expression plasmids. BBLF2, which is positionally related to the gene for the herpes simplex virus (HSV) UL9 origin-binding protein, was confirmed to be expressed as a spliced transcript with BBLF3 and not as an independent product. Examination of the requirement for the EBV transactivators revealed that Rta, while contributing to replication efficiency, was dispensable. Mta could be substituted by HSV IE63, and in complementation experiments with HSV replication genes, Mta was no longer required for replication of EBV oriLyt, suggesting that the contribution of Mta to replication may be indirect. Zta continued to be required for detectable oriLyt replication both with the EBV replication proteins and in the complementation assays with HSV replication proteins. We conclude that EBV does not encode an equivalent of HSV UL9 and that Zta is the sole virally encoded protein serving an essential origin-binding function.

Contribution of Epstein-Barr virus to development of malignant lymphoma of the thyroid

Epstein-Barr virus (EBV)-related mRNA, their products and apoptosis were investigated in 32 cases of malignant lymphoma of the thyroid (MLT) and 30 cases of Hashimoto's thyroiditis (HT) by in situ hybridization, immunohistochemistry and nick end labeling method on routinely processed tissue sections. In MLT, EBV-encoded small RNA (EBER) were detected in three cases, consisting of a follicular, predominantly large cell type (FL), a diffuse, large cell type (DL) and a large cell, immunoblastic type (IBL). In EBER-positive cases, IBL that was positive for T cell marker, exhibited neither BamHl H Left Frame 1 (BHLF1) transcript, EBV-encoded latent membrane protein (LMP) nor BamHl Z Left Frame 1 (BZLF1) gene product (ZEBRA), whereas both BHLF1 and ZEBRA were found in a small portion of the tumor cells in the FL and DL that expressed B cell marker and LMP. Apoptotic cells were observed in only a few lymphocytes in HT, and in a few non-neoplastic lymphocytes and various numbers of lymphoma cells in MLT. The apoptotic cell ratio of MLT tended to be higher in lower grade lymphomas. These results suggest that EBV may participate in the malignant transformation from HT to MLT.

Epstein-Barr virus efficiently immortalizes human B cells without neutralizing the function of p53

Epstein-Barr virus (EBV) efficiently converts resting human B cells into actively cycling, immortal, lymphoblastoid cell lines (LCLs). Here we show that LCLs expressing the full complement of latent viral genes are very sensitive to DNA-damaging agents such as cisplatin. The response includes a rapid accumulation of the tumour suppressor protein p53 and induction of the cellular genes mdm2 and WAF1/p21. Although the levels of Bcl2 protein and Bax mRNA appear unaltered by the activation of p53, within 24 h the majority of cells undergo apoptosis. Over-expression of wild-type p53 in an LCL also resulted in apoptosis; this was preceded by the dephosphorylation of the retinoblastoma gene product, pRb. Primary resting B cells showed no response to cisplatin and even after drug treatment, p53 remained undetectable. However, after infection with EBV, p53 gene expression was induced to a similar level to that found in mitogen-activated B cells. When the physiologically activated primary B cells were exposed to cisplatin, although p53 accumulated as in LCLs, the outcome was growth-arrest rather than gross cell death. We conclude that, in contrast to the transformation of fibroblasts by adenovirus, SV40 or HPV, when B cells become activated and immortalized by EBV they are sensitized to the p53-mediated damage response. When the resulting LCLs are treated with genotoxic agents such as cisplatin, they are unable to arrest like normal cells because they are driven to proliferate by EBV and consequently undergo apoptosis.

Epstein-Barr virus lytic replication is controlled by posttranscriptional negative regulation of BZLF1

Regulation of the immediate-early gene BZLF1 is assumed to play a key role in triggering the lytic replication of Epstein-Barr virus (EBV). The expression of BZLF1 is regulated on multiple levels, including control of transcription by several positive and negative cis-acting elements as well as posttranslational modifications and protein-protein interactions. Localization of BZLF1 on one strand of the genome and the latent EBNA1 transcription unit on the complementary strand suggests a regulatory mechanism via hybridization of antisense RNA. With a plasmid encoding a defective BZLF1 RNA, which could not be translated, we were able to induce expression of endogenous BZLF1 gene product Zta and other proteins of the lytic cycle. Our data show for the first time that latent replication is stabilized by negative regulation of an immediate-early gene of the lytic cycle by a posttranscriptional mechanism. This might be a common theme of herpes simplex virus and EBV latency.

