DNA Replication and Checkpoint Control in S Phase

Replicating DNA is fragile, and can break during the duplication process. In fact, broken chromosomes are often the source of DNA rearrangements and can change the genetic program of a cell. These changes can trigger a growth advantage in a single cell in your body, and when that cell continues to divide, tumors arise. Fortunately, our cells have defense mechanisms to shield us from these damaging events.

In the eukaryotic cell cycle, chromosome duplication occurs during "S phase" (the phase of DNA synthesis) and chromosome segregation occurs during "M phase" (the mitosis phase). During S phase, any problems with DNA replication trigger a ‘'checkpoint" — a cascade of signaling events that puts the phase on hold until the problem is resolved. The S phase checkpoint operates like a surveillance camera; we will explore how this camera works on the molecular level. The last 60 years of research in bacterial species (specifically, Escherichia coli) and fungal species (specifically, Saccharomyces cerevisiae), have continually demonstrated that several major processes during DNA replication are evolutionarily conserved from bacteria to higher eukaryotes.

Before delving into the intricacies of checkpoints, we must remind ourselves of the key molecules and processes of DNA replication. What happens to DNA when it is duplicated?

Amongst the array of proteins at the replication fork, DNA polymerases are central to the process of replication. These important enzymes can only add new nucleoside triphosphates onto an existing piece of DNA or RNA; they cannot synthesize DNA de novo (from scratch), for a given template. Another class of proteins fills this functional gap. Unlike DNA polymerases, RNA polymerases can synthesize RNA de novo, as long as a DNA template is available. This particular feature of de novo synthesis is similar to what happens during mRNA transcription.

Eukaryotic cells possess an enzyme complex that has RNA polymerase activity, but works in DNA replication. This unique enzyme complex is called DNA primase. Interestingly, this primase generates small 10-nucleotide-long RNA primers from a DNA template (the red portion of the Okazaki fragment in Figure 2). The RNA primers produced are later replaced by DNA, so that the newly-synthesized lagging strand is not a mixture of DNA and RNA, but consists exclusively of DNA. The chemical properties of DNA and RNA are quite different, and DNA is the preferred storage material for the genetic information of all cellular organisms, so this reinstallment of a continuous DNA strand is very important.

In prokaryotic cells, DNA primase is its own entity and works in a complex with the DNA helicase (Figure 2) (Alberts 2003; Langston & O'Donnell 2006). However, in eukaryotic cells DNA primase is associated with another polymerase, DNA polymerase-α | | | pol-α | | |, which initiates the leading strand and all Okazaki fragments (Pizzagalli, A. et al. 1988; Hubscher, Maga, & Spadari 2002). At present, we have no evidence that DNA primase binds to the DNA helicase in eukaryotic cells. But it is likely that some connector protein coordinates DNA unwinding and DNA synthesis initiation in eukaryotic cells.

Figure 2: Proteins at the Y-shaped DNA replication fork

These proteins are illustrated schematically in panel a of the figure below, but in reality, the fork is folded in three dimensions, producing a structure resembling that of the diagram in the inset b. Focusing on the schematic illustration in a, two DNA polymerase molecules are active at the fork at any one time. One moves continuously to produce the new daughter DNA molecule on the leading strand, whereas the other produces a long series of short Okazaki DNA fragments on the lagging strand. Both polymerases are anchored to their template by polymerase accessory proteins, in the form of a sliding clamp and a clamp loader. A DNA helicase, powered by ATP hydrolysis, propels itself rapidly along one of the template DNA strands (here the lagging strand), forcing open the DNA helix ahead of the replication fork. The helicase exposes the bases of the DNA helix for the leading-strand polymerase to copy. DNA topoisomerase enzymes facilitate DNA helix unwinding. In addition to the template, DNA polymerases need a pre-existing DNA or RNA chain end (a primer) onto which to add each nucleotide. For this reason, the lagging strand polymerase requires the action of a DNA primase enzyme before it can start each Okazaki fragment. The primase produces a very short RNA molecule (an RNA primer) at the 58 end of each Okazaki fragment onto which the DNA polymerase adds nucleotides. Finally, the single-stranded regions of DNA at the fork are covered by multiple copies of a single-strand DNA-binding protein, which hold the DNA template strands open with their bases exposed. In the folded fork structure shown in the inset, the lagging-strand DNA polymerase remains tied to the leading-strand DNA polymerase. This allows the lagging-strand polymerase to remain at the fork after it finishes the synthesis of each Okazaki fragment. As a result, this polymerase can be used over and over again to synthesize the large number of Okazaki fragments that are needed to produce a new DNA chain on the lagging strand. In addition to the above group of core proteins, other proteins (not shown) are needed for DNA replication. These include a set of initiator proteins to begin each new replication fork at a replication origin, an RNAseH enzyme to remove the RNA primers from the Okazaki fragments, and a DNA ligase to seal the adjacent Okazaki fragments together to form a continuous DNA strand.

