Figure 4.3. Mismatch repair of DNA.
We best understand this repair process in E. coli . The bacterium has an enzyme called Dam methylase that adds a ‐CH3 group, called a methyl group, onto the A of the sequence 5′ GATC3′. This sequence occurs very frequently in DNA, about once every 256 bp. The methylation of DNA happens very soon after a DNA strand has been replicated. However, for a short time during replication the double‐stranded DNA molecule will have one strand methylated (the parental strand) and one strand not methylated (the daughter strand). The DNA molecule is said to be hemi‐methylated (half methylated). Because the newly synthesized strand has not yet been methylated the cell knows that if a mismatch in base pairing has occurred between the two strands it is the nonmethylated, newly synthesized, strand that must carry the mistake.
A protein called MutH binds on the newly synthesized strand at a site opposite a methylated A in the template strand. If there is no mismatched base pair nearby then MutH does nothing. However, if two other proteins called MutL and MutS have detected a mismatched base pair then MutH, which is an endonuclease, is activated and nicks (cleaves a phosphodiester bond between two nucleotides in) the unmethylated newly synthesized strand. This allows a stretch of DNA containing the mismatched base pair to be removed. Two different proteins are involved in removing the stretch of DNA. If MutH nicks the DNA 5′ to the mismatch (Figure 4.3a), then exonuclease VII degrades the DNA strand in the 5′ to 3′ direction. However, if MutH nicks the DNA 3′ to the mismatch (Figure 4.3b), then the DNA strand is removed by exonuclease I in the 3′ to 5′ direction. In either case, the gap in the daughter strand is then replaced by DNA polymerase III.
Deoxyribonucleic acid can be damaged by a number of agents, which include oxygen, water, naturally occurring chemicals in our diet, and radiation. Because damage to DNA can change the sequence of bases, a cell must be able to repair alterations in the DNA code if it is to survive and pass on the DNA database unaltered to its daughter cells.
Spontaneous and Chemically Induced Base Changes
The most common damage suffered by a DNA molecule is depurination – the loss of an adenine or guanine because the bond between the purine base and the deoxyribose sugar to which it is attached spontaneously hydrolyzes (Figure 4.4). Within each human cell about 5000–10 000 depurinations occur every day.
Deamination is a less frequent event; it happens about 100 times a day in every human cell. Collision of H3O+ ions with the bond linking the amino group to carbon number 4 in cytosine sets off a spontaneous deamination that produces uracil (Figure 4.4). Cytosine base pairs with guanine, whereas uracil pairs with adenine. If this change were not corrected, then a CG base pair would mutate to a UA base pair the next time the DNA strand was replicated, introducing a mutation at this position in one of the two copies of the DNA double helix that are obtained post‐replication.
Ultraviolet light or chemical carcinogens such as benzopyrene, present in cigarette smoke, can also disrupt the structure of DNA. The absorption of ultraviolet light can cause two adjacent thymine residues to link and form a thymine dimer (Figure 4.5). If uncorrected, thymine dimers create a distortion in the DNA helix known as a bulky lesion. This inhibits normal base pairing between the two strands of the double helix and blocks the replication process. Ultraviolet light has a powerful germicidal action and is widely used to sterilize equipment. One of the reasons why bacteria are killed by this treatment is because the formation of large numbers of thymine dimers prevents replication.
Repair Processes
If there were no way to correct altered DNA, the rate of mutation would be intolerable. DNA excision and DNA repair enzymes have evolved to detect and to repair altered DNA. The role of the repair enzymes is to cut out (excise) the damaged portion of DNA and then to repair the base sequence. Much of our knowledge of DNA repair has been derived from studies on E. coli, but the general principles apply to other organisms such as ourselves. Repair is possible because DNA comprises two complementary strands. If the repair mechanisms can identify which of the two strands is the damaged one, it can be repaired to be as good as new by rebuilding it so that is again complementary to the undamaged strand.
Two types of excision repair are described in this section: base excision repair and nucleotide excision repair. The common themes for each of these repair mechanisms are: (i) an enzyme recognizes the damaged DNA, (ii) the damaged portion is removed, (iii) DNA polymerase inserts the correct nucleotide(s) into position (according to the base sequence of the second DNA strand), and (iv) DNA ligase joins the newly repaired section to the remainder of the DNA strand.
Base excision repair is needed to repair DNA that has lost a purine (depurination), or where a cytosine has been deaminated to uracil (U). Although uracil is a normal constituent of RNA, it does not form part of undamaged DNA and is recognized and removed by the repair enzyme uracil – DNA glycosidase (Figure 4.4). This leaves a gap in the DNA where the base had been attached to deoxyribose. There is no enzyme that can simply reattach a C into the vacant space on the sugar. Instead, an enzyme called AP endonuclease recognizes the gap and removes the sugar by breaking the phosphodiester bonds on either side (Figure 4.6). When DNA has been damaged by the loss of a purine (Figure 4.4), AP endonuclease also removes the sugar that has lost its base. The AP in the enzyme's name means apyrimidinic (without a pyrimidine) or apurinic (without a purine).
The repair process for reinserting a purine or a pyrimidine into DNA is now the same (Figure 4.6). DNA polymerase I replaces the appropriate deoxyribonucleotide into position. DNA ligase then seals the strand by catalyzing the reformation of a phosphodiester bond.
Nucleotide excision