Linear epitopes of the replication-activator protein of Epstein-Barr virus recognised by specific serum IgG in nasopharyngeal carcinoma

The linear antigenic epitopes of the Epstein-Barr virus replication activator protein (ZEBRA), recognised by specific serum IgG in nasopharyngeal carcinoma (NPC), were determined. This was achieved by synthesizing the entire amino acid sequence of ZEBRA as a set of 29, 22-residue peptides with an overlap of 14 amino acids. The ZEBRA peptides were tested in enzyme-linked immunosorbent assay (ELISA) for IgG binding in sera from 37 selected NPC patients who had IgG antibodies to the native ZEBRA protein. The most immunogenic epitope was peptide 1 at the amino-terminal end with 36 of the sera reactive against it. Further analysis of peptide 1, using the multipin peptide-scanning technique, defined a 10-amino-acid sequence FTPDPYQVPF, which was strongly bound by IgG. Two other regions of ZEBRA were also identified as immunodominant IgG epitopes, namely peptide 11 (amino acids 82-103) and peptide 19/20 (amino acids 146-175) with 8-13 of the NPC sera reactive against the peptides. The number of peptides reactive with individual NPC serum varies from 1 to 6 or more and there is some correlation between a greater number of peptide (at least 4) bound and a higher (at least 1:40) titre of serum IgA to viral capsid antigen. The immunodominant ZEBRA peptide 1 could be utilised in IgG ELISA for the detection of NPC.

Integral membrane protein 2 of Epstein-Barr virus regulates reactivation from latency through dominant negative effects on protein-tyrosine kinases

An Epstein-Barr virus-encoded protein, LMP2, blocks the effects of surface immunoglobulin (slg) cross-linking on calcium mobilization and on lytic reactivation of EBV in latently infected and growth-transformed primary human B lymphocytes. In wild-type EBV-transformed cells, LMP2 is constitutively tyrosine phosphorylated and is associated with Lyn and Syk protein-tyrosine kinases (PTKs). Baseline Lyn PTK activity is substantially reduced, and slg cross-linking fails to activate Lyn, Syk, Pl3-K, PLC gamma 2, Vav, Shc, and MAPK. Syk, Pl3-K, PLC gamma 2, and Vav are constitutively tyrosine phosphorylated, and their tyrosine phosphorylation does not change following slg cross-linking. In contrast, cross-linking slg on cells transformed by LMP2 null mutant EBV recombinants triggers the same protein tyrosine kinase cascade as in noninfected B lymphocytes. These data are consistent with a model in which LMP2 is a constitutive dominant negative modulator of slg receptor signaling through its effects on Lyn, Syk, or regulators of these kinases.

Epstein-Barr virus latent messages with shuffled leader exons: remnants of circumgenomic transcription?

The EBNA transcription unit which is active in Epstein-Barr virus-immortalized latently infected B lymphocytes covers approximately 60% of the 172-kb genome. Since the genome exists as a circular double-stranded DNA molecule in latently infected cells, it is conceivable that complete copies are made during transcription. Rather than attempt to detect gigantic RNA molecules directly, we used RNA-PCR to detect incorporation of leader exons into mRNA in a shuffled order. The downstream U leader exon was detected spliced upstream of the internal repeat leader exons W1 and W2 in the polyadenylated RNA fraction of spontaneous lymphoblastoid cell lines, restricted phenotype BL cell lines Wanyanyi and Wewak2, and in B95-8, Raji, and Akata cells. Quantitative competitive RNA-PCR showed that the ratio of U exon-containing EBNA1 messages to U exon-shuffled leader messages was approximately 10:1, with large variation from cell line to cell line, and was not affected by induction of the lytic cycle in B95-8, Raji, or Akata cells. Messages with shuffled exons contained only the C2W1 alternative splice, which does not produce an initiator AUG methionine codon for EBNA4 gene expression. The results provide evidence for long-range exon skipping and imply that genome-length transcripts may occur and participate in viral gene expression in latency.

