If you're seeing this message, it means we're having trouble loading external resources on our website.

If you're behind a web filter, please make sure that the domains *.kastatic.org and *.kasandbox.org are unblocked.

Main content

DNA proofreading and repair

Mechanisms to correct errors during DNA replication and to repair DNA damage over the cell's lifetime.

Key points:

  • Cells have a variety of mechanisms to prevent mutations, or permanent changes in DNA sequence.
  • During DNA synthesis, most DNA polymerases "check their work," fixing the majority of mispaired bases in a process called proofreading.
  • Immediately after DNA synthesis, any remaining mispaired bases can be detected and replaced in a process called mismatch repair.
  • If DNA gets damaged, it can be repaired by various mechanisms, including chemical reversal, excision repair, and double-stranded break repair.


What does DNA have to do with cancer? Cancer occurs when cells divide in an uncontrolled way, ignoring normal "stop" signals and producing a tumor. This bad behavior is caused by accumulated mutations, or permanent sequence changes in the cells' DNA.
Replication errors and DNA damage are actually happening in the cells of our bodies all the time. In most cases, however, they don’t cause cancer, or even mutations. That’s because they are usually detected and fixed by DNA proofreading and repair mechanisms. Or, if the damage cannot be fixed, the cell will undergo programmed cell death (apoptosis) to avoid passing on the faulty DNA.
Mutations happen, and get passed on to daughter cells, only when these mechanisms fail. Cancer, in turn, develops only when multiple mutations in division-related genes accumulate in the same cell.
In this article, we’ll take a closer look at the mechanisms used by cells to correct replication errors and fix DNA damage, including:
  • Proofreading, which corrects errors during DNA replication
  • Mismatch repair, which fixes mispaired bases right after DNA replication
  • DNA damage repair pathways, which detect and correct damage throughout the cell cycle


DNA polymerases are the enzymes that build DNA in cells. During DNA replication (copying), most DNA polymerases can “check their work” with each base that they add. This process is called proofreading. If the polymerase detects that a wrong (incorrectly paired) nucleotide has been added, it will remove and replace the nucleotide right away, before continuing with DNA synthesis1.
  1. DNA polymerase adds a new base to the 3' end of the growing, new strand. (The template has a G, and the polymerase incorrectly adds a T rather than a C to the new strand.)
  2. Polymerase detects that the bases are mispaired.
  3. Polymerase uses 3' to 5' exonuclease activity to remove the incorrect T from the 3' end of the new strand.

Mismatch repair

Many errors are corrected by proofreading, but a few slip through. Mismatch repair happens right after new DNA has been made, and its job is to remove and replace mis-paired bases (ones that were not fixed during proofreading). Mismatch repair can also detect and correct small insertions and deletions that happen when the polymerases "slips," losing its footing on the template2.
How does mismatch repair work? First, a protein complex (group of proteins) recognizes and binds to the mispaired base. A second complex cuts the DNA near the mismatch, and more enzymes chop out the incorrect nucleotide and a surrounding patch of DNA. A DNA polymerase then replaces the missing section with correct nucleotides, and an enzyme called a DNA ligase seals the gap2.
Mismatch repair.
  1. A mismatch is detected in newly synthesized DNA. There is a G in the new strand paired with a T in the template (old) strand.
  2. The new DNA strand is cut, and a patch of DNA containing the mispaired nucleotide and its neighbors is removed.
  3. The missing patch is replaced with correct nucleotides by a DNA polymerase.
  4. A DNA ligase seals the remaining gap in the DNA backbone.
One thing you may wonder is how the proteins involved in DNA repair can tell "who's right" during mismatch repair. That is, when two bases are mispaired (like the G and T in the drawing above), which of the two should be removed and replaced?
In bacteria, original and newly made strands of DNA can be told apart by a feature called methylation state. An old DNA strand will have methyl (CH3) groups attached to some of its bases, while a newly made DNA strand will not yet have gotten its methyl group3.
In eukaryotes, the processes that allow the original strand to be identified in mismatch repair involve recognition of nicks (single-stranded breaks) that are found only in the newly synthesized DNA3.

DNA damage repair mechanisms

Bad things can happen to DNA at almost any point in a cell's lifetime, not just during replication. In fact, your DNA is getting damaged all the time by outside factors like UV light, chemicals, and X-rays—not to mention spontaneous chemical reactions that happen even without environmental insults!4
Fortunately, your cells have repair mechanisms to detect and correct many types of DNA damage. Repair processes that help fix damaged DNA include:
  • Direct reversal: Some DNA-damaging chemical reactions can be directly "undone" by enzymes in the cell.
  • Excision repair: Damage to one or a few bases of DNA is often fixed by removal (excision) and replacement of the damaged region. In base excision repair, just the damaged base is removed. In nucleotide excision repair, as in the mismatch repair we saw above, a patch of nucleotides is removed.
  • Double-stranded break repair: Two major pathways, non-homologous end joining and homologous recombination, are used to repair double-stranded breaks in DNA (that is, when an entire chromosome splits into two pieces).

Reversal of damage

In some cases, a cell can fix DNA damage simply by reversing the chemical reaction that caused it. To understand this, we need to realize that "DNA damage" often just involves an extra group of atoms getting attached to DNA through a chemical reaction.
For example, guanine (G) can undergo a reaction that attaches a methyl (CH3) group to an oxygen atom in the base. The methyl-bearing guanine, if not fixed, will pair with thymine (T) rather than cytosine (C) during DNA replication. Luckily, humans and many other organisms have an enzyme that can remove the methyl group, reversing the reaction and returning the base to normal5.
Methylation of guanine
A normal guanine base undergoes a reaction with a harmful chemical, causing a methyl (CH3) group to be added to the carbonyl O found on one of the rings of the base.
The methyl group can be removed from the damaged, methylated base by an enzyme found in the cell.
Diagram based on similar figure in Cooper 5.

