Skip to content

DNA Replication, Repair and Recombination

Origins, forks, polymerases and fidelity, plasmid and organelle replicons, DNA repair pathways, and homologous and site-specific recombination for CSIR NET.

  • 24 min read
  • Updated
On this page · 14 sections
  1. Prerequisites
  2. The Hook: Three Bands in a Caesium Chloride Gradient
  3. The Logic of Replication
  4. Fidelity: How One Error in a Billion Is Achieved
  5. Extrachromosomal Replicons
  6. DNA Damage and Repair
  7. Recombination
  8. Memory Tricks
  9. Don’t Confuse
  10. Common Mistakes
  11. Practice
  12. Quick Revision Table
  13. Key Takeaways
  14. What’s Next

Prerequisites

The Hook: Three Bands in a Caesium Chloride Gradient


The Logic of Replication

Every DNA polymerase:

  1. needs a template and a primer with a free 3’-OH;
  2. adds dNMPs from dNTPs, releasing PPi (pyrophosphatase hydrolysis of PPi pulls the reaction forward);
  3. synthesises only 5’ → 3’.

Because the two strands are antiparallel, one daughter strand (the leading strand) is made continuously towards the fork, and the other (the lagging strand) is made discontinuously away from the fork as Okazaki fragments, each started by an RNA primer. Okazaki fragments are about 1,000–2,000 nt in bacteria and 100–200 nt in eukaryotes.

A replicon is a stretch of DNA replicated from one origin. The bacterial chromosome and most plasmids are single replicons; a eukaryotic chromosome has many. Replication from an origin is usually bidirectional (two forks moving apart), shown by autoradiography of replicating E. coli chromosomes (the “theta” structure) and by pulse-labelling experiments.

Origins and initiation in E. coli

oriC is a short region (a few hundred base pairs) containing:

  • several 9-bp DnaA boxes, bound by DnaA-ATP;
  • an AT-rich DNA unwinding element (a set of 13-mer repeats) that melts first, because A·T pairs have two H-bonds and stack less strongly;
  • GATC sites that are methylated on A by Dam methylase.
graph LR
    A["DnaA-ATP binds
DnaA boxes"] --> B["AT-rich 13-mers
melt"] B --> C["DnaC loads DnaB
helicase on each strand"] C --> D["DnaG primase
makes RNA primers"] D --> E["β-clamp loaded;
Pol III holoenzyme
extends primers"]

Preventing re-initiation (once per cell cycle):

  • SeqA binds newly replicated, hemimethylated oriC and sequesters it until Dam methylates the new strand.
  • RIDA (regulatory inactivation of DnaA): the loaded β-clamp and Hda stimulate DnaA’s ATPase, converting DnaA-ATP to inactive DnaA-ADP.

The enzymes at the bacterial fork

ProteinGeneJobExam point
DnaAdnaAInitiator; binds DnaA boxes as ATP formInitiation only; ts mutants stop at the end of the round
DnaBdnaBReplicative helicase, hexamer, moves 5’ → 3’ on the lagging-strand templateATP-driven unwinding
DnaCdnaCHelicase loaderInitiation
DnaGdnaGPrimase: short RNA primersRifampicin-resistant (not RNA polymerase)
SSBssbCoats single strands, prevents re-annealing and hairpinsCooperative binding
DNA gyrase (topo II)gyrA, gyrBRemoves positive supercoils ahead of the fork by introducing negative supercoils, ATP-dependentQuinolones (nalidixic acid, ciprofloxacin) target GyrA; novobiocin targets the GyrB ATPase
Pol III holoenzymednaE (α), dnaQ (ε)Main replicase: α polymerises, ε is the 3’ → 5’ proofreading exonucleaseVery processive with the β-clamp
β-clampdnaNRing-shaped sliding clamp (dimer) that tethers Pol III to DNAProcessivity factor; eukaryotic equivalent PCNA
Clamp loader (γ/τ complex)Opens and closes the clamp with ATP; τ links two cores (leading + lagging)Eukaryotic equivalent RFC
Pol IpolARemoves RNA primers with its 5’ → 3’ exonuclease and fills the gap (nick translation)Klenow fragment lacks the 5’ → 3’ exo
DNA ligaseligASeals nicks; E. coli ligase uses NAD⁺, T4 and eukaryotic ligases use ATPOkazaki fragments accumulate in ligase mutants
Topo IVparC, parEDecatenates interlinked daughter chromosomesNeeded for segregation
TustusBinds Ter sites to form polar fork traps in the terminus regionForks enter but cannot leave

