Differential expression of Epstein Barr viral transcripts for two proteins (TP1 and LMP) in lymphocyte and epithelial cells

Epstein-barr virus regulates c-MYC, apoptosis, and tumorigenicity in Burkitt lymphoma

Loss of the Epstein-Barr virus (EBV) genome from Akata Burkitt lymphoma (BL) cells is coincident with a loss of malignant phenotype, despite the fact that Akata and other EBV-positive BL cells express a restricted set of EBV gene products (type I latency) that are not known to overtly affect cell growth. Here we demonstrate that reestablishment of type I latency in EBV-negative Akata cells restores tumorigenicity and that tumorigenic potential correlates with an increased resistance to apoptosis under growth-limiting conditions. The antiapoptotic effect of EBV was associated with a higher level of Bcl-2 expression and an EBV-dependent decrease in steady-state levels of c-MYC protein. Although the EBV EBNA-1 protein is expressed in all EBV-associated tumors and is reported to have oncogenic potential, enforced expression of EBNA-1 alone in EBV-negative Akata cells failed to restore tumorigenicity or EBV-dependent down-regulation of c-MYC. These data provide direct evidence that EBV contributes to the tumorigenic potential of Burkitt lymphoma and suggest a novel model whereby a restricted latency program of EBV promotes B-cell survival, and thus virus persistence within an immune host, by selectively targeting the expression of c-MYC.

Mechanisms that regulate Epstein-Barr virus EBNA-1 gene transcription during restricted latency are conserved among lymphocryptoviruses of Old World primates

Epstein-Barr virus (EBV), the only known human lymphocryptovirus (LCV), displays a remarkable degree of genetic and biologic identity to LCVs that infect Old World primates. Within their natural hosts, infection by these viruses recapitulates many key aspects of EBV infection, including the establishment of long-term latency within B lymphocytes, and is therefore a potentially valuable animal model of EBV infection. However, it is unclear whether these LCVs have adopted or maintained the same mechanisms used by EBV to express essential viral proteins, such as EBNA-1, in the face of cell-mediated repression of EBV gene expression that occurs upon establishment of the asymptomatic carrier state. To address this issue, we determined whether the endogenous LCVs of baboon (Cercopithecine herpesvirus 12) and rhesus macaque (Cercopithecine herpesvirus 15) have the functional equivalent of the EBV promoter Qp, which mediates exclusive expression of EBNA-1 during the restricted programs of EBV latency associated with the carrier state. Our results indicate that (i) both the baboon and rhesus macaque LCVs have a genomic locus that is highly homologous to the EBV Qp region, (ii) key cis-regulatory elements of Qp are conserved in these LCV genomes and compose promoters that are functionally indistinguishable from EBV Qp, and (iii) EBNA-1 transcripts identical in structure to EBV Qp-specific EBNA-1 mRNAs are present in nonhuman LCV-infected cells, demonstrating that these Qp homologs are indeed utilized as alternative EBNA-1 promoters. These observations indicate that the molecular mechanisms which regulate EBV gene expression during restricted latency have been conserved among the LCVs. The contribution of these mechanisms to viral persistence in vivo can now be experimentally tested in nonhuman primate models of LCV infection.

Interferon-independent and -induced regulation of Epstein-Barr virus EBNA-1 gene transcription in Burkitt lymphoma

Replication of the Epstein-Barr virus (EBV) genome within latently infected cells is dependent on the EBV EBNA-1 protein. The objective of this study was to identify transcriptional regulatory proteins that mediate EBNA-1 expression via the viral promoter Qp, which is active in EBV-associated tumors such as Burkitt lymphoma and nasopharyngeal carcinoma. Results of a yeast one-hybrid screen suggested that a subset of the interferon regulatory factor (IRF) family may regulate EBNA-1 transcription by targeting an essential cis-regulatory element of Qp, QRE-2. Further investigation indicated that the transcriptional activator IRF-1 and the closely related IRF-2, a repressor of interferon-induced gene expression, are both capable of activating Qp. However, the major QRE-2-specific binding activity detected within extracts of Burkitt lymphoma cells was attributed to IRF-2, suggesting that interferon-independent activation of Qp is largely mediated by IRF-2 in these cells. We observed no effect of gamma interferon on Qp activity in transfection assays, whereas we observed a moderate but significant repression of Qp activity in response to alpha interferon, possibly mediated by either the interferon consensus sequence binding protein or IRF-7, a novel alpha interferon-inducible factor identified in this study. Since expression of IRF-1 and IRF-2 is increased in response to interferons, the Qp activity observed in the presence of interferon likely represented an equilibrium between IRF factors that activate and those that repress gene expression in response to interferon. Thus, by usurping both IRF-1 and its transcriptional antagonist IRF-2 to activate Qp, EBV has evolved not only a mechanism to constitutively express EBNA-1 but also one which may sustain EBNA-1 expression in the face of the antiviral effects of interferon.