Differential Epstein-Barr virus gene expression in B-cell subsets recovered from lymphomas in SCID mice after transplantation of human peripheral blood lymphocytes

We have analyzed the human B-cell tumors that arise spontaneously in SCID mice who have been given transplants of peripheral blood lymphocytes from Epstein-Barr virus (EBV)-seropositive donors to determine if patterns of EBV gene expression are correlated with phenotypic changes in the tumor B cells. Tumor cells were separated into two B-cell subsets by cell sorting on the basis of differential coexpression of membrane CD23 and CD38. One subset showed intermediate levels of CD23 and CD38 expression (CD23intCD38int), while a second subset had low-level CD23 but high-level CD38 expression (CD23loCD38hi). The CD23intCD38int cells had a high proliferative index and secreted little immunoglobulin in vitro; the CD23loCD38hi cells had a low proliferative index and high-level immunoglobulin secretion. We next analyzed the sorted cells for viral transcripts associated with latency (EBNA-1, EBNA-2, and LMP-1) or lytic cycle replication (ZEBRA and gp350 envelope protein). Only latent cycle transcripts were found in CD23intCD38int cells, whereas lytic cycle transcripts and transforming virus were present in the CD23loCD38hi cells. Finally, we generated short-term cell lines from the sorted CD23intCD38int cells and transferred these cells to SCID recipients. The resulting secondary tumors were predominantly CD23loCD38hi, suggesting that the CD23intCD38int lymphoblastoid cells are precursors to the well-differentiated, plasmacytoid CD23loCD38hi cells. These observations are discussed in the context of a three-step model for EBV-associated lymphomagenesis in humans.

Transcription of the E2F-1 gene is rendered cell cycle dependent by E2F DNA-binding sites within its promoter

The cell cycle-regulatory transcription factor E2F-1 is regulated by interactions with proteins such as the retinoblastoma gene product and by cell cycle-dependent alterations in E2F-1 mRNA abundance. To better understand this latter phenomenon, we have isolated the human E2F-1 promoter. The human E2F-1 promoter, fused to a luciferase cDNA, gave rise to cell cycle-dependent luciferase activity upon transfection into mammalian cells in a manner which paralleled previously reported changes in E2F-1 mRNA abundance. The E2F-1 promoter contains four potential E2F-binding sites organized as two imperfect palindromes. Gel shift and transactivation studies suggested that these sites can bind to E2F in vitro and in vivo. Mutation of the two E2F palindromes abolished the cell cycle dependence of the E2F-1 promoter. Thus, E2F-1 appears to be regulated at the level of transcription, and this regulation is due, at least in part, to binding of one or more E2F family members to the E2F-1 promoter.

Transcription of the E2F-1 gene is rendered cell cycle dependent by E2F DNA-binding sites within its promoter

The cell cycle-regulatory transcription factor E2F-1 is regulated by interactions with proteins such as the retinoblastoma gene product and by cell cycle-dependent alterations in E2F-1 mRNA abundance. To better understand this latter phenomenon, we have isolated the human E2F-1 promoter. The human E2F-1 promoter, fused to a luciferase cDNA, gave rise to cell cycle-dependent luciferase activity upon transfection into mammalian cells in a manner which paralleled previously reported changes in E2F-1 mRNA abundance. The E2F-1 promoter contains four potential E2F-binding sites organized as two imperfect palindromes. Gel shift and transactivation studies suggested that these sites can bind to E2F in vitro and in vivo. Mutation of the two E2F palindromes abolished the cell cycle dependence of the E2F-1 promoter. Thus, E2F-1 appears to be regulated at the level of transcription, and this regulation is due, at least in part, to binding of one or more E2F family members to the E2F-1 promoter.

