Comparative analysis of Epstein-Barr virus gene polymorphisms in nasal T/NK-cell lymphomas and normal nasal tissues: implications on virus strain selection in malignancy

Whether particular Epstein-Barr virus (EBV) strains are preferentially selected in malignant diseases remains controversial. Assessment of the importance of strain variation in the pathogenicity of EBV has been hampered principally by the lack of accurate data on the prevalence of virus variants in the normal population. To clarify this issue, a detailed comparative analysis of the EBV genomes contained in normal nasal and nasopharyngeal mucosal tissues and in nasal T/NK-cell lymphoma, which originates at these anatomic sites, was carried out by PCR amplification across the 30-bp deletion and the 33-bp repeat loci in the LMP1 gene and the type-specific polymorphic loci in the EBNA2 and EBNA3C genes and by sequence analysis of the 3' C-terminal region of the LMP1 gene. Whilst the majority of EBV strains in either normal or tumour tissues were type 1 viruses with similar numbers of LMP1 repeats, a marked predominance of LMP1 deletion (del-LMP1) over non-deleted/wild-type LMP1 (wt-LMP1) variants was observed in nasal T/NK-cell lymphoma. Although del-LMP1 variants were also prevalent in the normal carriers of our population, wt-LMP1 was detected at a significantly higher frequency in normal vs. tumour tissues (p = 0.036). More critically, wt-LMP1 variants were found frequently in mixed infection with del-LMP1 variants in the normal carriers. Sequence analysis identified 2 major del-LMP1 (and several wt-LMP1) variants containing signatory nucleotide changes in relation to the prototype B95-8 sequence in both normal and neoplastic nasal tissues. Together, our data provide strong evidence for a selection mechanism for del-LMP1 over the wt-LMP1 variants in tumours.     

Local mate competition, variable fecundity and information use in a parasitoid

A 30-bp deletion in the Epstein-Barr virus (EBV) latent membrane protein 1 (LMP1) gene has been reported in nasopharyngeal carcinoma and EBV-associated malignant lymphomas. Information on this deletion in EBV-associated gastric carcinoma (EBVaGC) is limited. The association of gastric carcinoma (GC) with EBV was examined by EBV-encoded RNA (EBER) in situ hybridization in 510 patients from Japan and 80 patients from Brazil. We studied the prevalence of 30-bp LMP1 gene deletion in EBVaGC in Japan (29 cases) and Brazil (four cases) in comparison with the corresponding EBER1-positive metastatic lesions in lymph nodes (10 cases) and EBV-infected reactive lymphocytes from dissected nonmetastatic lymph nodes (22 cases), microdissected non-neoplastic gastric mucosa of EBVaGC (five cases), and EBV-nonassociated GC (25 cases). We studied the status of the LMP1 gene by Southern blot hybridization of polymerase chain reaction products obtained after amplification with primers flanking the site of the deletion. We also performed EBV typing and LMP1 protein immunohistochemistry. EBV DNA was amplified by polymerase chain reaction in 30 of 33 EBVaGC cases, 8 of 10 metastatic carcinomas, 14 non-neoplastic tissues from 27 EBVaGC cases, and 12 of 25 non-EBV-associated GC cases with EBER1-positive lymphocytes. The 30-bp LMP1 gene deletion was observed in 23 of 26 (88.5%) cases of EBVaGC from Japan and two of four (50%) cases of Brazilian EBVaGC as compared with EBER1-positive reactive lymphocytes from 11 of 14 (78.6%) EBVaGC cases and 9 of 12 (75%) cases of non-EBV-associated GC. The variant type (the 30-bp deletion variant or nondeleted wild type) of LMP1 gene was the same among reactive lymphocytes, primary and secondary lesions of EBVaGC in all cases for which all three tissue types were studied (six of six). There was no correlation between the presence of the 30-bp deletion with depth of cancer invasion or presence of metastasis. Type A was detected in all available EBV-positive cases. The similar high incidence of 30-bp deletion in LMP1 gene in both carcinoma cells and reactive lymphocytes in EBVaGC cases suggests that this deletion may not be relevant to the pathogenesis of EBVaGC.     

Infection of leukaemic B lymphocytes by Epstein Barr virus

Epstein-Barr virus (EBV) infection of B lymphocytes in vitro gives rise to immortalized lymphoblastoid cell lines. Previous reports have shown that chronic lymphocytic leukaemia (CLL) cells, although infectable by EBV, are resistant to immortalization (1-4), although a small number of CLL cell lines have been reported (5-7). In the present study we have analysed early events occurring after EBV infection in 16 CLL samples. Out of 16 samples, 15 could be infected by the virus and expressed the full EB viral nuclear antigen (EBNA) complex but only one out of 16 expressed the latent membrane protein (LMP). The five CLLs in which we could investigate the presence of viral episomes showed circularized EBV by 16 hours after infection. The sequence of EBNA expression and genome circularization mirrored that seen in normal B cells, although genome amplification was not detected. The only CLL sample which expressed LMP after EBV infection was induced to proliferate for 2-3 weeks, but no cell line was established. Immortalized cell lines were obtained from three out of 16 samples tested, but all were polyclonal for light chain expression and had arisen from the CD5-negative, normal B-cell population. Thus the inability of EBV to induce proliferation of most CLL cells correlated with the absence of LMP expression which is invariably expressed during immortalization of normal B cells. This novel type of restricted gene expression could be compatible with evasion of host immune responses and consequent long-term survival of the cell in vivo.     

