Our DNA comes under constant attack from endogenous and exogenous mutagens. The most lethal forms of DNA damage generally arise during S phase, when the collision of DNA replication forks with lesions can cause DNA double-strand breaks (DSBs). Left unrepaired, DSBs lead to gross chromosomal rearrangements, a hallmark and likely driving force of cancer. Despite its central importance for human health, the mechanism by which forks bypass DNA damage (replication-coupled repair) is poorly understood. My laboratory has pioneered a cell-free system to elucidate the molecular events underlying vertebrate DNA replication and replication-coupled repair.
The Mechanism and Regulation of DNA Replication
The MCM2-7 helicase complex is the first violin in the orchestra of DNA replication: it unwinds DNA at the leading edge of the replisome, it is the first component of the replisome to interact with DNA damage in the template, and its access to chromatin is strictly regulated to prevent re-replication within the same cell cycle. My laboratory studies all aspects of MCM2-7's regulation and mechanism of action. We identified a novel, replication-coupled proteolysis pathway that destroys the MCM2-7 loading machinery in S phase, thereby preventing re-replication. By colliding replisomes with bulky roadblocks in the DNA template strands, we also showed that MCM2-7 unwinds DNA by translocating along the leading-strand template in the 3' to 5' direction. We also plan to study replication termination, the process by which the replisome is disassembled and the daughter molecules are decatenated after converging forks meet. We are particularly interested in the mechanism of MCM2-7 unloading during termination.
Replication-Coupled DNA Repair
DNA interstrand crosslinks (ICLs) are caused by widely used chemotherapeutics, including cisplatin and mitomycin C, but also by endogenous metabolites. Failure to repair endogenous ICLs is a likely cause of Fanconi anemia, a human cancer predisposition and bone marrow failure syndrome. ICLs are repaired primarily in S phase, suggesting a link to DNA replication. We recapitulated ICL repair in Xenopus egg extracts and showed that the process is initiated when two replication forks arrest on either side of the lesion (see figure), followed by helicase dissociation, incisions, lesion bypass, and homologous recombination. We are investigating the proteins that catalyze each step in this intricate pathway and whether one or two replication forks are needed to initiate ICL repair.
DNA protein crosslinks (DPCs) are caused by formaldehyde, a likely environmental carcinogen. We recently succeeded in recapitulating DPC repair in egg extracts and are elucidating the underlying mechanism. Like ICL repair, DPC repair involves collision of two replication forks with the damage, but downstream events differ.
Some forms of DNA damage are greatly exacerbated by DNA replication. For example, a single-stranded nick in the template is converted to a double-stranded DNA break when the adjacent DNA is unwound by an arriving replisome (fork "collapse"). We are seeking to achieve sequence-specific replication fork collapse in vitro so that we may study the architecture of collapsed forks, as well as the mechanism of their repair.
Well-known tumor suppressors such as the Fanconi anemia and BRCA proteins regulate the response to various forms of DNA damage. Remarkably, these tumor suppressor pathways are active in egg extracts, allowing us to elucidate how they promote repair. We used immunodepletion of the FANCI-FANCD2 complex from extracts to show that the Fanconi anemia pathway stimulates the programmed incisions involved in ICL repair (Figure). We are also investigating how BRCA1 and BRCA2 promote the repair of ICLs and DPCs. These experiments will help us understand how genome instability arises in breast cancer and Fanconi anemia.
Single-Molecule Analysis of DNA Replication and Repair
To deepen our understanding of genome maintenance, we also study DNA replication and repair at the single-molecule level. Real-time, single-molecule imaging can detect transient reaction intermediates, determine rate constants, and measure changes in protein stoichiometry during multistep reactions. In the past few years, we developed tools that allow us to examine vertebrate DNA replication at the single-molecule level. We showed that phage λ DNA molecules (48 kb long) stretched on the surface of a microfluidic flow cell undergo efficient DNA replication in Xenopus egg extract. This revealed that the two sister replisomes emanating from a single origin of replication need not remain physically coupled to function in replication. We further developed a new approach, called PhADE (photoactivation, diffusion, excitation) that can image single fluorescently labeled proteins in a crude-extract environment. Recently, we have begun to use this platform to tackle the mechanisms of homologous recombination, DNA interstrand cross-link repair, and non-homologous end joining.
In summary, by combining cell-free extracts, designer DNA templates, and single-molecule analysis, we are dissecting the function and dynamics of the vertebrate replisome and the consequences of replisome collision with DNA lesions. Our experiments will help illuminate the causes of genomic instability in hereditary cancer syndromes and thereby lay the foundation for a rational approach to cancer therapy.
Grants from the National Institutes of Health provided support for some of these projects.