The Epstein-Barr virus EBNA-1 promoter Qp requires an initiator-like element

Expression of the Epstein-Barr virus (EBV) EBNA-1 protein within EBV-positive tumor cells and subpopulations of latently infected B lymphocytes in vivo is mediated by the promoter Qp. Previous studies have established that Qp is a TATA-less promoter whose activation requires only proximal regulatory elements and that it is negatively autoregulated through two EBNA-1 binding sites downstream of the transcription initiation sites. The objective of this study was to better define the properties of an essential positive regulatory element (QRE-2) adjacent to a major transcription start site of Qp and to evaluate the contributions of other potential regulatory elements proximal to the Qp start site. Using DNA affinity purification and UV cross-linking, we have identified the QRE-2-binding protein as a single polypeptide of approximately 40 kDa. The DNA-binding properties of this protein are clearly distinct from those of the TATA-binding protein, suggesting that in the absence of a TATA box, QRE-2 may function as an initiator element to direct assembly of TFIID near the transcription start site. Mutational analysis of potential regulatory elements, furthermore, indicated that the putative E2F binding sites within the EBNA-1 binding domain can exert a positive influence on Qp that is EBNA-1 independent, suggesting that these regulatory elements play an additional if not different role in Qp regulation than previously proposed. A model for the regulation of Qp consistent with the current and previous findings which provides for a simple but efficient mechanism of ensuring the EBNA-1 expression necessary to sustain long-term latency is presented.

Transcription start sites downstream of the Epstein-Barr virus (EBV) Fp promoter in early-passage Burkitt lymphoma cells define a fourth promoter for expression of the EBV EBNA-1 protein

In Epstein-Barr virus (EBV)-transformed B lymphoblastoid and many Burkitt lymphoma cell lines, the EBV EBNA-1 protein is one of six viral nuclear antigens expressed from a common transcription unit under the control of one of two promoters, Wp or Cp. In contrast, EBNA-1 is the only EBV nuclear antigen expressed in Burkitt and other EBV-positive tumors. We previously identified a promoter of EBNA-1 transcription, designated Fp, in early-passage Mutu Burkitt tumor cells, and this promoter is also active in long-term Mutu and Akata Burkitt cell lines which maintain the exclusive expression of EBNA-1 characteristic of the tumor. However, transcription initiation within Fp reporter gene plasmids in EBV-negative cells occurs at positions 100 to 200 bases downstream of the Fp start site in the BamHI-Q restriction fragment. Here we demonstrate that transcription initiation within newly established Burkitt lymphoma cell lines is consistent with the transcription initiation we observed in reporter plasmids. Furthermore, previous observations of transcription from Fp to generate EBNA-1 transcripts can be attributed to lytic-cycle gene expression. These data, in conjunction with our previous characterization of promoter regulatory elements, define a fourth EBNA-1 promoter, Qp, that is active in latently infected Burkitt tumor cells.

Antibodies to LMP2A/2B in EBV-carrying malignancies

Antibodies to the Epstein-Barr virus (EBV)-encoded membrane proteins, LMP2A and LMP2B, were assayed in 540 individuals, including 154 patients with nasopharyngeal carcinoma, 16 with African Burkitt's lymphoma, 113 with Hodgkin's disease, 14 with EBV-carrying gastric carcinoma, 14 with oral hairy leucoplakia (HIV+ patients), 37 with non-Hodgkin's lymphoma, 49 with tumours of the head/neck, 19 with infectious mononucleosis, 62 with chronic illnesses with EBV titres consistent with re-activations, and 62 healthy controls. A novel assay, mouse monoclonal enhanced indirect immunofluorescence assay (MIFA) was designed and used to test the sera for antibodies to the LMP2A and 2B proteins, expressed in human keratinocytes. Antibody to both LMP2A and LMP2B was strikingly specific to NPC. Virtually all (99 of 101) of the LMP2 antibody positive individuals were NPC patients, 95% of whom had antibodies that reacted both with the LMP2A- and LMP2B-transfected indicator cells, while the remaining 5% reacted only with the LMP2B expressing cells.