© 2002 From Molecular Biology of the Cell, 4th Edition by Alberts et al. Reproduced with permission of Garland Science/Taylor & Francis LLC. All rights reserved.

After strand initiation, other DNA polymerases continue DNA elongation. In eukaryotic cells, these polymerases cooperate with a sliding clamp called proliferating cell nuclear antigen (PCNA). The regulation of PCNA is highly complex and important for DNA replication and repair (Moldovan, Pfander, & Jentsch 2007). There may be additional, yet undiscovered, parallel (or identical) mechanisms or proteins that coordinate DNA unwinding and DNA elongation. Observations in simpler model organisms strongly hint that eukaryotes too have a connecting mechanism that coordinates DNA helicase, and a DNA polymerase-a/DNA primase (pol-a/primase) complex.

How would you identify the protein that serves as a connector between DNA helicase and pol-a/primase? A simple yet often effective approach is to find proteins that directly bind to both enzymes. However, that requires us to understand the molecular architecture of DNA helicase.

In eukaryotes, the DNA helicase is comprised of a structural core and two regulatory subunits. The core, which contains the ATP hydrolysis activity, is a hexameric complex formed of the minichromosome maintenance proteins 2-7, called Mcm2-7 (Bochman & Schwacha 2008; Bochman & Schwacha 2009; Schwacha & Bell 2001). Mcm2-7 encircles dsDNA (Remus et al. 2009), but remains inactive until two additional regulatory subunits assemble onto it. Those factors are cell division cycle protein 45 (Cdc45) and GINS (Go, Ichi, Ni, and San; Japanese for "five, one, two, and three," which refers to the annotation of the genes that encode the complex). Scientists call this resulting functional DNA helicase a CMG complex (formed by Cdc45, Mcm2-7, GINS) (Moyer, Lewis, & Botchan 2006). In principle, any of these assembled components could be linked to pol-a/primase by a hypothetical connector protein. Scientists have actually identified two candidate connector proteins that directly bind to both helicase and primase: 1) Mcm10 (another Mcm protein that, despite its name, has no functional resemblance to any of the Mcm2-7 proteins) (Solomon et al. 1992.; Merchant et al. 1997) and 2) chromosome transmission fidelity protein 4 (Ctf4) (Kouprina et al. 1992). Specifically, both of these proteins interact with pol-a/primase (Fien et al. 2004; Ricke & Bielinsky 2004; Warren et al. 2009; Miles & Formosa 1992) and CMG complex subunits (Merchant et al. 1997; Gambus et al. 2009). In budding yeast, Mcm10 is essential for replication to occur. However, in these same cells DNA replication can function normally without Ctf4, which means that Ctf4 is not absolutely required (Kouprina et al. 1992). What about higher eukaryotes? Other experiments in human cells have shown that both proteins seem to be necessary, and work together during replication (Zhu, et al. 2007). Scientists are still actively investigating these complex mechanisms.

