6.7.2.3 Mismatch Repair. DNA polymerases^54 are extremely accurate at copying DNA templates
(10^7 error rate for E. coli) but they are not perfect. In part, this great accuracy involves a mechanism
called proofreading. The major DNA polymerases of prokaryotes and eukaryotes possess a 3–5exonu-
clease activity, which removes any nucleotide that has not been correctly base paired with the template
during the extension reaction. Nevertheless, very occasionally, an incorrect nucleotide is inserted into a
new DNA strand by DNA polymerase. This generates a sequence mismatch, which is corrected by mis-
match repair.^55 The sequence mismatch causes a small irregularity in the double helix, which leads to a
loss of base pairing around the mismatch. A short region on one of the two strands must be removed and
the lesion filled in by DNA polymerase. How does the repair machinery know which strand to remove?
Many organisms, such as E. coliand humans, have DNA methylation (Section 6.6.3), which distinguishes
old DNA from newly synthesised DNA. Other organisms lacking DNA methylation, such as yeast and
Drosophila, must use another way to recognise recently synthesised DNA, but this is not yet understood.
6.7.2.4 Repairing Double-Stranded DNA Breaks. A break in both DNA strands is extremely dan-
gerous for the cell. If such damage is not repaired the exposed ends might become degraded, leading to a
deletion of DNA at the break point. Alternatively, an entire chromosome segment may be lost along with
its genes or one chromosome segment translocated on to the telomere of another chromosome. In humans
Genes and Genomes 237
Figure 6.32 Types of DNA damage encountered by the DNA repair machinery. (a) Common types of DNA damage,
together with examples and their cause(s). (b) Formation of a cyclobutyl adduct (commonly called a
thymine dimer)