Lagging Behind

DNA polymerases can only synthesize DNA in the 5' to 3' direction. That is an absolute rule that's determined by the biochemistry of nucleotide addition. Given the antiparallel nature of DNA, a corollary is that DNA polymerases can only synthesize off template DNA that is oppositely oriented in the 3' to 5' direction.

Our leading strand exited the MCM2-7 ring of the CMG helicase 3' end first. So the DNA strand that will be synthesized using that strand as a template will be synthesized 5' to 3'--that is, toward the CMG helicase and in the same direction that the replisome moves.

In contrast, the lagging strand that was shunted out of the MCM2-7 hexamer led with its 5' end. That means that its replication must be performed in reverse: in the direction opposite replisome movement. There's no other option given this limitation of DNA polymerases.

In a nutshell, lagging strand synthesis is going to involve many short opposite direction syntheses that will create complementary fragments roughly 100-200 nucleotides long. These are called Okazaki fragments. These fragments will then be stitched together post synthesis to form a continuous strand complementary to the lagging strand. Let's take a closer look at this amazing process.

First, despite going in reverse, essentially all the molecular actors who starred in leading strand synthesis will star again in lagging strand synthesis. DNA polymerase alpha-primase will make the primers. Like leading strand synthesis, lagging strand synthesis will require a more accurate and processive DNA polymerase--DNA polymerase delta--to extend the primers. And, like DNA polymerase epsilon, DNA polymerase delta will require a PCNA sliding clamp and its RFC clamp loader to achieve acceptable speed and processivity.

Because the lagging strand was shunted to the outside of the MCM2-7 component of the CMG helicase, it will be exposed to enzymes that degrade single-stranded DNA. It will also be susceptible to the formation of secondary structures that can impede DNA synthesis. So the first step in lagging strand synthesis (as with leading strand synthesis) will be to protect the single-stranded DNA with RFA coating proteins.

Once the strand is protected, the next step will be the construction of the first primer by DNA polymerase alpha-primase, which remains tethered to the CMG helicase via the flexible AND-1 protein. I'll remind the reader here that polymerization by DNA polymerase alpha-primase is not very accurate compared to polymerization by DNA polymerases epsilon and delta. Thus, the DNA stretch of the hybrid primer might well contain errors. The cell has a way to deal with this, as we'll see.

As in leading strand synthesis, a PCNA sliding clamp will be loaded by the RFC loader onto the lagging strand right where the first primer ends. The PCNA sliding clamp will then recruit DNA polymerase delta, which will attach itself to the sliding clamp via its PIP-box. With the DNA polymerase and sliding clamp in place, replication will proceed in the backward direction, away from the CMG helicase.

Lagging Strand Synthesis.

But remember: while the first primer is being extended, the replisome is continuing to move forward at a clip of 50 nucleotides per second. So a second primer must quickly be built upstream of the first one--that is, closer to the moving helicase but, again, facing backwards. Another sliding clamp will then be quickly loaded onto this second primer and a second DNA polymerase delta will attach itself to this second sliding clamp. Together, they extend off the second primer.

Generally speaking, this is how lagging strand replication works. As the helicase moves forward, an ongoing series of new primers are constructed upstream, near the helicase, that extend backward, away from the helicase. Each will eventually run into the previous primer. So lagging strand synthesis consists of the construction of a series of Okazaki fragments that are subsequently spliced together to create one continuous DNA strand. Tricky!

As just mentioned, the strand being synthesized using the second primer is pretty quickly going to reach the first nucleotide of the first primer. But it doesn't crash into it. Instead, it displaces it. It pushes it out of the way. It then continues to synthesize below the primer. This creates a primer flap where the deoxyribonucleotides that have just been added by DNA polymerase delta have displaced the first primer.

The order of what happens here is still subject to debate, but I'll assume that before DNA polymerase delta reaches the previous primer and creates the flap, the RNA portion of the primer is removed via enzymatic degradation--that is, all but the last ribonucleotide. This is accomplished by RNase H2. RNase H2 is an endonuclease that only works on RNA that's hybridized to DNA, which is the case with the RNA portion of the primer. Because the primer is multiple ribonucleotides long, RNase H2 must make many endonucleolytic incisions to degrade the primer.

But, as mentioned, RNase H2 isn't able to remove the very last ribonucleotide from the primer. So what remains following RNase H2 endonucleolytic digestion is a primer consisting of mostly DNA but, at the very beginning of the primer, one RNA nucleotide, or ribonucleotide.

With the RNA portion of the primer almost completely degraded by RNase H2, when DNA polymerase delta reaches the nearly RNA-less primer it displaces it and continues to synthesize underneath it, generating a flap consisting mainly of DNA deoxyribonucleotides plus that one pesky ribonucleotide.

Enter now an enzyme called FEN1 (Flap Endonuclease 1) that cuts single-stranded flaps right where they meet double-stranded DNA. In what's referred to as the idling cycle, DNA polymerase delta and FEN1 are going to respectively and repeatedly perform: (1) strand displacement that creates a flap, and then (2) flap cleavage.

The polymerase synthesizes, creating a flap. The flap is cut. The polymerase progresses further. A new flap is generated. That new flap is cut. This happens repeatedly until the primer DNA is removed.

By the end of the idling cycle, we are left with new high quality DNA replacing the lower quality DNA portion of the original primer. That is, new DNA synthesized by the more accurate DNA polymerase delta will replace the less accurate primer DNA synthesized by DNA polymerase alpha-primase.

But once the idling cycle has finished and we're left with (hopefully) no errors in the primer sequence, there will still be a break in the DNA where the last flap was cut off by FEN1. The two Okazaki fragments have not yet been stitched together. Assigned to that task is another enzyme: DNA ligase 1. It arrives to stitch (the scientific term is "ligate") the two fragments together, creating one continuous stretch of new DNA. And that completes the process.

That's lagging strand synthesis. It is a frenetic process. A primer is constructed. It's then extended in the reverse direction to create an Okazaki fragment. That fragment eventually runs into the previous primer. DNase H2 has already degraded all but one nucleotide of the RNA in that primer. But when the polymerase runs into the remaining mostly-DNA portion of the primer, it displaces it, creating a flap that's then cut off by FEN1. This flap generation and flap cleavage repeats. Eventually, once all the primer DNA has been replaced with new DNA, the two abutting DNA strands are stitched together by DNA ligase 1.

When a single cell divides, it replicates it's genome so that there will be one for each of the two daughter cells. Doing a little math using the size of an Okazaki fragment and the size of the human genome as a whole we can calulate that over the course of a roughly eight hour S phase of one cell cycle, the cell generates and stitches together about 20 million Okazaki fragments! "Frenetic" is the right descriptor!

• RFA proteins are often called "SSBs" for "single-stranded binding" proteins. This is the accurate term for the equivalent proteins in E. coli. The nomenclature from bacteria seems to stick around, probably because it is so descriptive.

• A little about the nomenclature of RNase H2. Enzymes with names that end in "-ase" are always enzymes that degrade other molecules. Thus, "RNase" is the name given to enzymes that degrade RNA.