17. Building a DNA Copying Machine (1,497)

17. Assembling the Replisome

Our cell has now entered S phase. Across its genome, tens of thousands of starting points, or replication origins, have been prepared in advance. Each one has been “licensed”—approved for use.

Replication origin licensing is based on the presence of two MCM protein complexes lined up tail-to-tail at the origin in question. For the time being, they are inert. Their transformation into CMG helicases (the protein complexes that unwind the double helix to be copied) and the construction of two complete outward-facing DNA copying machines, or replisomes, around them is the topic of this chapter.

Assembling a replisome requires a precisely timed sequence of events that can be broken into four steps:

(1)  Transformation of both MCMs into complete but still inactive CMG helicases

(2)  Activation of the CMG helicases and positioning the DNA correctly within them.

(3)  Recruitment of various proteins that will stabilize and synchronize the replisome

(4)  Recruitment of the DNA polymerases—the enzymes that synthesize DNA.

Once these steps are complete, the cell will have constructed two fully functional replisomes led by two CMG helicases pointing in opposite directions away from the replication origin.

Let’s take a closer look at these steps—how the replisome is assembled.

Building the CMG Helicases

As the cell commits to DNA replication at the end of G2 and into S phase, a new wave of regulatory enzymes becomes active. These enzymes are called CDKs — cyclin-dependent kinases. Their job is to add small chemical tags called phosphates to specific proteins, flipping their molecular switches from “off” to “on” (or, in some cases, from “on” to “off”).

These phosphate signals accomplish two things at once.

First, CDK phosphorylations prevent re-licensing of replication origins by disabling or removing the proteins required to load new MCM complexes onto DNA in S phase. If origins could be re-licensed in S phase, the cell could mistakenly copy a section of the genome multiple times and create, in a word, a mess.

Second, CDK actions trigger the conversion of the two dormant MCM rings at the origin into active DNA-unwinding motors, or helicases—specifically, the human CMG helicase.

The transformation of an MCM ring into a CMG helicase involves the addition of the two other protein components: Cdc45 and GINS (“CMG” is an acronym for Cdc45-MCM-GINS).

But this transformation doesn’t happen with a snap of the fingers. Temporary scaffold proteins gather at the origin, forming a short-lived construction platform. Then multiple intermediary proteins recruit and attach the Cdc45 and GINS proteins to the MCM ring. Once these two additional protein species are added, the scaffolds leave.

What remains are two complete CMG helicase machines facing in opposite directions. But even now, they are not yet activated or moving. That will require the arrival of another protein, Mcm10. 

Activating the CMG helicase and re-positioning the DNA

Activation of the CMG helicase involves the transition to a protein complex that is beginning to unwind DNA. It occurs when the two tail-to-tail MCM components of the two CMG helicases separate, when the replication origin DNA is melted, and when each CMG helicase engages a single DNA strand and starts ATP-driven translocation (we’ll discuss this in detail in the next chapter). Activation establishes two outward-moving replication forks.

Mcm10 is a scaffolding protein required for this activation. It helps this delicate transition succeed by steadying the newly opened DNA and supporting the CMG helicase as it begins to move. In effect, it acts like a stabilizing hand during start-up, ensuring that the motor grips the correct strand and that a stable replication fork is successfully launched.

Now we have the leading strand protected inside the CMG helicase’s MCM ring and the lagging strand exposed on the outside of the ring. But this creates a problem. In the cell, single stranded DNA is susceptible to destruction—to being chopped up by proteins called nucleases. 

Single stranded DNA is also susceptible to the formation of secondary structures: regions where the DNA self-hybridizes based on standard A-T, G-C DNA base pairing rules. Secondary structures in the lagging strand impede replisome progression.

To address this, the cell calls into action proteins called RPAs (Replication Protein A). These smallish protein complexes coalesce around and coat the lagging strand, protecting it from nuclease digestion and from secondary structure formation. Since the other strand—the leading strand—is inside the helicase’s channel, it doesn’t require protection.