Enhancement of infectivity of hantavirus in cell culture by centrifugation

Centrifugation was introduced during virus adsorption to Vero E6 cells to improve the infectivity of hantavirus. Centrifugal adsorption of a stock solution of Hantaan virus strain 76-118 to a monolayer of Vero E6 cells enhanced virus infectivity depending on the centrifugation time and the centrifugal force. The maximum level of infectivity (3.1 x 10(6) FFU/ml) was enhanced after a 2 h centrifugation at 671 x g, which was almost 9-times higher than that of conventional adsorption of the virus at 37 degrees C for 1 h. Vero E6 cells were inoculated with a new hantavirus strain, KI-91-40, isolated with a low infectious titer (400 FFU/ml) from an urban rat and adsorbed by centrifugation. A higher virus titer was detected sooner compared to when using conventional adsorption. To analyze the mechanism of the enhancement, the centrifugation was carried out before and after virus adsorption. The infectivity was reduced when Vero E6 monolayers were centrifuged before virus inoculation. When the centrifugation proceeded after inoculation, the infectivity was almost equal to that without centrifugation. The infectivity was only enhanced when centrifugation was carried out during inoculation. These results indicate that centrifugation promotes a very early event of infection, probably attachment of the virus to cells.

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

During cultivation of the Epstein-Barr virus (EBV)-positive Burkitt's lymphoma (BL) line Akata, it was noted that EBV DNA is lost from some of the cells. Isolation of EBV-positive and EBV-negative clones with the same origin made it possible to examine the effects of EBV in BL cells. The results indicate that malignant phenotypes of BL, such as growth in low serum, anchorage-independent growth in soft agar, and tumorigenicity in nude mice, are dependent on the presence of EBV genomes and underline the oncogenic function of EBV in human cancer.

Determination of the role for CD21 during Epstein-Barr virus infection of B-lymphoblastoid cells

Epstein-Barr virus (EBV), a herpesvirus with oncogenic potential, is camouflaged with glycoprotein 350/220, which mimics the human ligand C3dg and thereby binds to and exploits complement receptor type 2 (CR2; CD21), the EBV receptor. It has not been possible to determine the role of CR2 during postbinding events of viral infection because all B lymphocytes express endogenous CR2, precluding an informative study of receptor mutants. We have overcome this obstacle through creation of a novel experimental system based on molecular dissection of the ligand-binding domains of human CR2 and murine CR2. Our results demonstrate first, that two discontinuous amino acid substitutions within the ligand-binding domain of murine CR2 render it capable of mediating EBV infection of human B-lymphoblastoid cells, and second, that the specific role of CR2 during EBV infection is to capture virions at the cell surface, after which cofactors not associated with CR2 mediate postbinding events. These are the first studies to be described in which a cell that is normally susceptible to viral infection can be manipulated so as to direct entry of virions via recombinant or endogenous receptors.

DNA-binding-defective mutants of the Epstein-Barr virus lytic switch activator Zta transactivate with altered specificities

The Epstein-Barr virus BRLF1 and BZLF1 genes are the first viral genes transcribed upon induction of the viral lytic cycle. The protein products of both genes (referred to here as Rta and Zta, respectively) activate expression of other viral genes, thereby initiating the lytic cascade. Among the viral antigens expressed upon induction of the lytic cycle, however, Zta is unique in its ability to disrupt viral latency; expression of the BZLF1 gene is both necessary and sufficient for triggering the viral lytic cascade. We have previously shown that Zta can activate its own promoter (Zp), through binding to two Zta recognition sequences (ZIIIA and ZIIIB). Here we describe mutant Zta proteins that do not bind DNA (referred to as Zta DNA-binding mutants [Zdbm]) but retain the ability to transactivate Zp. Consistent with the inability of these mutants to bind DNA, transactivation of Zp by Zdbm is not dependent on the Zta recognition sequences. Instead, transactivation by Zdbm is dependent upon promoter elements that bind cellular factors. An examination of other viral and cellular promoters identified promoters that are weakly responsive or unresponsive to Zdbm. An analysis of a panel of artificial promoters containing one copy of various promoter elements demonstrated a specificity for Zdbm activation that is distinct from that of Zta. These results suggest that non-DNA-binding forms of some transactivators retain the ability to transactivate specific target promoters without direct binding to DNA.