The effect of flumazenil on subclinical psychometric or neurophysiological alterations in cirrhotic patients: a double-blind placebo-controlled study

Posttransplant lymphoproliferative disorder (PTLD) is associated with Epstein-Barr virus (EBV), and may clinically resemble acute allograft rejection. Three methods to show EBV in tissue were evaluated in 15 liver allograft biopsies from 12 patients including four with PTLD: (1) semiquantitative polymerase chain reaction (PCR) for EBV DNA; (2) in situ hybridization for EBV RNA (EBER); and (3) immunoperoxidase for EBV latent membrane protein (LMP). Index cases had a PCR dot blot result of "positive" or "weak positive." Findings were correlated with histology, clinical data, therapy, and outcome. All four PTLD patients had a clinical diagnosis of acute rejection. All four showed EBV: PCR 4, EBER 4, LMP 3, Liver function tests were elevated in three, but EBV viral capsid antigen (VCA) IgM was not increased in three, but EBV viral capsid antigen (VCA) IgM was not increased in three. Immunosuppression was withdrawn and all four patients underwent a second transplantation. One died 4 days posttransplant with disseminated PTLD, two died of sepsis at 1.5 and 14 months, and one is well at 3 years without PTLD. Eleven biopsies without PTLD showed: acute rejection 7, acute rejection and hepatitis 1, hepatitis B 1, and non-inflammatory changes 2. In this group, EBV results included: PCR weak positive in 10 and 1+ in one, EBER negative in ten and rare positive cells in one, LMP negative in 11. Liver function tests were elevated in 10, whereas VCA IgM was not increased in three and increased in one. Patients with acute rejection were treated with increased immunosuppression: none developed PTLD, with follow-up of at least 6 months in nine cases. Two patients died within 4 months of biopsy. One patient with PTLD in tonsils had a liver biopsy showing both acute rejection and EBV (PCR 1+, rare EBER + small cells). Histological studies combined with special EBV detection methods, can be useful to evaluate atypical lymphoid infiltrates in liver allograft biopsies and confirmation of a diagnosis of PTLD. All three methods are useful; EBER and PCR are the most sensitive. EBER and LMP can use paraffin sections.     

Iron chelators for thalassaemia

We reported previously that a characteristic Epstein-Barr virus latent membrane protein 1 (EBV-LMP1) gene was associated with nasopharyngeal carcinoma (NPC) in Hong Kong. It showed a 30 bp deletion at the carboxyl terminus with specific amino acid substitution Asp at codon 335 with reference to Gly in B95-8 LMP1. This deletion variant Asp335 was present in over 90% of NPC biopsy specimens. The present study attempted to determine the whole encoding sequence of the LMP1 gene in different EBV isolates from NPC, and its relation with EBV types. We found that 92% (34/37) of primary NPC tumours harboured EBV-1 and possessed the LMP1 deletion variant, of which 86% were Asp335 and 6% were Gly335. EBV-2 was present in 8% (3/37) of tumours and all contained the retention variant of the LMP1 gene. Sequencing of the whole encoding region of the LMP1 gene revealed that the deletion variant Asp335 and deletion variant Gly335 carried similar sequences. They showed 43 common nucleotide substitutions in 41 codons with reference to B95-8. The retention variant showed 52 base changes in 46 codons compared with B95-8. The amino acid alterations in both the deletion and retention variants were mostly clustered at the transmembrane domain of the protein. Furthermore, half of the substitutions were common to both variants, suggesting a common evolutionary selection pressure. Nonetheless, the 2 LMP1 variants showed differences in nucleotide alterations and were associated with different EBV types, suggesting the presence of 2 distinct EBV strains in Hong Kong NPC.     

Distribution of monoclonal antibodies in intestinal and urogenital secretions of mice bearing hybridoma ''backpack'' tumours

Primary Epstein-Barr virus (EBV) infection may manifest itself as a benign lymphoproliferative disorder, infections mononucleosis (IM). EBV infection has been characterized in lymphoreticular tissues from nine patients with IM using the abundantly expressed EBV-encoded nuclear RNAs (EBERs) as a marker of latent infection. Expression of the virus-encoded nuclear antigen (EBNA) 2 and of the latent membrane protein (LMP) 1 was seen in variable proportions of cells in all cases. Double labelling revealed heterogeneous expression patterns of these proteins. Thus, in addition to cells revealing phenotypes consistent with latencies I (EBNA2-/LMP1-) and III (EBNA2+/LMP1+), cells displaying a latency II pattern (EBNA2-/LMP1+) were observed. Cells expressing EBNA2 but not LMP1 were also detected; whilst this may represent a transitory phenomenon, the exact significance of this observation is at present uncertain. EBER-specific in situ hybridization in conjunction with immunohistochemistry revealed expression of the EBERs mainly in B-lymphocytes, many of which showed features of plasma cell differentiation. By contrast, convincing evidence of latent EBV infection was not found in T-cells, epithelial or endothelial cells. Double-labelling immunohistochemistry revealed expression of the replication-associated BZLF1 protein in small lymphoid cells, often showing plasmacytoid differentiation. There was no unambiguous expression of this protein in other cell types. These results suggest that B-cells are the primary target of EBV infection and that plasma cells may be a source of infectious virus found in the saliva of IM patients.     