A subpopulation of normal B cells latently infected with Epstein-Barr virus resembles Burkitt lymphoma cells in expressing EBNA-1 but not EBNA-2 or LMP1

Using reverse transcription of whole cellular RNA and nested PCR, we have performed experiments mixing different proportions of Epstein-Barr virus (EBV)-carrying and EBV-negative cells. Based on the results, a method that detects viral transcripts for EBNA-1, EBNA-2, LMP1, and LMP2a from less than one positive cell among 10(5) negative cells was developed. With this method we have shown that the EBV DNA positive cells among small, high-density peripheral blood B-lymphocytes of normal healthy persons express EBNA-1-mRNA but not EBNA-2 or LMP1. A similar EBV expression pattern is found in type I Burkitt lymphoma cells. We suggest that the expression pattern in the lymphoma cells reflects the viral strategy in normal resting B cells and meets the requirements of latent persistence.

Expression of cyclin D2 in Epstein-Barr virus-positive Burkitt's lymphoma cell lines is related to methylation status of the gene

The cyclin D2 gene is not expressed in resting primary B lymphocytes or in group I Burkitt's lymphoma (BL) cell lines that retain the characteristics of authentic BL cells. Expression of cyclin D2 is induced in primary B lymphocytes following infection with Epstein-Barr virus (EBV) or transfection of the EBV genes EBNA-LP and EBNA-2. However, attempts to induce cyclin D2 expression in BL cell lines by the enforced expression of EBV genes were unsuccessful. Since the demethylation agent 5-azacytidine has been shown to modulate viral gene expression in BL cells, we explored the possibility that methylation plays a significant role in the control of cyclin D2 expression. We show that 5-azacytidine treatment of the Mutu CI 179 BL cell line led to expression of cyclin D2 RNA and that expression correlated with differences in the methylation status of a CCGG restriction enzyme site near the transcription initiation region of the cyclin D2 gene. Thus, methylation appears to play a direct role in the regulation of the cyclin D2 locus in BL.

Characterization of the Epstein-Barr virus Fp promoter

Expression of the Epstein-Barr virus nuclear antigen-1 (EBNA-1) protein is mediated by the virus Fp promoter in Burkitt lymphoma and nasopharyngeal carcinoma. This promoter is silent in latently infected B lymphoblastoid and most Burkitt lymphoma-derived cell lines in vitro, which utilize separate promoters approximately 50 kb upstream of Fp to express EBNA proteins. Fp-mediated activation of EBNA-1 expression is also activated upon induction of the virus replication cycle. We previously demonstrated that activation of Fp in Burkitt cells requires cis-regulatory elements downstream of the site of transcription initiation. We have now mapped two positive regulatory elements within the Fp promoter. One element contains two potential binding sites for the cellular transcription factor LBP-1 between +138 and +150. A second regulatory element was mapped between +177 and +192 and can be specifically bound in vitro by protein from nuclear extracts of Burkitt cells. Although this element overlaps two partial E2F binding sites and Fp reporter plasmids could be activated in trans by the adenovirus E1A protein in cotransfection experiments, mutational analysis and DNA binding studies suggest that these are unlikely to be functional E2F response elements within Fp. We also demonstrate that Fp-directed transcription initiates at multiple sites within both the genome and the Fp reporter plasmids. However, the principal site of transcription initiation within the genome is not utilized within reporter plasmids, in which the majority of transcripts initiate at multiple sites between +150 and +200. This finding suggests that additional elements may be necessary for Fp to function normally in these assays or that the context of Fp within the viral genome is critical to its regulation.

