Archival oral tissues comprising 51 squamous cell carcinomas, 18 non-malignant lesions and 7 normal mucosa samples were investigated for human herpesvirus-6 (HHV-6)-encoded antigens and HHV-6 DNA. The virus-specific antigens were detected by an immunohistochemical method using monoclonal antibodies. Two further techniques used for HHV-6 DNA detection included the polymerase chain reaction (PCR) with virus-specific primers and in situ hybridization using digoxigenin-labelled oligonucleotides specific for HHV-6A and HHV-6B genotypes. A high proportion (79-80%) of the squamous cell carcinomas were positive for HHV-6 with the various detection methods. In cases of lichen planus and leukoplakia a high prevalence rate (67-100%) was noted with in situ hybridization and immunohistochemical techniques but a lower proportion (22-33%) was detected with the PCR method. All 7 normal tissues tested were negative for HHV-6. The HHV-6 variant B was found in 60% of the oral carcinoma tissues analysed. The study demonstrates the frequent presence of HHV-6 in neoplastic and non-malignant lesions of the oral cavity. While the role of HHV-6 in oral mucosal tissues remains to be determined, the in vitro tumorigenic potential of the virus suggests a possible role in the etiopathogenesis of oral lesions.
In this study, we have sequenced the C-terminal part of the Epstein-Barr virus (EBV)-BNLF-1 gene encoding for the latent membrane protein-1 from tissues of EBV-positive Danish Hodgkin's disease (HD) and of Danish and Malaysian peripheral T-cell lymphomas (PTLs) and from tonsils of Danish infectious mononucleosis (IM). Our study showed that some of the 7 single-base mutations and the 30-bp deletion previously detected between codons of amino acid 322 and 366 in the BNLF-1 gene of the nasopharyngeal carcinoma cell line CAO were present in all Malaysian PTLs and in 60% of the Danish PTLs. In HD and the IM cases, the mutations were present in about 30%. The 30-bp deletion and the single base mutations occurred independently, and mutations were detectable in the majority of EBV type B-positive cases. These findings suggest that the 30-bp deletion and the 7 single-base mutations in the C-terminal part of the CAO-BNLF-1 gene do not characterize a new EBV type A substrain. Rather, some of the positions of single base mutations and the 30-bp deletion are hot spots that may have mutated independently through the evolution of EBV strains.
The BamHI Z EBV replication activator (ZEBRA) protein is involved in the switch from latency to productive cycle of Epstein-Barr virus. A recombinant ZEBRA protein was synthesized and assessed in enzyme-linked immunosorbent assay (ELISA) for serum IgG response in nasopharyngeal carcinoma (NPC) patients. In 100 NPC serum samples that were positive for IgA to the EBV viral capsid antigen (VCA), 75% had IgG anti-ZEBRA antibodies. In contrast, only 3/83 (3.6%) serum samples from healthy donors and 2/50 (4%) from other cancers were positive for IgG to ZEBRA. Interestingly, in a selected group of 100 NPC sera negative for IgA to VCA, 25% contained IgG anti-ZEBRA antibodies. This suggests that the ELISA for IgG anti-ZEBRA may also identify earlier cases of NPC not detected by the conventional immunofluorescence test for IgA to VCA.
LMP-1, an Epstein-Barr viral (EBV) latency protein, is considered a viral oncogene because of its ability to transform rodent fibroblasts in vivo and render them tumorigenic in nude mice. In human B cells, EBV LMP-1 induces DNA synthesis and abrogates apoptosis. LMP-1 is expressed in EBV-transformed lymphoblastoid cell lines, nasopharyngeal carcinoma (NPC), a subset of Hodgkin's disease (HD), and in EBV-associated lymphoproliferative disorders (EBV-LPDs). Recently, focused deletions near the 3' end of the LMP-1 gene (del-LMP-1, amino acids 346-355), in a region functionally related to the half-life to the LMP-1 protein, have been reported frequently in human immunodeficiency virus (HIV)-associated HD (100%) and EBV+ Malaysian and Danish peripheral T-cell lymphomas (100%, 61% respectively), but less frequently in cases of HD not associated with HIV (28%, 33%) and infectious mononucleosis (33%). To further investigate the potential relationship of del-LMP-1 to EBV-LPDs associated with immunosuppression or immunodeficiency, we studied 39 EBV-associated lymphoproliferations (10 benign, 29 malignant) from four distinct clinical settings: posttransplant (4 malignant, 1 reactive); HIV+ (18 malignant, 2 reactive); nonimmunodeficiency malignant lymphoma (ML) (7 cases); and sporadic EBV infection with lymphoid hyperplasia (7 cases). The presence of EBV within lymphoid cells was confirmed by EBV EBER1 RNA in situ hybridization or by polymerase chain reaction (PCR) analysis. EBV strain type and LMP-1 deletion status were determined by PCR. EBV strain types segregated into two distinct distributions: HIV+ (9 A; 11 B) and non-HIV (19 A, 0 B), consistent with previous reports. Overall, del-LMP-1 were found in 1 of 5 (20%) Burkitt lymphomas (BL); 17 of 24 (71%) aggressive non-Hodgkin's lymphoma (agg-NHL), and 2 of 10 (20%) reactive lymphoid proliferations. Of the agg-NHLs, del-LMP-1 were present in 4 of 4 PT-ML (100%); 10 of 15 HIV+ ML (67%); and 3 of 5 nonimmunodeficiency malignant lymphoma (ML, 60%). A total of 2 of 7 (28%) sporadic EBV-associated lymphoid hyperplasias contained a del-LMP-1. All del-LMP-1 were identical by DNA sequence analysis. No correlation was identified between the presence of del-LMP-1 and the EBV strain type observed. The high incidence of del-LMP-1 observed in agg-NHLs (71%), in contrast to the relatively low incidence observed in reactive lymphoid proliferations (28%), suggests that the deleted form may be preferentially selected in lymphomatous processes. All posttransplant agg-NHLs contained a del-LMP-1, and a similar frequency of del-LMP-1 was observed in both HIV-associated ML (66%) and nonimmunodeficiency ML (60%), suggesting that impairment of immune function alone is not a requirement for the expansion of malignant cells infected by EBV stains containing the deleted LMP-1 gene.