DNA Lesions II

In the last post, I introduced the cell's two main DNA lesion repair pathways--base excision repair (BER) and nucleotide excision repair (NER). Both operate beyond the replication fork during all phases of the cell cycle.

But what happens if a lesion escapes the BER and NER pathways and finds itself in front of a moving replication fork? Most lesions can't get past the DNA polymerase. And some very large lesions can't get past the CMG helicase.

Our focus here will be the majority of small- to medium-sized lesions that stall the DNA polymerase but not the helicase. These are usually handled by one of two quick lesion tolerance pathways (rather than repair pathways). Lesion tolerance pathways allow lesions to be bypassed now and fixed later. This strategy allows replication to proceed on schedule.

Replication Fork Stalling

Picture a replication fork humming along a double helix copying DNA at 50 nucleotides per second. Suddenly, there's a lesion in its path. Let's assume it's a chemical alteration to a base or a bond that's formed between neighboring pyrimidine bases. These are the types of lesions that stall a DNA polymerase but not the helicase.

When the DNA polymerase does stall at a lesion, the CMG helicase keeps moving and unwinding DNA. This generates two stretches of single-stranded DNA (ssDNA)--the leading strand and the lagging strand--hundreds to thousands of nucleotides long between the advancing helicase and the stalled polymerase.

Immediately, Replication Protein A (RPA) proteins arrive to coat and protect the exposed single strands. RPAs protect ssDNA against nucleases that destroy it and inhibit ssDNA secondary structures that can also form and stall DNA polymerases.

In addition, long stretches of RPA-coated single-stranded DNA (RPA-ssDNA) represent a very important signal to the cell. It communicates that a DNA lesion has stalled at the DNA polymerase and must be dealt with. This distress signal is powerfully agnostic with respect to the type of lesion. Any polymerase-blocking lesion will produce a long stretch of RPA-ssDNA. Large amounts of RPA-ssDNA are the cell's universal signal for replication stress.

At this point, the cell knows that a lesions is present (due to the presence of RPA-ssDNA) but it doesn't know the severity of the lesion and thus how challenging the repair or other response must be. Thus, the cell takes a staged approach to coping with lesions. It starts with easier responses but escalates its response in terms of both energy consumption and risk to itself if initial attempts fail.

The cell's response to RPA-ssDNA has a local component focused directly on addressing the lesion and a global component that communicates to the cell that there might be a larger, cell-wide problem with replication. This communication is used at the intra-S-phase cell cycle checkpoint to decide whether to continue genome replication or pause it. The local and global responses do overlap, but it's helpful to think in those terms.

Back to RPA-ssDNA. In all human cells, a heterodimeric protein complex called ATR-ATRIP is constantly scanning the nucleus for just that: RPA-ssDNA. When it finds some, its ATRIP subunit attaches to an RPA protein, placing the all-important (but not yet activated) ATR kinase component of the heterodimer near the site of the lesion. Soon the RPA-ssDNA will become, itself, coated with many ATR-ATRIP molecules.

Next, a ring-shaped protein complex called 9-1-1 is loaded onto the DNA near the lesion (and thus near the now ATR-ATRIP-coated RPA-ssDNA) by a specific ring loader. The 9-1-1 ring will serve as a general platform for checkpoint activation, if necessary. Its first job, though, is to recruit a protein called TOPBP1. TOPBP1 binds to the 9-1-1 ring but also has an ATR activation domain (AAD) that turns on the ATR kinase subunits of some of the ATR-ATRIPs attached to the RPA-ssDNA.

Once activated, ATR kinase--which remains tethered to RPA-ssDNA near the lesion--phosphorylates a key checkpoint protein kinase called CHK1, whose job it is to roam the nucleoplasm phosphorylating and thus activating or inactivating dozens of other proteins as part of the ATR kinase's distress signal. But I want to pause the CHK1 story here because, while critical for more serious lesions, CHK1 signaling will generally not be needed for the kind of modest lesions that we're focused on in this post.

Translesion Synthesis (TLS)

As I said, Translesion synthesis (TLS) (like repriming), is a lesion tolerance solution rather than a lesion repair solution. TLS and repriming both instantiate the notion that, during the 8-hour S phase of the cell cycle, the cell's highest priority is completing genome replication on time, even if that means delaying some DNA repairs.