Why is coordination between DNA unwinding and synthesis important? What would happen if you lose this coordination? Because pol-a/primase always requires CMG function to create the ssDNA template, it could never surpass the DNA helicase (Figure 2b). Without a connecting link, the CMG complex could just "run off" and leave pol-a/primase behind. This would create long regions of vulnerable ssDNA. Therefore, the second rule in DNA replication is that DNA unwinding and DNA synthesis have to be coordinated.

Figure 3: Single-stranded DNA (ssDNA) gaps with a 5' primer end are formed during nucleic acid metabolism

© 2008 Nature Publishing Group Cimprich, K. A. & Cortez, D. ATR: an essential regulator of genome integrity. Nature Reviews Molecular Cell Biology 9, 616–627 (2008) doi:10.1038/nrm2450. All rights reserved.

What is the mechanism of a red flag, or danger signal that activates a checkpoint? How does it alert the cell? Scientists who have asked this question don't know the entire answer, but they have learned that RPA-coated ssDNA attracts a specific protein with a complicated name: the ataxia telangiectasia mutated and Rad3 related kinase, also known as ATR (Cimprich & Cortez 2008). ATR associates with RPA and activates its intrinsic kinase activity. This starts a that temporarily halts S phase progression. Therefore, ATR is also known as the S phase "checkpoint kinase."

ATR kinase acts in several ways to keep the replication process intact. There is evidence that ATR also stabilizes replication forks that contain ssDNA (Katou et al. 2003). How this happens remains largely unclear, but recent evidence suggests that ATR may affect the Mcm2-7 proteins, the inner core of the CMG helicase mentioned above (Cortez, Glick, & Elledge 2004; Yoo et al. 2004). One hypothesis is that phosphorylation of one or several of the Mcm2-7 subunits prevents the CMG complex from unwinding more and more DNA. This action effectively stops the process so that it can be repaired before proceeding. Currently, many researchers are trying to better understand the mechanisms of crosstalk between ATR and the replication machinery (Forsburg 2008; Bailis et al. 2008).

Together with a variety of other molecules, ATR and ATM kinases are key factors for the surveillance of DNA replication, and prevent chromosome breakage in dividing cells. However, during repair processes, chromosome fragments can be improperly joined together. Indeed, some scientists consider that such mistakes enable some degree of genetic evolution by creating new and different genetic sequences. Nevertheless, if even a single cell in our body makes a mistake and fuses DNA fragments to each other that are not supposed to be joined, the rearrangement can be sufficient to deregulate normal cell division. If multiple changes of this type accumulate, then this single cell can eventually turn into a tumor.

Given this understanding, would it be true that people who carry a mutation in the ATM, ATR, CHK1, or CHK2 genes have a higher risk of developing cancer? Yes. In these affected individuals, the cellular surveillance system described above is defective and no longer provides full protection from random events that affect DNA replication. For example, the name of the ATM protein derives from the affliction that results from a mutated ATM protein: ataxia telangiectasia. In this disease, patients suffer from motor and neurological problems, and they also have what is known as a genome instability syndrome that genetically predisposes them to developing cancer (Shiloh 2003). In addition, when scientists examine cells directly, the experimental inhibition of ATM, ATR, Chk1, Chk2, or the connector protein Mcm10 causes a very dramatic increase of DSBs (Paulsen et al. 2009; Chattopadhyay & Bielinsky 2007). With these observations, it may be possible to create new ideas for novel diagnostics and therapies for cancer that specifically track these potent molecules.

The process of DNA replication is highly conserved throughout evolution. Investigating the replication machinery in simple organisms has helped tremendously to understand how the process works in human cells. Major replication features in simpler organisms extend uniformly to eukaryotic organisms, and replication follows fundamental rules. During replication, complex interactions between signaling and repair proteins act to keep the process from going awry, despite random events that can cause interruption and failures. Discovering the exact repair mechanisms that help keep DNA intact during replication may help us understand the mechanisms of tumor growth, as well as develop strategies to detect or treat cancer.