Base excision repair

Base excision repair is a mechanism used to detect and remove certain types of damaged bases. A group of enzymes called glycosylases play a key role in base excision repair. Each glycosylase detects and removes a specific kind of damaged base.
For example, a chemical reaction called deamination can convert a cytosine base into uracil, a base typically found only in RNA. During DNA replication, uracil will pair with adenine rather than guanine (as it would if the base was still cytosine), so an uncorrected cytosine-to-uracil change can lead to a mutation5.
To prevent such mutations, a glycosylase from the base excision repair pathway detects and removes deaminated cytosines. Once the base has been removed, the "empty" piece of DNA backbone is also removed, and the gap is filled and sealed by other enzymes6.
Base excision repair of a deaminated cytosine.
  1. Deamination converts a cytosine base into a uracil. This results in a double helix in which a G in one strand is paired with a U in the other. The U was formerly a C, but was converted to U via deamination.
  2. The uracil is detected and removed, leaving a base-less nucleotide.
  3. The base-less nucleotide is removed, leaving a 1-nucleotide hole in the DNA backbone.
  4. The hole is filled with the right base by a DNA polymerase, and the gap is sealed by a ligase.

Nucleotide excision repair

Nucleotide excision repair is another pathway used to remove and replace damaged bases. Nucleotide excision repair detects and corrects types of damage that distort the DNA double helix. For instance, this pathway detects bases that have been modified with bulky chemical groups, like the ones that get attached to your DNA when it's exposed to chemicals in cigarette smoke7.
Nucleotide excision repair is also used to fix some types of damage caused by UV radiation, for instance, when you get a sunburn. UV radiation can make cytosine and thymine bases react with neighboring bases that are also Cs or Ts, forming bonds that distort the double helix and cause errors in DNA replication. The most common type of linkage, a thymine dimer, consists of two thymine bases that react with each other and become chemically linked8.
Nucleotide excision repair of a thymine dimer.
  1. UV radiation produces a thymine dimer. In a thymine dimer, two Ts that are next to each other in the same strand link up via a chemical reaction between the bases. This creates a distortion in the shape of the double helix.
  2. Once the dimer has been detected, the surrounding DNA is opened to form a bubble.
  3. Enzymes cut the damaged region (thymine dimer plus neighboring regions of same strand) out of the bubble.
  4. A DNA polymerase replaces the excised (cut-out) DNA, and a ligase seals the backbone.
In nucleotide excision repair, the damaged nucleotide(s) are removed along with a surrounding patch of DNA. In this process, a helicase (DNA-opening enzyme) cranks open the DNA to form a bubble, and DNA-cutting enzymes chop out the damaged part of the bubble. A DNA polymerase replaces the missing DNA, and a DNA ligase seals the gap in the backbone of the strand9.

Double-stranded break repair

Some types of environmental factors, such as high-energy radiation, can cause double-stranded breaks in DNA (splitting a chromosome in two). This is the kind of DNA damage linked with superhero origin stories in comic books, and with disasters like Chernobyl in real life.
Double-stranded breaks are dangerous because large segments of chromosomes, and the hundreds of genes they contain, may be lost if the break is not repaired. Two pathways involved in the repair of double-stranded DNA breaks are the non-homologous end joining and homologous recombination pathways.
In non-homologous end joining, the two broken ends of the chromosome are simply glued back together. This repair mechanism is “messy” and typically involves the loss, or sometimes addition, of a few nucleotides at the cut site. So, non-homologous end joining tends to produce a mutation, but this is better than the alternative (loss of an entire chromosome arm)10.
A double-stranded break may be repaired by non-homologous end joining. The chromosome is "glued" back together, usually with a small mutation at the break site.
Diagram based on similar diagram in Alberts et al.10
In homologous recombination, information from the homologous chromosome that matches the damaged one (or from a sister chromatid, if the DNA has been copied) is used to repair the break. In this process, the two homologous chromosomes come together, and the undamaged region of the homologue or chromatid is used as a template to replace the damaged region of the broken chromosome. Homologous recombination is “cleaner” than non-homologous end joining and does not usually cause mutations11.
The double-stranded break may be repaired by homologous recombination. The broken chromosome pairs up with its homologue. The damaged region is replaced via recombination, using sequences copied from the homologue.
Diagram based on similar diagram in Alberts et al.10

DNA proofreading and repair in human disease

Evidence for the importance of proofreading and repair mechanisms comes from human genetic disorders. In many cases, mutations in genes that encode proofreading and repair proteins are associated with heredity cancers (cancers that run in families). For example:
  • Hereditary nonpolyposis colorectal cancer (also called Lynch syndrome) is caused by mutations in genes encoding certain mismatch repair proteins12,13. Since mismatched bases are not repaired in the cells of people with this syndrome, mutations accumulate much more rapidly than in the cells of an unaffected person. This can lead to the development of tumors in the colon.
  • People with xeroderma pigmentosum are extremely sensitive to UV light. This condition is caused by mutations affecting the nucleotide excision repair pathway. When this pathway doesn't work, thymine dimers and other forms of UV damage can't be repaired. People with xeroderma pigmentosum develop severe sunburns from just a few minutes in the sun, and about half will get skin cancer by the age of 10 unless they avoid the sun14.

Want to join the conversation?