DNA Replication and Mutation

Replication begins at defined chromosomal sites called origins of replication. In bacteria a single origin suffices; the human genome has thousands distributed across all chromosomes, allowing simultaneous initiation at many points and completion of the 6.4 billion base pairs within hours rather than years. At each origin, initiator proteins recruit helicase, which unwinds the double helix by breaking the hydrogen bonds between complementary base pairs. The unwinding creates a replication bubble with two diverging replication forks moving outward in opposite directions.

DNA polymerase cannot start a new chain from scratch. It requires a free 3-prime hydroxyl group on which to add the next nucleotide. Primase, an RNA polymerase, synthesises a short RNA primer (8 to 12 nucleotides) complementary to the template strand. DNA polymerase then extends the primer, adding deoxyribonucleotides one at a time. The leading strand template runs 3-prime to 5-prime relative to the direction of fork movement, so the leading strand is synthesised continuously from a single primer. The lagging strand template runs 5-prime to 3-prime, opposite to fork movement. It is copied in short Okazaki fragments, each requiring its own primer, each extended 5-prime to 3-prime away from the fork. The RNA primers are subsequently removed, the gaps are filled by DNA polymerase, and the nicks are sealed by DNA ligase.

In eukaryotes, the replicated strands must also be assembled into chromatin. Histone chaperones deposit histone octamers onto the new DNA almost immediately after synthesis. Parental histones carrying epigenetic marks are distributed between the two daughter strands, but the precise mechanism of epigenetic inheritance through replication remains an active area of research. Forensic applications that exploit epigenetic marks (for example, tissue-type prediction from methylation patterns) depend on understanding how faithfully those marks are copied.

DNA polymerase achieves high fidelity through two sequential mechanisms. First, nucleotide selection: the polymerase undergoes a conformational change around the incoming nucleotide before catalysis. The correct nucleotide, forming a standard Watson-Crick pair with the template base, induces a closed active-site conformation that positions the 3-prime hydroxyl for attack. A mismatch holds the active site open, delaying catalysis and giving the polymerase time to reject the incorrect nucleotide. This selectivity reduces the raw error rate to approximately one mistake per 100,000 insertions.

Second, proofreading: if a mismatched nucleotide is incorporated despite selectivity, the 3-prime to 5-prime exonuclease domain of the polymerase detects the distortion in the growing chain and excises the incorrect nucleotide before synthesis resumes. Proofreading reduces the error rate by roughly another 100-fold, to approximately one error per 10 million bases. Post-replication mismatch repair (MMR) catches the remaining errors: the MMR proteins scan newly synthesised DNA, recognise mispaired bases, excise a segment of the new strand containing the mismatch, and fill in the gap correctly. Together, these layers bring the final mutation rate in human somatic cells to approximately one base change per billion bases per division.

Inherited defects in MMR genes (MLH1, MSH2, MSH6, PMS2) cause Lynch syndrome, the most common hereditary colorectal cancer predisposition syndrome. Tumours from Lynch syndrome patients show microsatellite instability: the STR loci that are stable in normal tissue accumulate new alleles at high frequency. This is directly relevant to forensic biology: tumour DNA from a Lynch syndrome patient may yield STR profiles at some loci that differ from the constitutional profile obtained from blood.

Mutations are classified by the nature and scale of the sequence change. Point mutations affect a single base pair. A transition replaces one purine with the other purine (A for G or G for A) or one pyrimidine with the other (C for T or T for C). A transversion replaces a purine with a pyrimidine or vice versa. Transitions occur more frequently than transversions in spontaneous mutation, because the incorrect nucleotide more readily adopts a geometry that mimics a correct Watson-Crick pair when the change is within the same chemical class.

In a protein-coding context, a point mutation may be synonymous (the altered codon specifies the same amino acid, because the genetic code is degenerate), missense (the codon now specifies a different amino acid), or nonsense (the codon becomes a stop codon, truncating the protein). Synonymous mutations are largely invisible to natural selection and accumulate over time, contributing to the population-level sequence diversity captured by SNP panels.

Insertions and deletions (indels) add or remove one or more base pairs. In a coding region, an indel that is not a multiple of three shifts the reading frame, altering every codon downstream of the change. Frameshift mutations in essential genes are usually lethal when homozygous and are strongly selected against. In non-coding regions, indels can alter regulatory sequences or splice sites without directly changing any protein sequence. STR loci are non-coding, and indels at these loci through replication slippage are the primary source of allele-length variation exploited in forensic DNA profiling.

Spontaneous mutations arise without external chemical or physical insult. The two main sources are replication errors that escape repair and spontaneous chemical changes in DNA bases. Tautomeric shifts are transient alterations in the electronic structure of a base that change its hydrogen-bonding properties. Adenine normally pairs with thymine, but its imino tautomeric form pairs with cytosine. If the imino form is present during replication, the polymerase incorporates cytosine opposite an adenine template, creating an A-T to G-C transition after one further round of replication.

