18. Diving Deeper

Let's take a closer look at two amazing molecular machines found in the replisome: the CMG helicase and the PCNA sliding clamp. Both are multi-protein complexes arranged as rings that surround DNA. But they have different jobs, different shapes, and interact with the DNA they surround in different ways.

Quickly comparing them, the job of the CMG helicase is to propel the replisome forward along double-stranded DNA and at the same time separate that double-stranded DNA into two single strands. The job of PCNA (there are two per replisome) is to clamp the DNA polymerase to which it is attached to the DNA to improve the enzyme's performance. PCNA sliding clamps also serve as central nodes of attachment for other replisome proteins that come and go as needed.

The CMG helicase--and I'm focused on the MCM2-7 component of the helicase--is long and barrel-shaped, while the PCNA sliding clamp is shaped more like a flat disc or a washer. And, finally, the barrel-shaped CMG helicase has a motor that actively propels it along the DNA while the PCNA sliding clamp glides passively along DNA, pulled along by the DNA polymerase to which it is tethered.

CMG Helicase

Let's take a closer look at how CMG helicase accomplishes propulsion and DNA strand separation. Again, I'll be focusing on the MCM2-7 component of the helicase. It contains both the motor that propels the helicase along the DNA and the mechanism that splits the double helix into two single strands. We'll dissect at both. The Cdc45 and GINS components of the CMG helicase play critical, but supporting roles.

Recall that MCM2-7 is an elongated hexameric (six protein) ring. The six subunits are similar but not identical. Each has a top portion (the N-terminal domain) and a bottom portion (the C-terminal domain).* When the six MCM subunits are arranged as a ring, their N-terminal "heads" create an internal channel and their C-terminal "bodies" continue that channel all the way through the barrel.

Let me first provide a big picture of how the MCM2-7 portion of helicase works. There is an opening between the six N-terminal heads where double stranded DNA is going to enter the internal channel. Once inside the N-terminal channel, the six proteins are configured such that they are able to separate the strands. I'll discuss how in a moment. It is the C-terminal bodies of the six proteins that are going to make the helicase move, or "translocate," along the DNA. Let's look at the mechanism of strand separation first.

Double stranded DNA enters the N-terminal channel at the top of the barrel. Soon upon entering, six looping protrusions from the six subunits narrow the channel such that it can only fit one strand of DNA, not two. This will be the leading strand. The other lagging strand is blocked from entering. Instead, it is shunted to the outside of the MCM2-7 hexamer.

Thus, strand separation is based on "steric exclusion." This is a fancy way of saying that the channel becomes too small to fit both strands. The word "steric" in biochemistry refers to the shape or size of a molecule. But steric exclusion only works if the double stranded DNA is being forced through the central channel. Here's how that occurs.

CMG Helicase's MCM2-7 motor. Double stranded DNA enters the MCM2-7 hexamer ring through an upper channel. Immediately, loops from N-terminal domains of the proteins narrow the channel, shunting the lagging strand to the outside of the ring. The leading strand is pulled through by MCM2-7's motor: a helical arrangement of ATP-powered, positively charged loops that use a sequential, downward ratcheting motion to pull the DNA through..

MCM2-7 has a motor located on the inside of the channel created by the six lower C-terminal domains. There are two important components to these motor domains. First, if we were to look inside the channel, we would see six positively-charged loops sticking out from each of the six proteins, arranged in what I'll liken to a helical staircase. In other words, the loop sticking out from a each of the six proteins is a bit lower inside the channel than the loop sticking out from its immediately preceding neighbor. The result is a helical staircase of protruding loops.

Second, all six C-terminal domains contain ATP binding sites. These are required to get the loops to move, or pull the DNA. Recall that the cleavage of the last phosphate of the three-phosphate ATP molecule provides a burst of energy that can make molecules bend and shift and, in a word, do work. ATP provides the energy to run the CMG helicase motor.

DNA molecules are negatively-charged. In contrast, the six loops in the C-terminal portion of the channel are positively charged. This is based on the specific amino acids that make up the loops. The mechanism of translocation takes advantage of the fact that the positively-charged loops and the negatively charged phosphates on the DNA backbone attract each other. As the single leading strand enters the C-terminal portion of the channel, the loops transiently attract the DNA strand one after another to ratchet the single strand through the channel in a "hand-over-hand" motion.

In other words, in order, each loop grabs, pulls and then hands off the single leading strand of DNA to the next loop. This ratcheting action pulls the double-stranded DNA into the upper portion of the channel, propelling the CMG helicase along the double helix.

One more thing: the movement of the helicase along the DNA tracks the helical pitch of the double-stranded DNA that first enters the enzyme in a nut-and-bolt fashion. Think of the helicase as the nut and the double-stranded DNA entering it as the bolt--the threads of the bolt being the double helix itself. As the helicase moves along the DNA at 50 nucleotides per second, it rotates like a nut around its double helix DNA bolt.

