REPLICATION AND ORGANIZATION OF EUKARYOTIC GENOMIC DNA

HONG YAN, Ph.D., Associate Member

The initiation of DNA replication involves a minimum of four factors: a specific DNA sequence (origin), an initiator protein that binds to the origin, a helicase that unwinds the origin, and a single-stranded DNA binding protein that stabilizes the unwound origin. In eukaryotic cells, the initiator protein is the origin recognition complex (ORC) and the single-stranded DNA binding protein is the replication protein A (RPA). However, the helicase has not been identified and the nature of origins remains elusive except in the case of the budding yeast Saccharomyces cerevisiae. While some origin mapping studies suggest that metazoan cells utilize specific DNA sequences as origins, others indicate that replication can initiate at random sequences. Analysis of these conflicting data has led to the proposal that metazoan chromosomes contain many potential origin sequences, but only a small subset is actually selected and used during each cell cycle.It appears that sub-nuclear structure may play a role in determining which subset of origins is selected. Specifically, since DNA replication takes place at discrete locations (replication foci) within the nuclei of metazoan cells, an active replication origin may be one that not only contains sequence information but also is associated with a replication focus complex.

Our research is directed towards identifying the molecules responsible for the assembly of replication foci. Using Xenopus egg extracts as the model system, we have previously identified and purified a protein, focus forming activity 1 (FFA-1), which is required for the assembly of replication foci (Yan and Newport, Science 269:1883, 1995). Biochemical characterization has indicated that FFA-1 has DNA helicase activity. The FFA-1 gene encodes the Xenopus ortholog (true homolog) of the human Werner syndrome (WS) gene product (WRN), a member of the RecQ DNA helicase family (1). Mutations in the WS gene cause premature aging and increased risks for cancer. A common cause for these two phenotypes appears to be an inability of WS cells to stably maintain their genome; as a result, there are deletions and rearrangements of large DNA fragments, decreased number of replication initiation events, and misfiring of replication origins. Together these observations strongly suggest that FFA-1/WRN functions to coordinate the initiation of multiple replication origins and disruption of this coordination will lead to genomic instability.

The finding that FFA-1 is a DNA helicase suggests the following model for the assembly of RPA foci. 1) FFA-1 forms an oligomer, either by itself or in association with other proteins. 2) Each FFA-1 subunit binds to one replication origin, leading to the clustering of multiple origins. 3) FFA-1 partially unwinds origins and the unwound DNA attracts RPA. We have found that the native FFA-1 is indeed present as a large multi-protein complex based on gel filtration and sucrose gradient analyses. In order to identify the other subunits in the FFA-1 complex, we have initiated a yeast two-hybrid screen using the various regions of FFA-1 coding sequence as baits. At the same time, we have started to make monoclonal antibodies against FFA-1 so that we may be able to purify the FFA-1 complex by immuno-affinity column chromatography.

In addition to pursuing the FFA-1 protein complex, we are attempting to directly visualize RPA focus structure by electron microscopy. The prediction is that a core exists where FFA1 and RPA are localized and onto which multiple DNA loops are anchored. Such a result would provide not only the most definitive evidence for a loop structure for interphase chromatin but also valuable insights into the mechanism of loop formation.

To test the role of DNA helicase activity in RPA focus assembly, we have expressed recombinant FFA-1 in the baculovirus system. Our preliminary data suggested that the recombinant FFA-1 was active in promoting RPA focus formation. We then constructed a helicase-deficient mutant FFA-1 and are now testing its activity in focus formation. In addition to the helicase activity, FFA-1 also contains a conserved nuclease domain near the N-terminus. A similar domain in WRN protein has been confirmed biochemically to exhibit 3'Æ5' exonuclease activity. We have started to construct a nuclease-deficient FFA1 to test whether nuclease activity is required for the assembly of focus structure. Experiments such as these should help us to further understand the mechanism of focus assembly.

Helicases involved in DNA replication often interact with other replication proteins such as the single-stranded DNA binding protein and DNA polymerases. We have initiated a biochemical study on the potential interplay between FFA-1, RPA, and DNA polymerase alpha. Studies with the SV40 in vitro replication system have suggested that a high level of RPA can inhibit the activity of DNA polymerase alpha and this inhibition can be specifically relieved by large T antigen. If FFA-1 functions as the cellular counterpart of T antigen, one would predict a similar scenario. A positive effect of FFA-1 on DNA polymerase alpha would provide a strong evidence for the involvement of FFA1 in DNA replication.

Although our results and the phenotypes of WS cells indicate that FFA-1/WRN plays an important role in replication initiation, WRN is not an essential gene. WS patients develop symptoms starting at the second decade of life and many of these patients carry truncation mutations that result in proteins completely missing the helicase domain. A reasonable explanation for this apparent inconsistency is that there may be a second gene with redundant function to WRN. A good candidate for this second gene is the Bloom's syndrome (BS) gene, which encodes a helicase (BLM) of similar size with significant homology to WRN in the 300 amino acid helicase domain. BS is associated with small size, sun-sensitive facial erythema, immunodeficiency, and cancer. In BS cells, chromosome gaps and breakages are easily observed and interchanges between homologous chromosomes and sister chromatids are increased. The rate of nascent DNA chain elongation is retarded and abnormal replication intermediates (about 20 kilobasepairs in size) are accumulated. Together these observations suggest that BLM, like WRN, functions in the replication and organization of DNA.

We have isolated a Xenopus BLM homolog (xBLM) from an oocyte cDNA library. The complete sequence predicts a protein with extensive homology to the human BLM. We have expressed several fragments of xBLM as GST fusion proteins in E. coli, which have then been injected into rabbits to raise antibodies. Once the antibodies are obtained, we will do indirect immunofluorescence staining to see whether xBLM, like FFA-1 and RPA, binds to chromatin at discrete foci. We will also deplete xBLM from egg extracts to determine the effect on the formation of RPA foci. We expect that a double depletion, but not single depletions, of xBLM and FFA-1 will abolish the ability of egg extracts to form foci and replicate DNA. We believe that BLM is itself an important protein and, even if it plays no role in RPA focus assembly, merits an independent study using the Xenopus system.

1.   YAN, H., CHEN, C.Y., KOBAYASHI, R., NEWPORT, J. Replication focus activity 1 and the Werner syndrome gene product. Nature Genetics 19:375-378, 1998.

YOUNG, M.R., SUZUKI, K., YAN, H. GIBSON, S., TYE, B.K. Nuclear accumulation of Saccharomyces cerevisiae Mcm3 is dependent on its nuclear localization sequence. Genes Cells 2(10):631-643, 1997.

Illustrations or unpublished data in these reports should not be used without permission of the author.

Fox Chase Cancer Center

Scientific Report 1998