Depurination is the spontaneous hydrolysis of the N-glycosidic bond linking a purine base (adenine or guanine) to the deoxyribose sugar. Human cells lose an estimated 10,000 purines per cell per day through depurination. The resulting abasic (AP) site has no coding information and stalls DNA polymerase. Translesion synthesis polymerases can insert a nucleotide opposite the AP site, but they do so with low fidelity, usually incorporating adenine. Because the missing base was a purine, this insertion may be incorrect, generating a mutation.

Deamination converts cytosine to uracil by removing the amino group. Uracil pairs with adenine, not guanine, so an uncorrected deamination leads to a C-G to T-A transition. Uracil-DNA glycosylase normally removes uracil from DNA before replication, but if it fails, the mutation is fixed. In forensically relevant aged or degraded biological samples, depurination and deamination are significant sources of miscoding lesions that can generate artifactual base calls during PCR-based sequencing.

Environmental agents increase the mutation rate above the spontaneous background. They are classified by mechanism: agents that chemically modify DNA bases, agents that intercalate between base pairs, and agents that break the phosphodiester backbone.

Ultraviolet (UV) radiation at wavelengths around 260 nanometres is absorbed efficiently by pyrimidine bases. The absorbed energy drives covalent bond formation between adjacent pyrimidines on the same strand. The predominant product is a cyclobutane pyrimidine dimer (CPD), most often thymine-thymine, though cytosine-thymine and cytosine-cytosine dimers also form. CPDs distort the double helix and stall the replicative polymerase. Nucleotide excision repair (NER) is the primary removal pathway. When a CPD is replicated before repair, translesion synthesis polymerases typically insert adenine opposite both thymine residues of the dimer, which is correct for thymine but incorrect if a cytosine was involved, producing a characteristic C to T or CC to TT transition. These UV-signature mutations are used in cancer genomics to assign mutational aetiology, and they are relevant to the degradation of DNA in biological samples left in sunlight.

Alkylating agents transfer alkyl groups to DNA bases. Methyl methanesulphonate (MMS) and similar agents methylate guanine at the O6 position, producing O6-methylguanine. This lesion pairs with thymine instead of cytosine, generating a G-C to A-T transition. N-alkyl adducts on adenine block replication. Many alkylating agents found in tobacco smoke and industrial chemicals are potent mutagens. Ionising radiation (gamma rays, X-rays) deposits energy that generates reactive oxygen species within cells, which attack all four bases and the phosphodiester backbone. Double-strand breaks produced by ionising radiation are particularly mutagenic, because repair by non-homologous end joining is error-prone, and may join the wrong chromosome ends together.

Forensic DNA profiling rests on the existence of heritable variation at specific genomic loci. That variation is the accumulated product of mutations across many generations. The CODIS STR system used in the United States profiles 20 core loci; equivalent systems operate in the United Kingdom (National DNA Database, expanded from 10 to 16 STR loci in 2014), the European Union (ESS 7-plus-additional loci), and India's National DNA Data Bank, established under the DNA Technology (Use and Application) Regulation Act 2019. Each locus is polymorphic because replication slippage has created many alleles of different repeat-unit counts in the human population.

The probability that two unrelated individuals share the same genotype at all 20 CODIS loci is vanishingly small, typically less than one in a quadrillion, because each locus contributes an independent probability factor that compounds multiplicatively. This statistical power depends on the genetic independence of the loci (the loci are selected from different chromosomes or from well-separated positions on the same chromosome) and on accurate population allele frequency databases for each ethnic group. Mutation rates at STR loci are low enough (approximately 0.1 to 0.2 per cent per locus per generation) that parent-child pairs share the same allele at most loci, but that rate is high enough that a mutation is occasionally seen in paternity cases.

Single nucleotide polymorphisms (SNPs) are point mutations that have become fixed in at least one per cent of the population. The human genome contains more than 10 million common SNPs. SNP panels are used in forensic genetics for ancestry inference (informative SNPs differ systematically in frequency across continental population groups), for kinship analysis when STR typing fails or when samples are highly degraded, and for phenotypic prediction (hair colour, eye colour, skin tone). The Forensic Genetics Policy Initiative and regulatory frameworks in the UK (Protection of Freedoms Act 2012), EU, and the United States (Combined DNA Index System regulations) set limits on which applications of SNP data are permissible in law enforcement contexts. More detail on evidence collection and tissue types is at Scope of Biological Evidence; biotechnology applications including next-generation sequencing of SNP panels are covered at Forensic Biotechnology.