BHLF1, a Lytic-Cycle Gene Encoding a Long Non-Coding RNA Contributes to Epstein-Barr Virus Latency
Open Access
- Author:
- Yetming, Kristen Dominique
- Graduate Program:
- Microbiology and Immunology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- March 01, 2017
- Committee Members:
- Jeffery T. Sample, Dissertation Advisor/Co-Advisor
Jeffery T. Sample, Committee Chair/Co-Chair
Clare E. Sample, Committee Member
Todd D. Schell, Committee Member
Gregory S. Yochum, Outside Member
David J. Spector, Committee Member - Keywords:
- Epstein-Barr virus
EBV
BHLF1
long non-coding RNA
lncRNA
latency - Abstract:
- Epstein-Barr virus (EBV) is a lymphotropic, gammaherpesvirus which efficiently establishes a persistent, lifelong infection in humans. As is common with other herpesviruses, EBV has both replicative (lytic) and latent cycles of infection. EBV latency, in which there is no virus production, can be subdivided into several distinct latency programs which are characterized by differential expression patterns of the viral latency-associated proteins: EBNA1, -2, -3A, -3B, -3C, -LP and LMP1, -2A, -2B. The major long-term reservoir of EBV is the memory B cell, and as latently-infected B cells progress from an initial state of EBV-driven cell proliferation to a state of long-term viral latency within the memory B-cell pool, there is a restriction in the expression of the viral latency proteins, due to the epigenetic silencing of the EBNA promoters Wp and Cp. This restriction in latency is essential for the persistence of EBV infection as cytotoxic T lymphocytes can remove infected B cells that continue to express several of the latency-associated proteins. Historically, it has been thought that latency-associated EBV gene products only contribute to the latent cycle, and lytic-cycle genes only function during lytic infection. However, it has recently been found that there is a subset of “lytic” genes that are expressed upon the initiation of latency, one of which is BHLF1. This, along with several other lines of evidence, suggests that BHLF1 may have a latency-associated function. Thus, the goal of the work presented in this dissertation was to elucidate whether BHLF1 has a role in the establishment or maintenance of EBV latency. Using recombinant EBV (rEBV), we observed that infection of an EBV-negative cell line, BL2, with a wild-type (WT) rEBV is capable of sustaining latency III, whereas infection with mutant rEBVs, in which BHLF1 has been deleted, results in a transition from latency III to latency I within 3 months post-infection at both the protein and mRNA levels. Disruption of BHLF1 did not significantly influence the expression of other genes near the locus; thus, the phenotype observed is likely a direct consequence of the loss of BHLF1 function. In addition to the mutant phenotype in BL2 cells, we also observed a decrease in the efficiency of immortalization upon infection of primary B cells with the mutant rEBVs. Although BHLF1 contains a predicted translational open reading frame (ORF), attempts to transiently express a protein from this ORF failed unless we co-expressed the EBV SM protein, whose expression is normally restricted to the lytic cycle. We therefore hypothesize that during latent infection, BHLF1 functions as a long non-coding RNA (lncRNA). The likelihood that BHLF1 primarily functions as an lncRNA is further supported by our recent observation and that of others that the ORF is not conserved among all EBV isolates. Furthermore, RT-PCR analysis of the 5' ends of the EBNA cDNAs indicated a higher frequency of intron retention, possibly resulting in nonsense-mediated decay. Overall, these data suggest that BHLF1 may play a role in the maintenance of latency III, possibly through the regulation of splicing at the 5' ends of the EBNA RNAs.