With TLS, a new DNA polymerase--not a high-fidelity replicative polymerase like DNA polymerase epsilon, but one of the much lower fidelity "Y family" of DNA polymerases--is recruited to replace DNA polymerase epsilon. These new polymerases allow passage of many small to medium-sized DNA lesions. Their active sites are larger than that of DNA polymerase epsilon to allow lesions to get through. When they do, the Y-family polymerases also add one or a few bases across from the lesion. So while they're adept at letting lesions bypass, the Y-family polymerases are also more error prone. I'll quantify this in a moment.

There are four Y-family DNA polymerases: eta, iota, kappa, and one called REV1. Each accomodates specific kinds of distorted or bulky bases. For example, DNA polymerase eta deals with UV-induced pyrimidine dimers as we discussed in a previous post (i.e., when two neighboring pyrimidines--Cs and Ts--become bound together). Each Y-family DNA polymerase allows certain kinds of lesions to pass through unrepaired, to be repaired later. The Y-family polymerases are also called "TLS polymerases" or "bypass polymerases."

Translesion synthesis (TLS). In TLS a Y-family, or TLS polymerase replaces the normal replicative polymerase in order to bypass a DNA lesion

How does a cell know when a TLS polymerase is needed? It uses the universal replication distress signal: RPA-ssDNA! Like ATR-ATRIP, the protein complex RAD6-RAD18 is constantly scanning for RPA-ssDNA and rallies to it once it finds some. Once there, it attaches a specific chemical group (a ubiquitin molecule) on a specific amino acid of the PCNA sliding clamp. PCNA--which is located near the stalled polymerase--is now "monoubiquitinated." A mouthful. But this one ubiquitin attached to lysine 164 of the PCNA screams to the cell "bring on a TLS polymerase!"

Many lesions are bypassed via TLS but it carries a risk. Because the Y-family polymerases have low fidelity, the nucleotides across from the bypassed lesion may be wrong. DNA polymerase epsilon has an error rate of one error every 10 million to 100 million nucleotides. The TLS polymerases vary, but typically make an error once every 100-10,000 nucleotides. Basically, the cell says, "that's ok." It trades a few mutations for completing genome synthesis during the 8-hour S-phase of the cell cycle.

Once the TLS polymerase passes the lesion, it begins synthesizing off of a normal template strand. TLS polymerases aren't good at this. Their active sites don't fit unlesioned template strands as well as DNA polymerase epsilon's active site does. Thus, catalysis slows and it becomes easy for the enzyme with the better fit to supplant the TLS polymerase. Also, an enzyme complex called USP1-UAF1 arrives to de-monoubiquitinate PCNA, leaving it with little affinity for Y-family polymerases.

Thus, DNA polymerase epsilon replaces the transiently-needed TLS polymerase and normal replication restarts. The lesion remains, though. It will have to be repaired later. And the accuracy of the few nucleotides across from the lesion isn't guaranteed.

Repriming

The cell's other DNA lesion tolerance pathway is repriming. This is almost as simple as it sounds. Once DNA polymerase epsilon reaches a lesion, it stalls. Then yet another DNA polymerase called PRIMPOL arrives. It, too, is attracted to the universal replication distress signal: RPA-ssDNA.

PRIMPOL is a special kind of DNA polymerase in that it can synthesize its own primer. It doesn't need a 3' end to begin synthesizing. There is only one other human DNA polymerase that can do this and we've come across it: DNA polymerase alpha-primase.

PRIMPOL synthesizes an 8-10 nucleotide hybrid RNA-DNA primer about 100-1,000 nucleotides in front of the lesion. It sometimes extends its primer a short distance. Then, it leaves and RFC loads a PCNA sliding clamp. The PCNA partners with a DNA polymerase epsilon molecule and leading strand synthesis restarts, leaving that 100-1,000 nucleotide single-stranded gap in its wake.

This gap must be filled or the cell risks DNA breaks and genome instability. It will be filled later either by TLS or by another more complex repair pathway that we'll cover in a couple posts called homologous recombination (HR). These two pathways address the gap but not the lesion. The lesion will be repaired by BER or NER.

So TLS and repriming are "first responses" to a lesion that has stalled a DNA polymerase. But when I digressed to describe TLS and repriming, we were in the middle of describing CHK1 phosphorylation and how the cell can escalate its response to a lesion. That's where we'll pick things up in the next post.