DNA-binding-defective mutants of the Epstein-Barr virus lytic switch activator Zta transactivate with altered specificities

The Epstein-Barr virus BRLF1 and BZLF1 genes are the first viral genes transcribed upon induction of the viral lytic cycle. The protein products of both genes (referred to here as Rta and Zta, respectively) activate expression of other viral genes, thereby initiating the lytic cascade. Among the viral antigens expressed upon induction of the lytic cycle, however, Zta is unique in its ability to disrupt viral latency; expression of the BZLF1 gene is both necessary and sufficient for triggering the viral lytic cascade. We have previously shown that Zta can activate its own promoter (Zp), through binding to two Zta recognition sequences (ZIIIA and ZIIIB). Here we describe mutant Zta proteins that do not bind DNA (referred to as Zta DNA-binding mutants [Zdbm]) but retain the ability to transactivate Zp. Consistent with the inability of these mutants to bind DNA, transactivation of Zp by Zdbm is not dependent on the Zta recognition sequences. Instead, transactivation by Zdbm is dependent upon promoter elements that bind cellular factors. An examination of other viral and cellular promoters identified promoters that are weakly responsive or unresponsive to Zdbm. An analysis of a panel of artificial promoters containing one copy of various promoter elements demonstrated a specificity for Zdbm activation that is distinct from that of Zta. These results suggest that non-DNA-binding forms of some transactivators retain the ability to transactivate specific target promoters without direct binding to DNA.

RAZ, an Epstein-Barr virus transdominant repressor that modulates the viral reactivation mechanism

Epstein-Barr virus (EBV) is associated with the development of several types of human cancers and is an important cause of lymphomas in immunocompromised hosts. Expression of the EBV BZLF1 immediate-early gene product (Z) triggers disruption of latency in EBV-infected cells. Z is a member of the b-Zip family of proteins and binds to AP-1-like sites in early viral promoters. Here we show that a viral RNA related to Z, in which there is replacement of the transactivation domain of Z by fusion through alternate splicing with a portion of another EBV transactivator, BRLF1 (R), can repress Z function. This differentially spliced mRNA is predicted to express a novel chimeric protein which we call RAZ for R and Z. RAZ retains the dimerization and DNA-binding domains of Z but loses its transactivation domain. We show that in vitro the RAZ protein acts transdominantly to repress transactivation of early promoters by Z. Repression is produced by dimerization of RAZ with Z resulting in RAZ:Z heterodimers that can no longer bind to Z-binding sites despite retention of the DNA-binding domains in both proteins. Deletion of the R domain of RAZ restores the ability of the truncated RAZ homodimers and RAZ:Z heterodimers to bind to DNA. A biologic effect of RAZ was shown by cotransfection of latently infected Raji cells with Z and RAZ expression clones; RAZ diminished viral reactivation induced by Z, as indicated by amount of early replicative antigens (EA-D) detected. The RAZ protein presents a model for transcriptional control unique among the herpesvirus and distinct from analogous viral and cellular repressors. RAZ, by limiting the availability of Z protein, is likely to modulate EBV reactivation.

Different activation of Epstein-Barr virus immediate-early and early genes in Burkitt lymphoma cells and lymphoblastoid cell lines

Specific expression of the Epstein-Barr virus (EBV) immediate-early and early gene products Zta, Rta, I'ta, and MSta by a recombinant vaccinia virus system allowed us to analyze the first steps in the induction of the lytic cycle in EBV-infected Burkitt lymphoma (BL) cells and lymphoblastoid cell lines (LCLs). Significant differences in the induction of early genes were found between these cell types: whereas in BL cells the trans activator Zta was found to induce key steps of the early lytic cycle, only minor activities of Zta were noted in LCLs. Contrary to Zta, the trans activator Rta was found to be highly effective in LCLs. These observations suggest that Rta may play an important role in the activation of the early lytic cycle in LCLs, although it cannot be activated by Zta. The latter may be a reason for the lower tendency of LCLs to switch into the lytic cycle compared with BL cells or differentiated epithelial cells.

An integral membrane protein (LMP2) blocks reactivation of Epstein-Barr virus from latency following surface immunoglobulin crosslinking

The role of latent membrane protein 2 (LMP2) in Epstein-Barr virus (EBV) infection was evaluated by using latently infected primary B lymphocytes that had been growth transformed by wild-type or specifically mutated EBV recombinants. LMP2 null mutant recombinant EBV-infected cells were similar to normal B lymphocytes in their rapid increase in intracellular free calcium after surface immunoglobulin crosslinking. These cells also became more permissive for lytic EBV replication. In sharp contrast, wild-type control infected cells had little or no increase in intracellular free calcium or in permissivity for EBV replication. The block to surface immunoglobulin crosslinking-induced permissivity in cells expressing wild-type LMP2 could be bypassed by raising intracellular free calcium levels with an ionophore and by activating protein kinase C with phorbol 12-myristate 13-acetate. LMP2A, not LMP2B, mediates this effect on calcium mobilization. Genetic and biochemical data are consistent with these effects being due to the interaction of the LMP2A N-terminal cytoplasmic domain with B lymphocyte src family tyrosine kinases.

