1. ¹Research Institute of Radiology and Roentgenology, St Petersburg, Russia
2. ²CCRI, Children's Cancer Research Institute, St Anna Kinderspital, Vienna, Austria
3. ³Division of Hematology, St Paul's Hospital, Vancouver, BC, Canada
4. ⁴Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada
5. ⁵Department of Pathology and Laboratory Medicine, BC Cancer Agency, Vancouver, BC, Canada

Correspondence: Dr VS Lestou, Department of Pathology and Laboratory Medicine, British Columbia Cancer Agency, 600 W 10th Ave, Vancouver, BC, Canada V5Z 4E6. E-mail: vlestou@bccancer.bc.ca

Received 26 March 2004; Revised 20 May 2004; Accepted 21 May 2004.

Abstract

Epstein–Barr (EBV) virus is associated with malignancies such as lymphoma and carcinoma. Infection of cells with EBV may result in either lytic infection with production of viral particles, characterized by the presence of linear DNA forms, or latent infection, characterized by either episomal or integrated DNA forms. To examine whether the different lytic and latent EBV DNA forms can reliably be distinguished in single human cells, in situ hybridization was performed in EBV-positive cell lines. Immunocytochemistry and Southern blot analysis were performed supplementary to in situ hybridization. In latent infection, three in situ hybridization patterns were observed: large-disperse (episomal), small-punctate (integrated) and combined (both), signal types 1, 2 and 3 respectively. These were associated with expression of latent membrane protein 1, but not with Z fragment of Epstein–Barr replication activator or viral capsid antigen. In lytic infection, three additional in situ hybridization patterns were observed: nuclear membrane associated, bubble (filling up the nucleus) and spillover (covering the lysed cells) signals types 4, 5 and 6 respectively. Signal types 4 and 5 were associated with expression of latent membrane protein 1 and Z fragment of Epstein–Barr replication activator but not viral capsid antigen, whereas type 6 was associated with expression of viral capsid antigen only. Southern blot analysis confirmed these results; however, low copy numbers of integrated virus were often missed by Southern blot, confirming that in situ hybridization is more sensitive in determining the presence of all types of EBV DNA. In situ hybridization may prove useful in rapidly screening large series of tissue microarrays and other clinical specimens for the presence of lytic or latent EBV.