Let's take stock. We now have a replication origin with two functional CMG helicases beginning to move in opposite directions. The DNA located near the helicases has been partially unwound. The leading strand remains inside the helicase ring. The lagging strand is outside of the helicase and is protected by numerous RPA proteins.

As the CMG helicase begins to move, the next step is to fill out the replisome with proteins that provide structure and that make sure the activities of the helicase and the DNA polymerases (yet to arrive) are synchronized.

Coordinating DNA unwinding and DNA synthesis

When genome replication begins, the CMG helicase progresses down the double stranded DNA, separating it into two single strands. Protein enzymes called DNA polymerases (there are three, each with a specific role) will follow closely behind, synthesizing new strands off the two existing strands. These two activities—DNA strand separation and DNA polymerization—must be coordinated spatially and temporally. 

The next set of proteins to arrive play structural roles in the replisome. They help organize, stabilize and physically coordinate the replication machinery rather than catalyzing chemical reactions themselves. They basically maintain the architecture of what I’ll now refer to as the replication fork—the Y-shaped region where DNA is being unwound and copied.

The proteins that serve these structural and coordinating roles are referred to in the aggregate as the fork protection complex, or FPC. As we'll see later, the FPC also stabilizes stalled forks, preventing the CMG helicase from running too far ahead of the polymerase if a fork encounters damage or stress.

Recruitment of the DNA polymerases

Finally, the enzymes that will synthesize the DNA—the DNA polymerases—arrive at the replisome. In fact, there are three different polymerases with different roles. They are DNA polymerases alpha-primase, epsilon and delta. 

The first of these enzymes—DNA polymerase alpha-primase—synthesizes short starter segments of DNA called primers. The other two DNA polymerases can’t start synthesizing without a roughly 30- to 40-nucleotide stretch of DNA laid down by this first DNA polymerase. In other words, the other two DNA polymerases can add to, or extend, a pre-existing DNA strand. But they can’t initiate synthesis without the primers generated by DNA polymerase alpha-primase.

Next, enter the other two so-called replicative DNA polymerases that will carry out the bulk of DNA synthesis during genome replication. DNA polymerase epsilon synthesizes continuously on one of the strands, which I will now refer to as the leading strand. DNA polymerase delta synthesizes shorter segments on the opposite strand, the lagging strand. (We’ll explore the leading and lagging strands and why this asymmetry exists in the next few chapters.)

Next, to keep the DNA polymerases from repeatedly falling off the DNA during synthesis, the cell deploys one of the most elegant machines in molecular biology: a ring-shaped sliding clamp. Without the sliding clamp, DNA polymerases fall off the DNA frequently. Molecular biologists would say that they have low “processivity.” With their partners, the sliding clamps, the two replicative polymerases stay on the DNA for a long time, greatly increasing their processivity.

The sliding clamp is called PCNA (Proliferating Cell Nuclear Antigen)—a name with only historical significance. PCNA also has its own dedicated loader protein. The loader literally opens the PCNA sliding clamp ring, inserts the double stranded DNA inside of it, and then closes the ring around the DNA like a bracelet.

On the leading strand, DNA polymerase epsilon tethers directly to the CMG helicase at the front of the replisome and the sliding clamp encircles and rides the DNA immediately behind it. On the lagging strand, DNA polymerase delta doesn’t attach to the CMG helicase. Rather, it attaches directly to its own PCNA.

Two machines, not one

Because initially two MCM rings were positioned back-to-back at the replication origin, two complete replisomes will form at each origin. These replisomes will move away from each other in opposite directions, copying the genome bidirectionally.

As we’ve seen, these replisomes are far from single enzymes. I haven’t come close to mentioning every replisome protein in this chapter. In fact, about 30-40 different protein species make up an active replisome. Building one is a carefully staged molecular construction project—one that takes place tens of thousands of times during just one cell division.

If you don't yet have a clear picture of the replisome, I encourage you to look at the figure above. And if you're not yet clear on leading strand versus lagging strand DNA synthesis, don't worry. I haven't explained it yet. I've only set the stage for its explanation.