DNA Lesions I

In addition to DNA polymerase errors, the human genome--even in a perfectly healthy cell--is under constant assault from endogenous (internal to the cell) and exogenous (external to the cell, or environmental) chemicals and other agents that alter its chemistry, break its backbone, and distort its helical structure.

In this post, we'll first consider what lesions are. We'll see that some are modest (e.g., a small chemical change to the base on a nucleotide) and some are more serious (e.g., a DNA-protein crosslink, or DPC, where a large protein becomes strongly bound to DNA). Most of this post, though, will be focused on discussing the two DNA repair pathways that correct most DNA lesions: base excision repair (BER) and nucleotide excision repair (NER).

I'll note here that the BER pathway is really two pathways: short-patch BER and long-patch BER. Long-patch BER, which repairs larger lesions and lesions that get through short-patch BER, is linked to the replication fork. There, it leverages many of the proteins involved in Okazaki fragment processing. However, short-patch BER (which repairs a single base) and NER (which repairs larger, helix-distorting lesions) occur well beyond the replication fork and throughout the cell cycle, not just during S-phase.

DNA Lesions

Chemical alterations to DNA are referred to as "lesions" to distinguish them from DNA polymerase "errors." A DNA lesion is any change to a DNA molecule that disrupts its normal chemical structure. In the case of substitutions, insertions and deletions, the chemical structure of the nucleotide molecules remains intact. And with ribonucleotide incorporation, the ribonucleotide also remains chemically correct. It's just somewhere it shouldn't be. None of these are lesions.

Most lesions (90-99%) are the result of endogenous rather than environmental insults. They're caused by the natural byproducts of cell metabolism and by the cell's aqueous environment. Many metabolic byproducts are reactive and chemically alter DNA, especially the base component of the nucleotide. They can also cause complete removal of a base from a nucleotide as well as single- and double-stranded breaks. Lesions are not rare. Even a healthy human cell accumulates tens of thousands of endogenous DNA lesions every day.

Causes of less frequent exogenously-caused DNA lesions (1-10% of the total) include UV radiation from the sun and reactive chemicals from environmental pollutants, industrial chemicals, dietary products, cigarette smoke and air pollution. UV exposure is responsible for a common kind of lesion that we'll be discussing later: an unwanted bond that forms between neighboring bases. Exogenous insults may be rarer, but they cause some of the most serious DNA lesions, which we'll discuss last.

Base Excision Repair (BER)

For the sake of comparison, consider RER. An unwanted ribonucleotide differs from a deoxyribonucleotide only by virtue of a single extra chemical group (scientifically speaking, an -OH, or hydroxyl group) attached to its ribose sugar. So, in effect, by replacing a ribonucleotide with a deoxyribonucleotide, RER fixes the sugar component of a nucleotide.

Base excision repair (BER), on the other hand, fixes not the sugar, but a damaged or abnormal base portion of a nucleotide--that is, the A, T, G and Cs. If left uncorrected, these can cause mutations during the next round of replication and also stall DNA polymerases. So the cell needs a way to remove lesions that arise on DNA bases.

To complicate things, many different small chemical lesions can arise on bases. I won't describe them; there are too many and I'd have to use words like "deamination," "alkylation," "oxidation," and "hydrolysis." Fortunately, all we need to be aware of is that the majority of base lesions involve small chemical groups being added to, removed from, or changed on the base part of a nucleotide.

Base lesions occur constantly cells. A single human cell experiences 20,000-50,000 de novo base lesion events per day! So, like RER, BER is a very active pathway. Also, like RER, BER embodies a "cut-and-patch" approach to DNA repair (as we'll see). In fact, it features two versions because, as mentioned, there are two versions of BER: short-patch BER (which replaces a single base) and long-patch BER (which replaces 2-10 bases). But unlike RER, BER is not exclusively tied to the replication fork or replication at all.

Short-patch BER

Let's start with short-patch BER. In short-patch BER, instead of replacing a section of DNA, just the one nucleotide with the base lesion is removed and replaced. Short-patch BER has no association with replication or the replisome. It takes place throughout the cell cycle and anywhere on the genome. It'san ongoing molecular spot repair system.

Short-patch BER begins with a family of proteins called DNA glycosylases. Because there are so many different kinds of base lesions, the cell needs an entire family of inspectors to find them. There are about 10-12 different DNA glycosylase detector proteins, each of which detects one or a small number of base lesions.

These different DNA glycosylases are constantly scanning the genome for damaged bases. The enzymes ride along a DNA strand--like a train on a train track--bending the DNA strand just a bit as they pass to test for the kind of local instability that would be characteristic of a base lesion.

