EBV reactivation causes widespread host de novo promoter transcription and transcriptional interference

Summary The Epstein Barr virus (EBV) is a DNA tumor virus that causes an elevated incidence of lymphoma in the HIV/AIDS population. While more than 90% of the world's population carries EBV, the virus typically exists in a “latent” state with little impact on the host. In response to certain stimuli or local microenvironmental cues, however, EBV enters the lytic viral replication program, leading to viral spread both between and within hosts. In addition to the known role of viral latency proteins in EBV associated cancers, there are well-established links between lytic replication and EBV associated cancers; and elevated EBV lytic replication in HIV co- infection (+ or – ART) likely contributes to the increased susceptibility of HIV infected individuals to EBV associated lymphomas. With minimal genetic content, viruses are inexorably dependent on host cell resources for their replication and they evolve mechanisms to modulate host cell metabolic processes to facilitate efficient virus production. One of the most conserved virus-host interactions in herpesvirus replication is “host shut off” where virus encoded factors degrade host cell mRNAs, freeing up translation resources for dedicated production of viral structural proteins. Recently, the Glaunsinger lab showed that the murine γ-herpesvirus, MHV68 also inhibits Pol II loading on cell genes. This illustrates the utilization of layered host shut off mechanisms that together overcome the rate-limiting step of viral structural protein production. Using EBV reactivation models that facilitate assessment of transcriptome changes in pure reactivating cell populations, we gained new and unexpected insights into EBV's interactions with the host cell transcriptome. Among a number of observed transcriptome alterations is our finding that EBV causes transcription initiation at more than 25,000 cell genomic sites that have no previous evidence of promoter activity. These “de novo” promoters are unusually simple, being composed of only one or two short viral transcription factor binding motif(s) that due to their short nature, occur randomly at a high frequency across the cell genome. Our preliminary data suggests that one of these motifs is recognized by the viral preinitiation complex, vPIC, which binds a “TATA”-like motif (TATT(TA)AA) and has been shown in some cases to activate viral late genes without apparent support from other transcription factors. We hypothesize that EBV utilizes the ability of vPIC to single-handedly activate transcription through these simple motifs to spawn thousands of host de novo promoters and substantially alter the cell transcriptome and chromatin architecture. Our preliminary studies also show that the most highly active de novo promoters have enriched localization within 1.5 kb of canonical cell promoters, with transcriptional orientations toward the respective canonical promoters. We hypothesize that while EBV utilizes the primal nature of late viral promoter motifs to target thousands of sites throughout the cell genome, it leverages chromatin context near canonical promoters to drive high-level de novo promoter transcription, leading to targeted transcriptional interference (TI) of host promoters. This represents a new mechanistic concept for virus-host interactions that contributes to disrupted cell gene expression and dedicated production of viral structural proteins. In this proposal, we will 1) investigate the underlying mechanisms of de novo promoter specification and local chromatin context driving high level transcription near existing canonical promoters, and 2) investigate the impact of de novo transcription and transcriptional interference in downregulating cell mRNA expression.