Base excision repair (BER) repairs many small, chemically altered DNA bases that do not strongly distort the overall double helix.

A key idea is that the damaged base can be removed while leaving the surrounding DNA largely intact.

Step 1: a DNA glycosylase removes the damaged base

A DNA glycosylase recognizes a particular kind of abnormal base and cleaves the bond connecting that base to its sugar.

The sugar-phosphate backbone remains present, but the nucleotide now contains no base. This intermediate is an abasic site, also called an AP site.

For example, spontaneous deamination can convert cytosine to uracil. Because uracil is not normally used as a DNA base, a uracil-DNA glycosylase can recognize and remove it.

Step 2: the damaged sugar-phosphate residue is removed

An endonuclease is an enzyme that cuts within a nucleic-acid strand rather than removing residues only from an end. An AP endonuclease cuts the DNA backbone near the abasic site. Additional enzymatic steps remove the remaining damaged sugar-phosphate residue.

Step 3: the sequence is restored from the opposite strand

DNA polymerase inserts the nucleotide specified by the intact complementary strand, and DNA ligase seals the remaining nick.

Schematically:

damaged base
   ↓ glycosylase
abasic site
   ↓ backbone cleavage and cleanup
gap
   ↓ DNA polymerase
correct nucleotide restored
   ↓ ligase
continuous DNA

Why the opposite strand is valuable

Double-stranded DNA stores redundant complementary information. If only one strand contains the lesion, the undamaged strand can identify which base belongs at that position.

BER therefore illustrates a general repair principle: recognize abnormal chemistry, remove the damaged material, and reconstruct sequence from an intact template.

This pathway is different from mismatch repair. Mismatch repair corrects a wrong pairing between otherwise ordinary bases after replication; base excision repair begins by recognizing a chemically abnormal base itself.