EBV-immortalized isogenic human B-cell clones exhibit differences in DNA-protein complex formation on the BZLF1 and BRLF1 promoter regions among latent, lytic and TPA-activated cell lines

The reactivation of Epstein-Barr virus from latency requires the transcriptional induction of the viral encoded lytic cycle initiator gene, BZLF1, and a concomitant switch from OriP to OriLyt directed viral DNA replication. To investigate the role of host cell factors in these events, a series of EBV-immortalized clonal lymphoblastoid cell lines (LCL) were derived from the spontaneous outgrowth of peripheral blood lymphocytes from a single EBV-seropositive individual. We show that the state of virus activation among this family of isogenic clonal LCL differs, suggesting that each B-cell clone expresses a different complement of cellular factors that influence the state of viral activation. As a first step in the identification of factors involved in EBV reactivation, nuclear extracts were prepared from tightly latent, spontaneously replicating and latent LCL treated with 12-O-tetradecanoyl-phorbol-13-acetate (TPA) and sodium butyrate. The extracts were used in gel mobility shift analyses to compare DNA-protein complex formation among a series of target DNA sequences, including OriLyt and promoter sequences from BZLF1 and BRLF1. The DNA-protein complex patterns were reproducible and indistinguishable among extracts obtained from the latent LCL, but differed from those observed in extracts obtained from the spontaneously replicating LCL, particularly in regard to the binding of a CREB protein to the BZLF1 promoter. Moreover, extracts prepared from LCL treated with TPA to induce virus reactivation resulted in the formation of complexes that differed from those prepared from the spontaneously replicating LCL. Taken together, these data suggest that B-cell factors govern the state of viral activation and that EBV may be reactivated by more than one pathway.

Epstein-Barr virus induces fragmentation of chromosomal DNA during lytic infection

Pulsed-field agarose gel electrophoresis showed that fragmentation of chromosomal DNA in Raji cells was induced by infection with the P3HR-1 strain of Epstein-Barr virus (EBV). S1 nuclease treatment of the agarose plugs containing cells suggested that the majority of DNA fragments did not contain single-strand gaps. Chromosomal DNA fragmentation was inhibited by cycloheximide, indicating that protein synthesis was required for DNA fragmentation. Phosphonoacetic acid, an inhibitor of EBV DNA polymerase, did not inhibit fragmentation of chromosomal DNA. These findings suggest that EBV-specific early proteins participate in fragmentation of chromosomal DNA. Chromosomal DNA of P3HR-1 cells was also fragmented by treatment with n-butyrate plus 12-O-tetradecanoylphorbol-13-acetate (TPA), which induced activation of latent EBV genome following viral replication. In addition, fragmentation of DNA preceded cell death during lytic infection. These results suggest that fragmentation of chromosomal DNA is generally induced during EBV replication and probably contributes to the cytopathic effect of EBV. The role of DNA fragmentation in death of infected cells is discussed in relation to apoptosis.

Frequent occurrence of in vivo clonal expansion of CD4- CD8- T cells bearing T cell receptor alpha beta chains in adult humans

We have previously reported 2 cases of healthy men showing in vivo monoclonal expansion of mature CD4- CD8- alpha beta T cells. In the present study, an additional 3 adults were found to exhibit such an expansion, among a total 464 adult donors studied. These 5 individuals were otherwise physiologically normal, with no history of severe illness and autoimmune disease at the time of examination. To investigate the mechanisms of the clonal expansion, further characterization of the clonal cells was attempted. No apparent preference for usage of the T cell receptor beta chain variable region was observed in the clonal T cells. These clonal T cells showed lectin-dependent or redirected antibody-dependent cell-mediated cytotoxicities, whereas they could not lyse autologous lymphoblastoid cell lines. Failure of Fas antigen expression was not observed for any of these clones. These results suggest that clonal expansion of CD4- CD8- alpha beta T cells frequently occurs in the periphery without any T cell abnormalities.

