Associate Member, (Genetics)                                             Department of Cancer Biology
Roswell Park Cancer Institute
Elm and Carlton Streets
Buffalo, NY 14263
Telephone: (716) 845-8964
Fax: (716) 845-8449
E-mail: John.Yates@RoswellPark.org

General Research Interest

Epstein-Barr virus and the control of DNA replication

Laboratory Personnel

Theresa Hodin, PhD
Tanbir Najrana, PhD
Liam Coyne - Student

Current Program

Epstein-Barr virus (EBV) is a herpesvirus which causes infectious mononucleosis, can immortalize B cells, and is associated with several human malignancies, usually of B-cell or of epithelial-cell origin. Perhaps the most salient feature of the life cycle and the pathogenesis of EBV is that it establishes latent states of infection, in which virus is not produced and cells are not killed. Instead, latently infected cells may be altered by a small number of EBV genes that are expressed, and the 165-kb (kilobase-pair) EBV chromosome is maintained stably in a circular (episomal) form. The EBV episome, which in most respects resembles a small piece of human chromosome, is replicated just once per cell cycle in much the same way that human chromosomes are replicated. Our goal is to learn how the EBV chromosome is replicated during latent infection, because by understanding this central aspect of EBV infection we will also gain insight into some fundamental questions concerning the replication of human chromosomes.

The main focus of our investigation is an origin of replication on the EBV chromosome, oriP, a 1.7-kb region that allows plasmids to be replicated and to be stably maintained in human cells that contain the EBV-encoded protein, EBNA1. In the sixteen years or so since its discovery, oriP has been used to support the autonomous maintenance of artificial plasmids in human cells for a variety of purposes. OriP is somewhat misnamed, because it is really two distinct genetic components that are almost 1 kb apart, called FR and DS, only one of which supports replication directly. FR (family of repeats) is an array of 20 EBNA1 binding sites, through which EBNA1 tethers oriP-containing plasmids to human chromosomes during mitosis, thus performing an essential segregation function. DS (for dyad symmetry) contains four EBNA1 binding sites that serve as a replicator, i.e., direct bidirectional replication to initiate in the vicinity.

How does DS function as a replicator? 1. Circumstantial evidence indicates that replication at oriP is governed by "licensing" and "delicensing", the cellular mechanisms that limit chromosomal replication to one round per S phase, suggesting that initiation at oriP involves some or all of the factors and mechanisms that are expected to govern initiation of replication on human chromosomes (Yates and Guan, 1991). 2. Specific replication activity is entirely dependent on EBNA1 (Yates et al., 2000), implying that EBNA1 may recruit replication licensing (initiation) factors to DS. (EBNA1-independent activity has been noted, but such activity is only marginally above background levels, does not require the EBNA1 binding sites, and appears to be rather nonspecific.) 3. Replication activity requires at least one pair of EBNA1 sites spaced exactly 21-bp apart, center-to-center (our unpublished data.) The requirement for exact spacing implies that a precise structure is important, and each such active pair of bound EBNA1 dimers bends DS DNA by approximately 90 degrees (our unpublished data.) 4. EBNA1 is bound to its DS sites throughout all phases of the cell cycle (Hsieh et al., 1993.) 5. Our recent work indicates that certain replication initiation factors, known for their roles at replication origins in yeast, associate in some manner with DS.

The present opportunity is to investigate how replication initiation factors are recruited to DS or its vicinity and how EBNA1 is involved. It is not known how replication origins are specified on human chromosomes. In contrast to chromosomal loci, oriP is easy to manipulate and to test functionally, so it is likely to draw the attention of scientists who want to understand how human chromosomes are replicated. It will be important to map exactly where synthesis begins for the leading strands of the two diverging replication forks.

While oriP may be the most tractable replication origin of EBV, it is not the only place on the latent EBV chromosome where replication initiates. Studies in the laboratory of Carl Schildkraut have shown that replication initiates well away from oriP on most copies of the EBV chromosome most of the time, in at least one large zone that resembles the delocalized initiation that is typical of certain mammalian chromsomal loci. Replication appears to initiate within oriP less than half of the time. This prompted us to delete the replicator (DS) of oriP from the EBV chromosome. Such DS-deleted EBV mutants can establish latent infections of cell lines in culture without difficulty, and the mutant EBV chromosome is replicated solely by replication forks that initiate away from oriP (Norio et al., 2000.) EBNA1-deficient mutants, on the other hand, cannot establish latent infections, implying that it is the segregation function of EBNA1 and FR of oriP that is essential (Lee et al., 1999.) Nevertheless, the replicator function of oriP has been conserved during evolution and is likely to be important in nature. It is interesting to note that two independent DS-deleted EBV mutants are not able to establish immortalizing infections of native human B cells. It remains to be determined whether DS function is essential in this regard or merely contributes to efficiency, and whether it is required for replication initiation or in some unknown way. However it is tempting to speculate that native B cells might have more stringent requirements for replication initiation than is the case for established cell lines, which are either highly transformed or of oncogenic origin. 

Publications

Click here for a list of current publications