Constitutive Activation of Epstein-Barr Virus (EBV) Nuclear Antigen 1 Gene Transcription by IRF1 and IRF2 during Restricted EBV Latency

Abstract

Epstein-Barr virus (EBV) infection in immunocompetenthumans is predominantly latent and persists for the life of the individual (reviewed in reference 41). Recently, it has been shown that EBV is capable of adopting at least three distinct forms of latency (27). Type III latency is observed upon in vitro infection of B lymphocytes and results in immortalization and continuous proliferation of the infected B cells via the action of a subset of the six EBV nuclear antigens (EBNAs; EBNA1, EBNA2, EBNA3a, EBNA3b, EBNA3c, and EBNA4) and three membrane proteins (LMP1, LMP2a, and LMP2b) expressed in the type III (immortalizing) program. However, in vivo, the type III latency program is likely to be only transiently observed upon initial infection of a naive host, since several of the type III latent antigens elicit a potent cytotoxic T-lymphocyte response resulting in very effective elimination of type III latently infected B cells (reviewed in reference 33).EBV is also capable of entering programs of latency in which the expression of latent antigens is restricted with respect to the type III program. In type I latency, only the EBNA1 protein is expressed (64). It has recently been shown that EBNA1 peptides do not enter the major histocompatibility complex class I antigen processing and presentation pathway (29, 39, 52), and cells which express only EBNA1 are not detected by host cellular immune surveillance mechanisms (63). The type II latency program differs from type I latency only in the expression of variable combinations of LMP1, LMP2a, and LMP2b, in addition to EBNA1. Since the type I latency program was identified, there has been speculation that type I latently infected B lymphocytes represent the lifelong reservoir of virus in immunocompetent, seropositive individuals, and there have recently been several reports which provide evidence that a type I-like form of restricted viral latency does indeed exist in healthy carriers of EBV (6,51,60,85).Investigations into the molecular basis of type III and type I latency have demonstrated that distinct promoters are used to drive transcription of the EBNAs in each latent program. In type III latency, transcription of all six nuclear antigens is initiated from either Wp (during initial infection of resting B cells) or Cp (in cycling B cells), and the long primary transcripts are differentially spliced to generate the mature EBNA transcripts (75, 94; see Fig. 1A for a schematic representation of EBV latent transcription and locations of promoters). Recent investigations have shown that transcription of the EBNA1 gene in type I latency is driven by a promoter designated Qp, which is considerably downstream of the type III latency EBNA gene promoters (57,(70)(71)(72). Qp has an architecture with numerous similarities to eukaryotic housekeeping

Co-occupancy of the interferon regulatory element of the class II transactivator (CIITA) Type IV promoter by interferon regulatory factors 1 and 2

“…In this respect, the CIITA Type IV IRF-E and human GBP IRF-E represent a unique class of IRF-Es because they are both responsive to IFN-g, i.e., responsive to IRF-1, and activated by IRF-2. The IRF-E of the EBNA1 promoter, which is activated by IRF-2, is not responsive to IFN-g (Schaefer et al, 1997). The IRF-Es of the promoters of histone H4 (Vaughan et al, 1995(Vaughan et al, , 1998 and VCAM-1 (Jesse et al, 1998), the other two genes that are activated by IRF-2, according to current understanding, are also not responsive to IFN-g (Schaefer et al, 1997).…”

“…The IRF-E of the EBNA1 promoter, which is activated by IRF-2, is not responsive to IFN-g (Schaefer et al, 1997). The IRF-Es of the promoters of histone H4 (Vaughan et al, 1995(Vaughan et al, , 1998 and VCAM-1 (Jesse et al, 1998), the other two genes that are activated by IRF-2, according to current understanding, are also not responsive to IFN-g (Schaefer et al, 1997). One interesting question is whether the IRF-Es of these latter three promoters are adjacent to an IRF-1 inhibitory region, present in the CIITA Type IV promoter between nucleotides 7309 and 785, as discussed above.…”

“…Cooperative activation of human GBP promoter and co-occupancy of the GBP IRF-E by IRF-1 and IRF-2 CIITA is among a group of genes, including EBNA1 (Schaefer et al, 1997), histone H4 (Vaughan et al, 1995(Vaughan et al, , 1998, and murine muscle VCAM-1 (Jesse et al, 1998), which are activated by IRF-2 and contain a G residue at position 11 of their IRF-Es. The human guanylate binding protein (GBP) promoter IRF-E also has a G residue at position 11 and is thought to be strongly activated by IRF-1.…”

“…In addition, other recent evidence demonstrates that IRF-2 is not always a transcriptional repressor. IRF-2 activates promoters for Epstein-Barr virus nuclear antigen-1 (EBNA1) (Schaefer et al, 1997), histone H4 (Vaughan et al, 1995(Vaughan et al, , 1998, and murine muscle vascular cell adhesion molecule-1 (VCAM-1) (Jesse et al, 1998).…”

“…The IRF-E is also found in the promoters of three constitutively expressed genes: EBNA1 (Schaefer et al, 1997), histone H4 (Vaughan et al, 1995(Vaughan et al, , 1998, and murine muscle VCAM-1 (Jesse et al, 1998), which are activated by IRF-2. The consensus IRF-E sequence has been established as G(A)AAA G / C T / C GAAA G / C T / C .…”