Epstein-Barr virus-coded BHRF1 protein, a viral homologue of Bcl-2, protects human B cells from programmed cell death

Epstein-Barr virus, a human herpesvirus that persists within the B-lymphoid system, can enhance the survival potential of latently infected B cells in vitro through up-regulation of the cellular survival protein Bcl-2. The possibility that an analogous effect is operative in lytically infected cells was suggested by the observation of distant sequence homology between an Epstein-Barr virus-coded early lytic cycle protein, BHRF1, and Bcl-2. Here we show by gene transfer that BHRF1 resembles Bcl-2 both in its subcellular localization and in its capacity to enhance B-cell survival. Thus confocal microscopic analysis of cells acutely cotransfected with BHRF1 and Bcl-2 expression vectors revealed substantial colocalization of the two proteins in the cytoplasm. In subsequent experiments, stable BHRF1 gene transfectants of Burkitt lymphoma cells paralleled Bcl-2 transfectants in their enhanced survival under conditions that induce cell death by apoptosis. Despite their limited sequence conservation, therefore, the two proteins appear to be functionally homologous. We suggest that BHRF1 provides an alternative, Bcl-2-independent, means of enhancing B-cell survival that may operate during the virus lytic cycle.

Analysis of the BZLF1 promoter of Epstein-Barr virus: identification of an anti-immunoglobulin response sequence

The induction of the viral lytic cycle in latently Epstein-Barr virus (EBV)-infected B cells is initiated by activation of the BZLF1 gene, whose expression is sufficient to disrupt EBV latency, suggesting that BZLF1 acts as the switch to change from a latent to a lytic replicative cycle. In the present studies, a series of deletion plasmids encompassing positions bp -552 to +12 of the BZLF1 promoter were constructed and tested for their response to anti-immunoglobulin (anti-Ig), an inducer of the viral lytic cycle, upon transfection into EBV-negative and -positive lymphoid cells. The promoter consisted of three functionally distinct regions. Region I (bp -552 to -221) had a negative influence on promoter activity; its deletion made the promoter highly responsive to anti-Ig. Region II (bp -203 to -177) was important for conferring responsiveness to anti-Ig. The response to anti-Ig did not require the presence of the EBV genome or EBV gene products. This sequence also enhanced expression of the chloramphenicol acetyltransferase (cat) gene from the simian virus 40 promoter in response to anti-Ig, even when inserted downstream of the cat gene. Region III (-134 to -116) was a positive element that was transactivated by the BZLF1 gene product.

The lytic transition of Epstein-Barr virus is imitated by recombinant B-cells

Lytic transition of Epstein-Barr virus (EBV) is initiated by distinct immediate early regulators of the viral cycle, in synchronization to temporary, permissive conditions during host cell differentiation. We developed eukaryotic vectors suitable to imitate the processes involved in lytic transition in cell culture systems. Two stable B cell lines were established: R59Z activator cells were used to induce lytic EBV expression in a constitutive manner by the production of the BZLF 1 trans-activator (Zta). R7-57 reporter cells, on the other hand, signaled induced activity of the lytic origin of EBV replication (ori Lyt). Different modes, like chemical induction, lytic superinfection with EBV and single gene trans-activation converted the recombinant ori Lyt element in R7-57 reporter cells. BZLF 1, transiently expressed in R7-57 reporter cells, was the only EBV trans-activator found, sufficient in inducing the viral lytic cycle. Basing on these experiments, trans-cellular activation of EBV was tested by cocultivation of BZLF 1-expressing R59Z activator cells with the R7-57 reporter line. No lytic effect on the reporter cells could be measured, neither by cocultivation of activator cells nor by coincubation of BZLF 1-containing cell lysates. Latency breaking activity, however, was transferred from activator to reporter cells when active, exogenous virus was added. The cell system described in these experiments provides a tool for the detection of EBV reactivation and demonstrates the potential of the lytic regulatory gene BZLF 1.