Epstein-Barr virus latency in blood mononuclear cells: analysis of viral gene transcription during primary infection and in the carrier state

Epstein-Barr virus (EBV) can display different forms of latent infection in B-cell lines in vitro; however, the types of infection normally established by the virus in vivo remain largely unexplored. Here we have approached this question by analyzing the types of viral RNAs present in mononuclear cells freshly isolated from the blood of 14 infectious mononucleosis patients undergoing primary EBV infection and 6 long-term virus carriers. Reverse transcription-PCR amplifications were carried out with a panel of oligonucleotide primers and probes which specifically detect (i) the EBER1 RNA common to all forms of latency, (ii) transcripts either from the Cp and Wp promoters generating all six nuclear antigen (EBNA1, -2, -3A, -3B, -3C, -LP) mRNAs or from the Fp promoter generating a uniquely spliced EBNA1 mRNA, (iii) the latent membrane protein (LMP1 and 2A) mRNAs, and (iv) the BZLF1 mRNA, an immediate-early marker of lytic cycle. Viral transcription in infectious mononucleosis mononuclear cells (and in the B-cell-enriched fraction) regularly included the full spectrum of latent RNAs seen during EBV-induced B-cell growth transformation in vitro, i.e., EBER1, Cp/Wp-initiated EBNA mRNAs, and LMP1/LMP2 mRNAs, in the absence of lytic BZLF1 transcripts. In addition, transcripts with the splice pattern of Fp-initiated EBNA1 mRNA, hitherto seen only in vivo in certain EBV-positive tumors, were frequently detected. In long-term virus carriers, the mononuclear cells were again positive for latent (EBER1) and negative for lytic (BZLF1) markers; Cp/Wp-initiated RNAs were not detected in these samples, but in several individuals it was possible to amplify both Fp-initiated EBNA1 mRNA and LMP2A mRNA signals. We suggest (i) that primary infection is associated with a transient virus-driven expansion of the infected B-cell pool through a program of virus gene expression like that seen in in vitro-transformed cells and (ii) that long-term virus carriage is associated with a switch from Cp/Wp to Fp usage and thus to a more restricted form of latent protein expression that may render the infected cells less susceptible to recognition by the virus-specific cytotoxic T-cell response.

Varicella-zoster virus DNA polymerase and major DNA-binding protein genes have overlapping divergent promoters

A detailed analysis of the transcriptionally divergent promoters of varicella-zoster virus (VZV) open reading frames (ORFs) 28 and 29, encoding the DNA polymerase and major DNA-binding proteins, respectively, was performed. We found that the 221-bp ORF 28-29 intergenic domain contains overlapping divergent promoters; these promoters have TATA boxes and cap sites arranged closely back-to-back, have highly concordant patterns of responsiveness to transactivation by VZV ORFs 4 and/or 62, and could not be separated without abolishing the effects that VZV trans activators imparted to them. Mutation of the ORF 28 TATA box rendered this promoter unresponsive to ORF 62 and the combination of ORFs 4 and 62 without altering ORF 29 promoter activity. Mutations of all potential ORF 29 TATA boxes collectively failed to abolish this promoter's responsiveness to either ORF 4 or ORF 62, suggesting a mechanism of gene regulation for ORF 29 that differs from that of ORF 28. These findings are concordant with the observation that both genes are expressed in productive infection, but only ORF 29 expression has been identified in latency.

HLA A2.1-restricted cytotoxic T cells recognizing a range of Epstein-Barr virus isolates through a defined epitope in latent membrane protein LMP2

Cytotoxic T-lymphocyte (CTL) responses induced by persistent Epstein-Barr virus (EBV) infection in normal B-lymphoid tissues could potentially be directed against EBV-positive malignancies if expression of the relevant viral target proteins is maintained in tumor cells. For malignancies such as nasopharyngeal carcinoma and Hodgkin's disease, this will require CTL targeting against the nuclear antigen EBNA1 or the latent membrane proteins LMP1 and LMP2. Here we analyze in detail a B95.8 EBV-reactivated CTL response which is specific for LMP2 and restricted through a common HLA allele, A2.1. We found that in vitro-reactivated CTL preparations from several A2.1-positive virus-immune donors contained detectable reactivity against A2.1-bearing target cells expressing either LMP2A or the smaller LMP2B protein from recombinant vaccinia virus vectors. Peptide sensitization experiments then mapped the A2.1-restricted response to a single epitope, the nonamer CLGGLLTMV (LMP2A residues 426 to 434), whose sequence accords well with the proposed peptide binding motif for A2.1. Most Caucasian and African virus isolates (whether of type 1 or type 2) were identical in sequence to B95.8 across this LMP2 epitope region, although 2 of 12 such isolates encoded a Leu-->Ile change at epitope position 6. In contrast, most Southeast Asian and New Guinean isolates (whether of type 1 or type 2) constituted a different virus group with a Cys-->Ser mutation at epitope position 1. CTLs raised against the B95.8-encoded epitope were nevertheless able to recognize these variant epitope sequences in the context of A2.1 whether they were provided exogenously as synthetic peptides or generated endogenously in B cells transformed with the variant viruses. A CTL response of this kind could have therapeutic potential in that it is directed against a protein expressed in many EBV-positive malignancies, is reactive across a range of virus isolates, and is restricted through a relatively common HLA allele.