Epstein-Barr virus-associated dural leiomyosarcoma in a man infected with human immunodeficiency virus. Case report

A 35-year-old man infected with human immunodeficiency virus presented with cervical myelopathy of 2 months duration. Clinical and radiographic evaluation revealed a discrete, subdural mass at C-6. At surgery, the mass proved to have a dural attachment and thus clinically, radiographically, and grossly, it resembled meningioma. Histopathological analysis revealed a leiomyosarcoma that stained diffusely for muscle-specific actin. Electron microscopy revealed basal lamina surrounding the tumor cells and intracytoplasmic bundles of myofilaments. Epstein-Barr virus (EBV) was demonstrated within tumor cell nuclei by in situ hybridization for EBER1 messenger RNA and immunohistochemical staining for EBNA2 protein. Epstein-Barr virus latent membrane protein (LMP1) was not detected. This is the first documentation of an EBV-associated smooth-muscle tumor of the dura, and the first demonstration that tumors in this location contain EBV in an unusual form of latency not seen in lymphoid cell lines. With increasing numbers of individuals being afflicted with long-term immunosuppression, EBV-associated dural leiomyoma and leiomyosarcoma may be encountered more frequently in the future.     

Presence of Epstein-Barr virus in lymphoepithelioma-like carcinoma of the middle ear

AIM: To examine the association of Epstein-Barr virus (EBV) with carcinoma of the ear. METHODS: Five non-keratinising squamous cell carcinomas and two undifferentiated carcinomas of the ear were examined. In situ hybridisation was used to localised EBV-encoded RNAs (EBER). Immunohistochemical methods to detect LMP-1 and EBNA2 were performed in the EBER positive cases. RESULTS: Two cases were EBER positive, including one non-keratinising and one undifferentiated carcinoma. Both showed identical morphology to those arising from the nasopharynx, with abundant lymphoid stroma. They were both negative for LMP-1 and EBNA2. CONCLUSIONS: EBV associated carcinoma with the morphology of lymphoepithelioma can also arise from the middle ear.     

In vivo growth of Epstein-Barr virus transformed B cells with mutations in latent membrane protein 2 (LMP2)

Epstein-Barr virus (EBV) causes infectious mononucleosis in adolescents and is associated with malignant B lymphocyte proliferation in AIDS patients, patients undergoing immune suppression for organ transplantation, and SCID mice. In vitro, EBV transformed, latently infected lymphoblastoid B cell lines (LCLs) contain EBV episomes and express nine virus encoded proteins. Six are nuclear proteins (EBNAs) and three are the integral membrane proteins, LMP1, LMP2A, and LMP2B. To determine if LMP2 was essential for in vivo growth, SCID mice were injected with LCLs containing wild-type EBV (LMP2+) or with LCLs transformed with EBV containing mutations in either LMP2A or LMP2B (LMP2-). SCID mice injected with the LMP2+ or LMP2- LCLs were monitored for tumor development, length of time to tumor development, and phenotypic characterization of the resulting tumors. No difference was observed in any of the above parameters between LMP2+ and LMP2- LCLs demonstrating that LMP2 is not essential for the in vivo growth of EBV transformed B lymphocytes in SCID mice.     

Both Epstein-Barr viral nuclear antigen 2 (EBNA2) and activated Notch1 transactivate genes by interacting with the cellular protein RBP-J kappa

The Epstein-Barr viral nuclear antigen 2 (EBNA2) plays a key role during establishment and maintenance of B cell immortalization after Epstein-Barr virus (EBV) infection. EBNA2 acts as a transactivator of cellular and viral genes. We studied two EBNA2 regulated viral promoters (TP1 promoter and LMP/TP2 promoter) in detail to learn more about the molecular mechanisms of EBNA2-mediated transactivation. In both promoters we could identify at least one binding site for the cellular repressor protein RBP-J kappa. EBNA2 is tethered to the EBNA2 responsive promoter elements by interaction with this cellular protein. Although necessary, the binding of RBP-J kappa is not sufficient for EBNA2-mediated transactivation. At least two further cellular proteins, which are different in the studied promoters are important for efficient transactivation. The identification of RBP-J kappa as central mediator of EBNA2 transactivation suggested an interference of EBNA2 with the highly conserved Notch receptor signal transduction pathway. We could show that an activated form of the Notch receptor can transactivate a reporter construct containing a hexamer of the two RBP-J kappa binding sites of the TP1 promoter supporting the idea that EBNA2 acts as a functional equivalent of an activated Notch receptor.