As mentioned, the C-terminal loops pull the single stranded DNA in the C-terminal section of the helicase due to ATP hydrolysis in its six C-domains. When an ATP molecule binds to its binding site in one of the six protomers, it causes a conformational change that puts the positively-charged loop into the "up" position where it can attract and transiently attach to the negatively charged DNA. Then, when that MCM2-7 subunit cleaves the ATP molecule into ADP and Pi, it cause its loop to sweep down, pulling the single stranded DNA with it. At this point, the first loop hands the DNA off to the next loop on the helical staircase and the process continues. After six pulls, the process restarts from the top-most loop.

Picture it. The first loop ratchets the DNA down by about one nucleotide. Then it hands the strand off to the nearest neighbor loop, which ratchets the strand down by another nucleotide, etc. So it is the coordinated, sequential action of the C-terminal loops on the single-stranded DNA that pulls the leading strand through the channel, moves the CMG helicase along the double stranded DNA, and separates the strands at the narrowing of the channel in the N-terminal section of the barrel.

During active human genome replication in the S phase of just one cell in your body, about 50,000 CMG helicases are using this rapid rotary ratcheting motion to move along DNA at a clip of about 50 nucleotides per second, separating the DNA into single strands that are immediately replicated by the DNA polymerases following close behind.

PCNA Sliding Clamp a and the Clamp Loader

The PCNA sliding clamp is a homotrimeric flat-ish ring that, like the CMG helicase, encircles DNA. "Homotrimeric" means the ring is made of three identical protein subunits. Unlike the CMG helicase, though, the sliding clamp moves passively along DNA. It has no motor. On both the leading and lagging strands, the clamp is attached to a DNA polymerase. As the polymerase synthesizes DNA, it actively moves along the strand being replicated. The PCNA sliding clamp is tethered to a polymerase, so it moves, too.

As I just mentioned, there are two PCNAs in a replisome. One tethers to DNA polymerase epsilon, following behind it as it replicates the leading strand. The other tethers to DNA polymerase delta and is pulled along as that polymerase works on the lagging strand.

They're called clamps because they literally clamp DNA polymerases to the DNA they are replicating. Clamping is required for high DNA polymerase speed and processivity. Recall that processivity is defined as the number of nucleotides a DNA polymerase can synthesize before it falls off the single strand that it's replicating. For example, without PCNA, DNA polymerase epsilon will synthesize tens to 100 nucleotides of the leading strand before it falls off. But when clamped by PCNA, the processivity of DNA polymerase epsilon increases to thousands to tens of thousands of nucleotides. That's a 100-1,000-fold increase!

In addition to enhancing DNA polymerase performance, PCNA also serves as a kind of replisome "toolbelt"in that it serves as a central node of attraction for other replisome proteins. I'll introduce some of these other proteins when we discuss leading and lagging strand synthesis.

To make the toolbelt work, each of PCNA's three ring proteins contains what's called a PIP-box (for PCNA Interacting Protein box). Other proteins needed at the replisome have hooks that specifically latch onto PIP boxes, tethering the proteins near the point of DNA replication. So PCNA is essential both in DNA polymerization and also as an organizational hub within the replisome.

At this point you might be wondering how a closed ring comes to surround a long strand of DNA. It is through the action of another protein complex: the PCNA clamp loader, which is formally called RFC (for Replication Factor C). RFC's sole job is to grab a free floating PCNA sliding clamp ring, open it up, place it around a strand of DNA, and then re-close the ring. RFC needs energy to accomplish this. The energy, not surprisingly, comes from the attachment and then the hydrolysis of ATP.

The RFC clamp loader is a heteropentamer complex (made up of five different proteins) with extended domains in some of the proteins that act as arms. All five subunits have ATP binding sites. When an ATP binds to RFC, it causes a conformation change that in turn causes it to both grab a PCNA ring and open it. RFC then threads DNA--specifically DNA at the the primer-template junction where DNA polymerization will begin--into the PCNA ring. Upon ATP hydrolysis the RFC clamp loader releases the PCNA ring and the ring snaps closed. With the clamp loader gone, DNA polymerase can now take its place attached to the sliding clamp.

Clamp Loader. The clamp and clamp loader start as separate protein complexes. When the clamp loader binds ATP, it causes it to grab a clamp ring and open it up. The clamp loader then threads DNA at the primer-template junction into the ring. Upon ATP hydrolysis (formation of ADP and Pi), the clamp loader disengages and the clamp closes shut. With the clamp loader gone, DNA polymerase can get into position next to the clamp.

So the cell not only requires PCNA sliding clamps, it also requires the RFC clamp loaders. We'll see in lagging strand synthesis that the clamp loader can be a very busy molecule, constantly opening, threading and closing PCNA sliding clamp rings one after another as synthesis progresses.

Let's leave it there for now. Over the past few posts I've been referring to leading strand synthesis and lagging strand synthesis. It's time to look at both in more detail.