Transcripts from the Epstein-Barr virus BamHI A fragment are detectable in all three forms of virus latency

An unexpected feature of the latency II form of Epstein-Barr virus (EBV) infection seen in the epithelial tumor nasopharyngeal carcinoma (NPC) is the presence of spliced polyadenylated RNAs encoded from the BamHI A fragment of the viral genome and running in the opposite orientation to several BamHI-A lytic cycle genes. The importance of these BamHI-A transcripts and the specificity of their association with NPC remain to be determined. In this study, we examined the extent to which such RNAs are present in other transcriptionally distinct forms of EBV latency seen in B cells. Two independent assays of BamHI-A transcription were employed: amplification across defined splice junctions in cDNAs, using the polymerase chain reaction, and in situ hybridization with a radiolabeled riboprobe specific for a putative open reading frame downstream of these splice junctions. Such methods, which easily detected BamHI-A RNAs in fresh NPC biopsies and transplantable NPC lines, also revealed consistent expression of these transcripts in all EBV-positive Burkitt's lymphoma cell lines displaying the highly restricted latency I form of infection (BamHI-F promoter usage) as well as in all EBV-transformed lymphoblastoid cell lines (LCLs) displaying the latency III form of infection (BamHI-C/W promoter usage). Expression in established LCLs, occurring irrespective of virus producer status, was not a consequence of continued in vitro passage; thus, appropriately spliced BamHI-A transcripts could be amplified from normal B cells within 1 day of their experimental infection in vitro, along with BamHI-C/W promoter-initiated but not BamHI-F promoter-initiated mRNAs. In situ hybridization both on Burkitt's lymphoma cell lines and on LCLs showed that essentially every cell contained BamHI-A transcripts, although at levels apparently lower than those observed in NPC. We conclude that expression of the BamHI-A RNAs is a consistent feature shared by all known forms of latent EBV infection.

Complex nature of the major viral polyadenylated transcripts in Epstein-Barr virus-associated tumors

The most abundant polyadenylated viral transcripts in the Epstein-Barr virus (EBV)-associated tumor nasopharyngeal carcinoma are a family (apparent sizes, 4.8, 5.2, 6.2, and 7.0 kb) of highly spliced cytoplasmic RNAs expressed from the BamHI-I and -A regions of the viral genome in an antisense direction with respect to several viral lytic functions encoded within the same region and concerned with the lytic cycle of the virus. We have called these complementary-strand transcripts. They are also expressed in B cells, including Burkitt's lymphoma and EBV-immortalized marmoset cell lines, and tumors generated in cottontop tamarins in response to EBV infection, but at a lower level. The complete structure of the major 4.8-kb RNAs (seven or eight exons) was determined in this study; the larger, but related, transcripts appear to be produced by differential splicing. The transcriptional promoter for the major complementary-strand transcripts, located in BamHI-I, contains several well-characterized transcriptional control elements (E2A, SP1, and AP1) and is functionally active in both B lymphocytes and epithelial cells. It appears to be a bifunctional viral promoter, as it also contains the initiation codon for a gene (BILF2) that encodes a glycoprotein that is expressed off the other strand. Splicing events create a number of small AUG-initiated open reading frames, one of which has homology to functionally significant regions of the EBV-encoded nuclear antigen 2 and to E2 (in papillomavirus). The complex nature of these transcripts and their potential role in the virus association with malignancy are considered.

