eduKateSG · How Science Works
Choose your reading route
Cells use several molecular systems to recognise damaged DNA or copying errors, remove or process the affected region, and restore a usable strand. Many repairs use an intact complementary strand or another matching DNA template. Other pathways join broken ends. Repair is active maintenance, but it is not always successful or error-free.
Full contents · Check your understanding · How Science Works master guide
DNA is often pictured as a durable instruction book. The picture misses an important fact: the material carrying those instructions is chemically vulnerable. Cells preserve genetic information not because DNA never changes, but because surveillance, repair and controlled copying repeatedly reduce the consequences of damage.
Part of the How Science Works collection. Parent guide: Molecular Biology.
Contents
Understand the mechanism · Chapters 1–3
1. Damage and mutation are not the same event
Follow the reasoning · Chapters 4–6
4. Nucleotide excision repair removes a damaged stretch
Evidence and checks · Chapters 7–8
7. What would demonstrate repair rather than survival alone?
DNA damage is a physical or chemical alteration, such as a modified base, an abnormal linkage or a strand break. A mutation is a change in sequence that can persist through copying. Damage can lead to a mutation, but the two terms are not synonyms.
A lesion may be repaired before copying fixes an altered sequence. Alternatively, a copying error may escape correction. Once a changed base pair is established as an ordinary-looking pair, the cell may have no general way to identify it as a historical mistake.
That distinction prevents a misleading claim that repair machinery possesses a perfect master copy of every intended sequence. It recognises molecular features and uses available information; it does not compare the genome against an ideal blueprint stored elsewhere.
The Molecular Biology guide explains the wider information system. Here the focus is the maintenance mechanisms and the evidence that separates them.
In ordinary double-stranded DNA, bases on one strand pair with corresponding bases on the other. If one strand contains a damaged region while the opposite strand remains usable, the intact strand can guide replacement synthesis.
The reasoning resembles repairing a corrupted line from a matching record, but the match is complementary rather than identical. A polymerase adds nucleotides according to the template, and other proteins process and seal the repaired region.
This advantage has limits. Both strands can be damaged, a template can be unavailable, or the surrounding structure can make access difficult. DNA is packaged with proteins and participates in ongoing copying and transcription; repair occurs within that busy environment.
Different lesion types therefore need different routes. Removing one altered base, cutting out a short damaged stretch, correcting a replication mismatch and joining a broken chromosome are not all the same operation with different names.
Some lesions change the chemistry of an individual base without creating a large distortion. In base excision repair, a DNA glycosylase recognises a suitable altered base and removes it, initially leaving a site without that base. Further enzymes process the site, replace the missing information and seal the backbone.
An important example concerns uracil arising when cytosine is deaminated. In Lindahl’s 1974 study, an enzyme from E. coli released uracil from appropriate DNA substrates. This identified a concrete chemical activity rather than merely suggesting that cells must somehow maintain DNA.
The recogniser and the full repair pathway should not be conflated. Detecting removal of a damaged base establishes one step; it does not by itself demonstrate completion of every downstream step.
Different glycosylases recognise different substrates, and repair can involve short or longer replacement patches. The compact classroom sequence is recognition, base removal, backbone processing, synthesis and sealing—not one universal enzyme doing all five jobs.
Bulky lesions that distort DNA can be handled through nucleotide excision repair. A region around the lesion is opened and cut, a short segment containing the damage is removed, and the gap is filled using the intact strand as a template before sealing.
This differs from initially removing a single base. The unit excised includes a stretch of nucleotides containing the lesion. Both routes restore continuity, but their recognition and processing mechanisms differ.
Some nucleotide excision repair begins through surveillance of the genome; another route is coupled to transcription when a lesion stalls the transcription machinery. The initial problem is detected in different contexts, while downstream repair steps share components.
The Nobel committee’s scientific background on DNA repair explains the mechanistic research recognised in 2015. The existence of multiple routes is central: a lesion that fits one pathway is not evidence that every form of damage is handled identically.
Replication can leave a mismatched base pair or a small insertion/deletion loop. Proofreading by a replicative polymerase removes many errors during synthesis, but mismatch repair provides an additional layer of correction after errors have escaped that first check.
A mismatch presents a special information problem: which member of the pair should be replaced? Repair needs signals associated with the newly synthesised strand and replication context. The detailed signals differ among organisms; a bacterial textbook mechanism should not simply be copied onto human cells.
The affected region of the newly made strand is removed and resynthesised using the other strand as the template. This reduces replication errors but does not guarantee a zero mutation rate.
Consider a hypothetical G–T mismatch. Its appearance alone does not announce whether G or T was the original template base. Correct repair depends on strand discrimination, not on a rule that one of those letters is always wrong.
A double-strand break interrupts both strands and removes the straightforward option of using the opposite strand across the break. Cells can use different pathways depending on the break, cell-cycle stage and available templates.
Homologous recombination uses a matching DNA sequence, often a sister chromatid in the appropriate cell-cycle context, to guide repair. It can restore sequence accurately when the necessary information and processing are available.
Non-homologous end joining brings broken ends together and seals them. Compatible ends can sometimes be joined without changing sequence, while processing damaged or incompatible ends can introduce insertions or deletions. Describing every end-joining event as inevitably inaccurate is too crude.
No simple ranking makes one route universally best. Speed, template availability and the actual molecular problem matter. The article explains the normal conceptual routes; it does not provide experimental instructions for editing or damaging DNA.
Cells remaining alive after a challenge does not establish which repair pathway acted. They might tolerate damage, delay division or differ in how much damage they initially received. The mechanism requires measurements closer to the proposed steps.
Researchers can examine defined DNA substrates, repair intermediates, restoration of strand continuity and sequence outcomes. Comparing a system with and without a particular protein can test its role, provided the comparison controls other changes.
Hypothetical example: one preparation removes an altered base but does not seal the resulting gap; another completes both processes. An assay measuring only base removal would make the preparations look more alike than an assay measuring completed repair.
An informative report therefore states the endpoint: lesion removal, gap filling, ligation, sequence restoration or survival. These measurements answer different questions. A successful early step is not proof of a fully restored chromosome.
Restoring a missing component and recovering the predicted activity strengthens a causal argument. Even then, a purified biochemical system and a living cell have different levels of complexity. Moving between them requires additional evidence.
Repair pathways interact with damage signalling, checkpoints, transcription and replication. A checkpoint can delay progression while a problem is addressed, but signalling that a problem exists is not itself the chemical repair.
The existing CRISPR guide discusses how genome-editing tools can engage cellular repair. This article instead explains why cells already need repair systems in ordinary life. It is not an editing protocol or a health recommendation.
Does all damage become a mutation? No. Some damage is repaired before it becomes a persistent sequence change. Other damage may block a process without producing that particular mutation.
Does an intact-looking DNA strand prove the original sequence was restored? No. Continuity and sequence fidelity are separate endpoints.
Does repair reverse every existing mutation? No. A stable, normally paired sequence change may not present a recognisable lesion or reveal which sequence existed before.
The central chain is recognise the molecular problem → choose a compatible pathway → process and rebuild → seal → verify the outcome. The persistence of genetic information depends on repeated maintenance, not indestructible molecules. Return to How Science Works to follow how such mechanisms are established experimentally.