The DNA-binding domain of two bZIP transcription factors, the Epstein-Barr virus switch gene product EB1 and Jun, is a bipartite nuclear targeting sequence

The Epstein-Barr virus BZLF1 gene product EB1 (also called ZEBRA and Zta), is a transcription factor belonging to the bZIP (basic domain leucine zipper) family of nuclear proteins. Translocation to the nucleus of EB1 (J. Becker, U. Leser, M. Marschall, A. Langford, W. Jilg, H. Gelderblom, P. Reichart, and H. Wolf, Proc. Natl. Acad. Sci. USA 88:8332-8336, 1991) and of two other bZIP proteins, c-Jun and c-Fos (P. Roux, J.-M. Blanchard, A. Fernandez, N. Lamb, P. Jeanteur, and M. Piechaczyk, Cell 63:341-351, 1990), has been shown to be subject to regulation. We show here that for both EB1 and Jun the nuclear targeting signals (NTS) in the proteins' primary sequences are two clusters of positively charged amino acids. These clusters, called BRA and BRB, are necessary and sufficient to direct beta-galactosidase to the nuclear compartment and act as a bipartite NTS. They are conserved among all the bZIP proteins, and although they are not identical, they probably share the same function. Site-directed mutagenesis studies made on these basic clusters suggest that they also act as a bipartite NTS in the EB1 protein. Our results also demonstrate that in EB1 and Jun, these bipartite NTS are superimposed with bipartite DNA-binding domains, since BRA and BRB are required in vitro for direct and specific contact between these proteins and their DNA-binding sites.

The Epstein-Barr virus immediate-early promoter BRLF1 can be activated by the cellular Sp1 transcription factor

Disruption of viral latency in Epstein-Barr virus-infected cells is mediated through the activation of the BZLF1 (Z) immediate-early gene product. The Z protein can be derived from either of two promoters: the BZLF1 promoter, which directs transcription of a 1.0-kb mRNA encoding the Z gene product alone, or the upstream BRLF1 promoter, which directs transcription of a 2.8-kb bicistronic mRNA encoding the BRLF1 and BZLF1 immediate-early proteins. In this study we have examined the regulation of the BRLF1 promoter by viral and cellular factors. We found that the BRLF1 promoter is autoregulated by the BRLF1 transactivator through a nonbinding mechanism. We show that the BRLF1 (but not the BZLF1) promoter is highly responsive to the Sp1 transcription factor. Sp1 activation of the BRLF1 promoter is mediated through a consensus Sp1-binding site located from -39 to -44 (relative to the mRNA start site). We demonstrate that the BRLF1 promoter has high constitutive activity in C-33 cells (an epithelial cell line) and that the proximal Sp1-binding site is required for this activity. Despite the ubiquitous presence of Sp1 in many cell types, we found that the BRLF1 promoter has essentially no activity in lymphoid cell lines, suggesting that factors other than Sp1 may negatively regulate the BRLF1 promoter in these cells. Our findings demonstrate that the two potential promoters directing BZLF1 transcription are differentially regulated and that Sp1 can activate the BRLF1 promoter but not the BZLF1 promoter.

Differential splicing of Epstein-Barr virus immediate-early RNA

A variant form of splicing of RNA crossing the Epstein-Barr virus (EBV) BZLF1 gene was observed in the late productive cycle of EBV. This splice omits the middle exon of BZLF1 and joins the outer two exons of BZLF1 in frame, but the shortened form of BZLF1 protein (Z delta) could not be detected in natural EBV infections.

Partial elimination of Epstein-Barr virus plasmids from Burkitt's lymphoma cells by transfecting the BZLF1 gene

Epstein-Barr virus (EBV) nonproducer Raji cells stably maintain approximately 45 copies of the EBV genome per cell, depending on the presence of the EBV-determined nuclear antigen 1 (EBNA-1) protein. We found that transfection of the EBV BZLF1 gene causes the disappearance of EBNA proteins on Western blots (immunoblots). On the basis of these results, we attempted to eliminate EBV plasmids in Raji cells by transfecting a BZLF1 plasmid. Among 33 clones that were cotransfected with a BZLF1 plasmid and a hygromycin B resistance plasmid and selected resistant for hygromycin B, 24 clones had decreased numbers of EBV plasmids, as revealed by the decrease in the intensity of the EBV band on Southern blots compared with that of nontransfected Raji cells.