The Epstein-Barr virus (EBV) nuclear antigen 1 BamHI F promoter is activated on entry of EBV-transformed B cells into the lytic cycle

In Epstein-Barr virus (EBV)-positive Burkitt's lymphoma cell lines exhibiting the latency I form of infection (i.e., EBV nuclear antigen 1 [EBNA1] positive in the absence of other latent proteins), the EBNA1 mRNA has a unique BamHI Q/U/K splice structure and is expressed from a novel promoter, Fp, located near the BamHI FQ boundary. This contrasts with the situation in EBV-transformed lymphoblastoid cell lines (LCLs) exhibiting the latency III form of infection (i.e., positive for all latent proteins), in which transcription from the upstream Cp or Wp promoters is the principal source of EBNA mRNAs. We carried out cDNA amplifications with oligonucleotide primer-probe combinations to determine whether Fp is ever active in an LCL environment. The results clearly showed that some LCLs express a Q/U/K-spliced EBNA1 mRNA in addition to the expected Cp/Wp-initiated transcripts; this seemed inconsistent with the concept of Cp/Wp and Fp as mutually exclusive promoters. Here we show that Fp is indeed silent in latency III cells but is activated at an early stage following the switch from latency III into the virus lytic cycle. Four pieces of evidence support this conclusion: (i) examples of coincident Cp/Wp and Fp usage in LCLs are restricted to those lines in which a small subpopulation of cells have spontaneously entered the lytic cycle; (ii) transcripts initiating from Fp can readily be demonstrated in spontaneously productive lines by S1 nuclease protection; (iii) the presence of Fp-initiated transcripts is not affected by acyclovir blockade of the late lytic cycle; and (iv) infection of latently infected LCLs with a recombinant vaccinia virus encoding the EBV immediate-early protein BZLF1, a transcriptional transactivator which normally initiates the lytic cycle, results in the appearance of the diagnostic Q/U/K-spliced transcripts.

The Epstein-Barr virus nuclear protein 1 promoter active in type I latency is autoregulated

The only member of the Epstein-Barr virus family of nuclear proteins (EBNAs) expressed during type I and type II latent infections is EBNA-1. This is in contrast to type III latency, during which all six nuclear proteins are expressed from a common transcription unit. The exclusive expression of EBNA-1 during type I and II latency is mediated through a recently identified promoter, Fp. The objective of this study was to characterize Fp in the Burkitt lymphoma cell background, where it is known to be differentially utilized. Using a short-term transfection assay and reporter gene plasmids containing Fp linked to the human growth hormone, we examined Fp activity in type I and type III latently infected and virus-negative Burkitt lymphoma cells. The data suggested that Fp is predominantly regulated through two distinct elements located between +24 and +270 relative to the transcription start site. One element positively mediates Fp activity, probably at the level of transcription, and acts in a virus-independent manner. The second element contains the EBNA-1 DNA binding domain III and negatively regulates Fp-directed gene expression in trans with EBNA-1 in type III as well as type I latency. Thus, we have identified a third function of EBNA-1, i.e., that of a repressor of gene expression, in addition to its known role in viral DNA replication and its ability to trans-activate gene expression. The overall activity of Fp in type I latently infected Burkitt cells was approximately sixfold lower than in virus-negative Burkitt cells, in which there is no autoregulation, suggesting that there is a fine balance between these two opposing regulatory elements during type I latency.

Epstein-Barr virus latent gene expression in uncultured peripheral blood lymphocytes

In this study of Epstein-Barr virus (EBV) latency, the polymerase chain reaction was used in modified form for amplification and detection of viral mRNA sequences in peripheral blood lymphocytes from healthy seropositive adults. Six known promoters for latent gene expression and eight known gene products were identified in in vitro-immortalized lymphocytes and in the cell lines established spontaneously from seropositive adults. We examined whether mRNA expression in uncultured B cells from four seropositive adults was the same as that which occurred in spontaneously established EBV-positive B-cell lines from the same individuals. A minimum of 17 polymerase chain reaction targets was required to circumscribe the known latent mRNA structures. Expression of the C promoter for the EBNA genes was detected in B-cell RNA from three of the four subjects. Transcripts initiated from the alternative W promoter for EBNA expression were not detected. The spliced transcripts detected in the B cells contained only the C2-to-W1 alternative splice, which was nonproductive for EBNA4 gene expression. None of the other EBNA open reading frames were detected spliced onto the 3' ends of the C promoter-initiated RNAs. Spliced RNA from the TP gene was detected in all four subjects. Expression of the TP gene was restricted to TP1 promoter-initiated RNAs, as no TP2 promoter-initiated transcripts were detected. Expression of RNA from the LMP gene was not detected. The F promoter which is active in the restricted expression latency that occurs in Burkitt's lymphoma cells was not detected being expressed in peripheral blood B cells. This pattern of latent gene expression is unique to uncultured B cells, indicating that there are profound differences between viral latent states in vitro and in situ and suggesting a central role for the TP gene in the latency of EBV.

