Epstein-Barr virus (EBV) is an important cause of human cancers world-wide, including both B cell and epithelial cell malignancies. Although the role of EBV-encoded latency proteins in driving these cancers is well accepted, the importance of lytic viral proteins remains relatively controversial. Nevertheless, lytic infection in a subset of EBV-infected cells in vivo may be required for efficient EBV-induced lymphoma formation. We previously showed that a lytic-defective EBV mutant (missing the BZLF1 (Z) immediate-early (IE) gene) is impaired for the ability to form lymphomas in a humanized mouse model that allows horizontal virus transmission. In addition, we and others have identified growth factors and immunosuppressive factors released from lytically infected B cells that likely enhance the growth and survival of nearby latently infected B cells. Furthermore, we have recently discovered that a BZLF1 promoter variant (known as Zp-V3) that is over-represented in certain EBV-positive malignancies (including NPC and AIDS-related lymphomas) relative to its frequency in non-malignant tissues confers enhanced lytic viral reactivation in vitro due to binding of the cellular NFAT transcription factor to the Zp-V3 variant (but not the prototype promoter, Zp-P). Although the Zp-V3 form of the promoter is relatively uncommon in type 1 EBV strains, it is present in all type 2 strains (which are very common in the malaria belt of Africa). These results suggest that enhanced lytic EBV infection increases the likelihood of EBV-induced lymphomas in vivo, and that a particular cancer-associated variant of the Z promoter promotes lytic infection in EBV-infected B cells. In this proposal, we will use two different humanized mouse models to compare the phenotypes of EBV containing the Zp-P form versus the cancer- associated (Zp-V3) form of the BZLF1 promoter, and to explore mechanism(s) by which lytic EBV infection promotes lymphomagenesis. We will also determine whether EBV loads are higher in malaria-infected children co-infected with Zp-V3 containing EBV strains versus Zp-P containing strains.
Our Specific Aims are 1) to use humanized mouse models to examine the in vivo phenotypes of Zp-V3 versus Zp-P containing type 1 or type 2 EBV strains; 2) to compare the phenotypes of BALF5 (the viral DNA polymerase)-deleted versus BZLF1-deleted EBV mutants in humanized mice, and explore whether blocking lytic EBV DNA replication with the antiviral drug, acyclovir, inhibits the development of EBV-induced lymphomas; and 3) to determine whether the Zp-V3 promoter variant is associated with higher plasma levels of EBV in malaria-infected African children. We hypothesize that the EBV Zp-V3 variant will enhance EBV-induced lymphomas in humanized mice by increasing lytic EBV infection, and that this variant is also associated with enhanced lytic EBV replication in malaria-infected children. If so, these results will suggest that the presence of the Zp-V3 variant may be a useful biomarker for predicting increased risk of EBV-induced malignancy in humans.
While latent Epstein-Barr virus (EBV) infection is essential for the development of EBV-induced tumors, whether lytic viral infection also contributes (and through what mechanism(s)) is not clear. In this proposal, we will use two different humanized mouse models to determine if EBV strains that contain a variant of the viral BZLF1 promoter (Zp-V3) that is over-represented in human tumors, and which promotes lytic viral reactivation in B cells in vitro, is more highly transforming in vivo, and will define mechanism(s) by which lytic EBV infection contributes to EBV-induced tumors in vivo. In addition, we will determine if patients infected with malaria are more likely to have high EBV viral loads when infected with EBV strains containing the Zp-V3 form of the BZLF1 promoter. These studies should reveal whether excessive lytic EBV replication promotes EBV-induced lymphomas in vivo, and whether patients infected with Zp-V3 containing EBV have an enhanced risk for EBV-induced tumors.
