Analysis of Epstein-Barr Virus Infections in Lymphoma Cell Lines.

Epstein-Barr virus (EBV, human herpesvirus type 4) is ubiquitously distributed in all human populations, reaching infection rates of more than 90%. EBV is known to infect B- lymphocytes and mucosal epithelium cells and to establish latent or productive infections. The virus is the causative agent of infectious mononucleosis and closely associated with the endemic form of Burkitt lymphoma (BL). EBV has also been associated with various lymphoid and epithelial malignancies, such as Hodgkin, T-cell, and AIDS-related lymphomas, and lymphoepithelioma-like carcinomas of several organs. In vitro, B- lymphocytes are transformed by EBV into permanent lymphoblastoid cell lines (B-LCL). We investigated the EBV infection status of primate cell lines by PCR (406 human, 4 monkey). This method detects EBV genomes integrated into the eukaryotic chromosomes, non-integrated EBV episomes, and linear genomes of active EBV particles. The analyses revealed that 38/410 cell lines contain the EBV genome. All EBV+ cell lines were established from B- lineage leukemia/lymphoma cells (13/52 B-non-Hodgkin cell lines, 10/13 BL cell lines, 2/2 hairy cell leukemia cell lines, 1/6 plasma cell leukemia/myeloma cell lines) or are B-LCLs (9/9), natural killer cells (2/2), and one monkey cell line. No cell lines from other tissues were found to be EBV+. To further examine the production of EBV particles in the PCR+ cell lines, we analyzed the expression of the BZLF1 protein by Western blotting applying a ZEBRA monoclonal antibody. The cell lines were analyzed untreated as well as treated with the phorbol ester TPA for 3 days to induce the lytic phase of the EBV infection. Four cell lines exhibited a BZLF1 specific band a priori; after stimulation with TPA, 4 further cell lines expressed BZLF1 protein to various extents. To distinguish between linear DNA of herpesviruses (DNA form of active viruses) and covalently closed circles of episomal DNA, we performed Gardella gels applying crude lysates from cell cultures. Except for cell line NAMALWA and its subclones, DG-75, DOHH-2, and OCI-LY19 (all EBV-PCR+ cell lines) showed at least one band of episomal genomes. Some cell lines showed two episomal bands pointing to a double infection or to mutated episomes. The amount of linear DNA does not correlate with the number of episomes. Southern blots of genomic DNA revealed different genotypes of EBV, except for those cell lines which were established with B95-8 virus particles. To determine distribution of EBV genomes in single cells, we established a fluorescence in situ hybridization (FISH) method with a Cy3-labeled cosmid clone containing a genomic EBV fragment. The method showed for various cell lines that only a few cells contain high amounts of EBV genomes (several hundred) whereas the vast majority harbors only a few genomes in the nuclei. FISH appears to be superior to other methods, allowing for EBV analysis at the single cell level to determine the cellular permissiveness. In summary, we could show that EBV is constitutively produced in a few B-lymphoma derived cell lines and can be induced in several other cell lines. These cell lines represent valuable tools for further investigation into the biology of EBV infection.