If the DNA glycosylase finds a local instability, it flips that base into a special pocket on the enzyme. And, if the base's fit in the pocket is still consistent with it being a base lesion, then the DNA glycosylase cleaves the chemical bond between the base and the rest of the nucleotide, creating what's referred to as an "abasic site" (also known as an "AP site").*

Next, APE1 (AP endonuclease 1) recognizes the abasic site and cuts the DNA strand there creating a single strand nick 5' of the site. This provides a DNA polymerase--in the case of short-patch BER it will be a new one: DNA polymerase-beta--with a 3' end to extend. DNA polymerase beta actually performs two roles. It first removes the remaining portion of the abasic nucleotide--that is, the sugar and the phosphate. Then it adds a correct nucleotide using the undamaged strand as the template.

Finally, a scaffold protein called XRCC1 links DNA polymerase-beta with a DNA ligase we haven't come across yet: DNA ligase III. DNA ligase III connects the two ends of the still-disconnected single strand and the repair is complete.

Long-patch BER

Long-patch BER targets the same kinds of lesions as short-patch BER and, in some cases, lesions that short-patch BER failed to repair. But the long-patch BER pathway is quite different than the short-patch BER pathway. Long-patch BER not only removes a much larger patch of DNA, it also takes place at the replication fork and uses many of the same enzymes as RER and Okazaki fragment processing.

Long-patch BER still begins with a DNA glycosylase identifying a specific base lesion and nicking the single strand just 5' of that lesion. From that point on, though, long-patch BER is reminiscent of RER and Okazaki fragment. Starting at the nick, DNA polymerase delta or epsilon synthesizes 2-10 nucleotides, displacing the strand with the lesion. This creates a flap that's removed by FEN1 (just like Okazaki fragment processing). Once the flap is gone, DNA ligase I or III does its thing connecting the strand ends and the repair is complete.

The Base Excision Repair (BER) and Nucleotide Excision Repair (NER) pathways.

Nucleotide Excision Repair (NER)

Like BER and several of the repair pathways we've seen, nucleotide excision repair (NER) embodies a modified version of cut-and-patch repair. But while BER typically repairs small chemical lesions on bases, NER removes bulky, helix distorting lesions that affect entire nucleotides. In fact, NER repair involves the complete removal of a 24-32 nucleotide-long single-stranded DNA that's atttached to the lesion.

Two examples of the kinds of lesions targeted by NER are: (1) a "pyrimidine dimer," in which UV radiation causes two neighboring pyrimidine nucleotides (that is, Cs and/or Ts) to bind to each other, and (2) a large, bulky chemical adduct to a nucleotide caused by carcinogens in cigarette smoke. Both of these serious, helix-distorting lesions must be removed. The former would likely stall the DNA polymerase. The latter could even stall the CMG helicase.

In discussing NER, I'll focus mainly on its uniquenesses as a cut-and-patch repair pathway. NER's first unique aspect is its lesion detector, XPC-RAD23B, which patrols the genome constantly, looking for any kind of helix distortion. Importantly, the detector does not seek out any specific type of lesion. Most large ones will distort the double helix, forming XPC-RAD23B's target.

Once XPC-RAD23B identifies a distortion, it inserts an arm between the two DNA strands there. Then it flips out the two bases of the two nucleotides just across from the lesion, creating a small bubble of unwound double helix. At that point, it attaches itself to the unlesioned strand, identifying the lesioned strand for the NER repair enzymes.

XPC-RAD23B then initiates assembly of the full NER complex at the site. Most importantly for the next step, it recruits a large 10-subunit protein complex called TFIIH. TFIIH contains two helicases, XPD and XPB. The TFIIH complex will have a few roles. It will further unwind the DNA. It will verify that the lesion warrants NER. And it will assist in removing the lesion-containing single strand from the double helix..

Starting at the unwound bubble created by XPC-RAD23B, the XPB subunit unwinds about 20-30 nucleotides around the lesion, creating single-stranded DNA. Then the XPD subunit moves along the presumably lesioned single strand created by the helicase. If XPD pauses or halts, that indicates that the lesion is real and NER should proceed. If the helicase moves along the single strand without incident, TFIIH leaves and NER is aborted.

Assuming NER proceeds, the next two proteins to arrive will be endonucleases. These two cutting enzymes will generate single stranded nicks on the lesioned single strand on both the 5' side of the lesion about 5 nucleotides from the lesion (XPF-ERCC1 endonuclease does this) and about 25-30 nucleotides on the 3' side of the lesion (the job of XPG endonuclease).

Once these nicks, or incisions, are made to the lesioned strand, it still remains weakly bound to its complementary strand. Removing this oligonucleotide will be facilitated by RPA proteins that will coat the "good" strand, thus keeping the lesioned strand from reannealing to it. Also, as its last act, TFIIH's helicases will distrupt the binding of, and effectively force the dually-nicked lesioned strand to leave. Once it leaves, the cell degrades it.

The rest of NER takes the standard "cut-and-patch" approach. With the strand containing the lesion gone, DNA polymerase delta or epsilon arrives and, held in place by the PCNA sliding clamp, re-synthesize the now-missing strand using the "good" strand as the template (step 4). Then, either DNA ligase I or III will arrive to stitch the two strand ends together, restoring a continuous and now repaired double helix (step 5).