The importance of epigenetic alterations in the development of epstein-barr virus-related lymphomas

Abstract

Epstein-Barr virus (EBV), a human gammaherpesvirus, is associated with a series of malignant tumors. These include lymphomas (Burkitt's lymphoma, Hodgkin's disease, T/NK-cell lymphoma, post-transplant lymphoproliferative disease, AIDS-associated lymphoma, X-linked lymphoproliferative syndrome), carcinomas (nasopharyngeal carcinoma, gastric carcinoma, carcinomas of major salivary glands, thymic carcinoma, mammary carcinoma) and a sarcoma (leiomyosarcoma). The latent EBV genomes persist in the tumor cells as circular episomes, co-replicating with the cellular DNA once per cell cycle. The expression of latent EBV genes is cell type specific due to the strict epigenetic control of their promoters. DNA methylation, histone modifications and binding of key cellular regulatory proteins contribute to the regulation of alternative promoters for transcripts encoding the nuclear antigens EBNA1 to 6 and affect the activity of promoters for transcripts encoding transmembrane proteins (LMP1, LMP2A, LMP2B). In addition to genes transcribed by RNA polymerase II, there are also two RNA polymerase III transcribed genes in the EBV genome (EBER 1 and 2). The 5' and internal regulatory sequences of EBER 1 and 2 transcription units are invariably unmethylated. The highly abundant EBER 1 and 2 RNAs are not translated to protein. Based on the cell type specific epigenetic marks associated with latent EBV genomes one can distinguish between viral epigenotypes that differ in transcriptional activity in spite of having an identical (or nearly identical) DNA sequence. Whereas latent EBV genomes are regularly targeted by epigenetic control mechanisms in different cell types, EBV encoded proteins may, in turn, affect the activity of a set of cellular promoters by interacting with the very same epigenetic regulatory machinery. There are EBNA1 binding sites in the human genome. Because high affinity binding of EBNA1 to its recognition sites is known to specify sites of DNA demethylation, we suggest that binding of EBNA1 to its cellular target sites may elicit local demethylation and contribute thereby to the activation of silent cellular promoters. EBNA2 interacts with histone acetyltransferases, and EBNALP (EBNA5) coactivates transcription by displacing histone deacetylase 4 from EBNA2-bound promoter sites. EBNA3C (EBNA6) seems to be associated both with histone acetylases and deacetylases, although in separate complexes. LMP1, a transmembrane protein involved in malignant transformation, can affect both alternative systems of epigenetic memory, DNA methylation and the Polycomb-trithorax group of protein complexes. In epithelial cells LMP1 can up-regulate DNA methyltransferases and, in Hodgkin lymphoma cells, induce the Polycomb group protein Bmi-1. In addition, LMP1 can also modulate cellular gene expression programs by affecting, via the NF-κB pathway, levels of cellular microRNAs miR-146a and miR-155. These interactions may result in epigenetic dysregulation and subsequent cellular dysfunctions that may manifest in or contribute to the development of pathological changes (e.g. initiation and progression of malignant neoplasms, autoimmune phenomena, immunodeficiency). Thus, Epstein-Barr virus, similarly to other viruses and certain bacteria, may induce pathological changes by epigenetic reprogramming of host cells. Elucidation of the epigenetic consequences of EBV-host interactions (within the framework of the emerging new field of patho-epigenetics) may have important implications for therapy and disease prevention, because epigenetic processes are reversible and continuous silencing of EBV genes contributing to patho-epigenetic changes may prevent disease development.

Figure 1.

Alternative origins of EBV DNA replication: In A, the synchronous replication of latent EBV episomes and the cellular DNA (i.e. the chromosomal site where the episome is teethered by EBNA1) is demonstrated. EBNA1 (not shown on the figure) binds to oriP, the latent origin of EBV DNA replication, and it is anchored to a possibly AT rich chromosomal site as well. After replication, duplicated EBV episomes bind to the duplicated sites on adjacent sister chromatids (based on Nanbo et al.10). In B, concatemeric replication is initiated during the lytic cycle at oriLyt. After nuclease cleavage, double stranded linear EBV genomes with variable number of terminal repeats (TRs; their borders are indicated by vertical bars) are generated (based on Sato et al.7).

Figure 2.

Clonality of EBV-associated neoplasms: The double stranded linear EBV genomes packaged into virions have variable numbers of terminal repeats (TRs, bordered by vertical bars on the figure) at their ends. After infection of cells, the termini fuse forming a circular episome harbouring distinct numbers of TRs. Clones of transformed cells can be characterized by determining the size of the BamHI fragment carrying the TRs (see Raab-Traub and Flynn11).

Figure 3.

Promoter switch in latent EBV episomes: After initial EBV infection of B cells in vitro, Wp (a promoter located to the BamHI W fragment of the EBV genome) is switched on. A giant transcript coding for 6 nuclear antigens (EBNA1-6) is generated but the activity of Wp is transient. The transactivator protein EBNA2 and the EBNA1-bound oriP enhancer may contribute to switching on Cp, a B lymphoblast specific promoter. Burkitt’s lymphoma (BL) cells use a different promoter, Qp, to generate a transcript for EBNA1 only. In vitro cultivated BL cells may switch from Qp to Cp (arrow to the left). This promoter switch results in a phenotypic drift (from memory B cell to activated B cell phenotype). A Cp to Qp switch has been predicted but never observed (broken arrow) (see Niller et al.13).