Consistent transcription of the Epstein-Barr virus LMP2 gene in nasopharyngeal carcinoma

Two species of the Epstein-Barr virus-encoded latent membrane protein 2, LMP2A and LMP2B, are generated by alternative splicing, each species having a distinct first exon. LMP2 transcription in undifferentiated nasopharyngeal carcinoma (NPC), which is consistently associated with the Epstein-Barr virus, was investigated. Fifteen NPC specimens were analyzed by Northern (RNA) blot and RNA-based polymerase chain reaction; the LMP2A transcript was present in all specimens except one. In some specimens the LMP2B transcript was also detected. Sequence analysis of LMP2 cDNAs obtained from two NPC specimens revealed four mutations in exon 1 of the LMP2A transcript, which were present in both tumors and which resulted in nonconservative amino acid changes. These data suggest that the LMP2 expressed in NPC is distinct from that which is expressed in lymphoid cells.

Epstein-Barr virus latent gene transcription in nasopharyngeal carcinoma cells: coexpression of EBNA1, LMP1, and LMP2 transcripts

Epstein-Barr virus (EBV) genome-positive nasopharyngeal carcinomas (NPCs) regularly express the virus-coded nuclear antigen EBNA1, but not other EBNAs, and a subset of tumors also appear to be latent membrane protein LMP1 positive; the status of NPCs with respect to a second virus-coded latent membrane protein LMP2 is unknown. In the present work the EBV-NPC cell interaction has been analyzed at the RNA level with reverse transcription and polymerase chain reaction-based amplification to detect specific latent viral mRNAs. All four transplantable NPC cell lines studied and 17 of 18 fresh snap-frozen NPC biopsy specimens expressed an EBNA1 mRNA with a BamHI Q/U/K splice structure exactly like that recently identified in group I Burkitt's lymphoma (BL) cell lines and shown to be driven from a novel viral promoter, Fp. The BamHI Y3/U/K-spliced EBNA1 mRNA characteristic of virus-transformed B-lymphoblastoid cell lines (LCLs) was never found in NPCs. These same NPC biopsy specimens were then analyzed for evidence of the various LMP transcripts which are constitutively expressed in LCLs but down-regulated in BL cells. While only 3 of 18 tumors gave a clear LMP1 mRNA-specific signal after first-round amplification with either of two sets of polymerase chain reaction primers, the majority proved to be LMP1 mRNA positive after second-round amplification with nested primers. A rather similar pattern of results was obtained with respect to LMP2B mRNA expression, such transcripts being detectable only in a subset of tumors, and then at apparently low levels. In contrast, clear evidence of LMP2A mRNA expression was obtained in 17 of 17 fresh biopsies. The predominant form of EBV infection in NPCs, with coexpression of EBNA1 and LMP mRNAs, is therefore quite distinct from that seen in BL cells (in which EBNA1 is the only expressed mRNA) and in LCL cells (in which all six EBNA and three LMP transcripts are present). This third form of EBV latency may not be restricted to NPC but may have more general relevance in the context of EBV infection in vivo.

Transcription of the Epstein-Barr virus gene EBNA-1 from different promoters in nasopharyngeal carcinoma and B-lymphoblastoid cells

Transcriptional expression of the Epstein-Barr virus (EBV) genome has been shown to differ markedly between nasopharyngeal carcinoma (NPC) cells and latent B-cell lines, with a more limited pattern of gene expression seen in NPC. EBNA-1 is the only nuclear antigen so far detected in both NPC and Burkitt's lymphoma cells. We found previously that in a human NPC tumor passaged in nude mice, designated C15, the EBNA-1 mRNA contained a novel splice site in the BamHI Q region of EBV which had not previously been described for B-cell lines. This lies within a region of the EBV genome to which EBNA-1 binds. Here, we further characterize the 5' region of EBNA-1 transcripts and identify two splicing patterns in C15 cells; we show that they are derived from a common promoter region in the BamHI F region of the viral genome. We also demonstrate that this region can function to initiate transcription of the chloramphenicol acetyltransferase gene in epithelial cells and that the promoter region is only partially methylated at CpG sites in the tumor. In contrast, a B-lymphoblastoid cell line derived from C15 uses a conventional promoter in BamHI-C/W for expression of EBNA-1.