Pol I was the first DNA polymerase discovered, but a polA mutant that lacked most Pol I activity still replicated its DNA (it was, however, sensitive to UV). That result is the classic argument that Pol I is not the main replicase; its essential role is primer removal and repair.

Eukaryotic replication

StepProteinsNote
Licensing (G1)ORC (Orc1–6) marks origins; Cdc6 and Cdt1 load the MCM2–7 helicase as an inactive double hexamerOnly in G1, when CDK activity is low
Firing (S)S-CDK and DDK (Cdc7–Dbf4) recruit Cdc45 and GINS → active CMG helicaseEach origin fires at most once
PrimingPol α–primase: RNA primer extended by a short stretch of DNAPol α has no proofreading
Leading strandPol εHas 3’ → 5’ exo
Lagging strandPol δ with PCNA (loaded by RFC)Has 3’ → 5’ exo
Primer removalRNase H and FEN1 (flap endonuclease)Pol δ displaces the primer as a flap
SealingDNA ligase IATP-dependent

Re-replication block: licensing (MCM loading) needs low CDK, firing needs high CDK, so an origin cannot be reloaded until the next G1. In metazoans geminin also binds Cdt1 in S/G2 and blocks reloading.

Origins: budding yeast origins are short, defined sequences (ARS, autonomously replicating sequences, with an ORC-binding consensus). Metazoan origins have no strict consensus sequence and are defined more by chromatin context. Origins fire at different times in S phase: euchromatin tends to replicate early, heterochromatin late.

The end-replication problem and telomerase

When the last RNA primer at a linear chromosome end is removed, there is no upstream 3’-OH from which to fill the gap, so each round would shorten the lagging-strand product. Telomerase, a ribonucleoprotein reverse transcriptase (TERT protein + TR/TERC RNA template), extends the 3’ overhang by copying its own RNA template; the complementary strand is then filled by primase and Pol α. Human telomeres are tandem repeats of TTAGGG; the ends are capped by the shelterin complex and form a t-loop. Most somatic cells have little telomerase and their telomeres shorten with divisions; germ cells, stem cells and most cancer cells keep telomerase active (some cancers use recombination instead, called ALT).

Other polymerases you must know

PolymeraseRole
E. coli Pol IIRepair; restart of stalled forks
E. coli Pol IV (DinB) and Pol V (UmuD’₂C)SOS translesion synthesis, error-prone
Pol βBase excision repair gap filling (has dRP lyase)
Pol γMitochondrial DNA replication
Pol ηTranslesion synthesis across thymine dimers, inserting A opposite each T; loss causes the variant form of xeroderma pigmentosum (XP-V)
Pol ζExtension after translesion insertion
Reverse transcriptaseRNA-dependent DNA polymerase (retroviruses, retrotransposons, telomerase)

Fidelity: How One Error in a Billion Is Achieved

Overall replication errors are about one per 10⁹–10¹⁰ nucleotides. Three layers multiply:

LayerMechanismApproximate error rate after this layer
Base selectionCorrect Watson–Crick geometry fits the active site; induced fit; dNTP pool balanceabout 10⁻⁴–10⁻⁵
Proofreading3’ → 5’ exonuclease removes a mismatched 3’ terminal nucleotide (ε subunit of Pol III; Pol δ, Pol ε)about 100-fold better: 10⁻⁶–10⁻⁷
Mismatch repairPost-replicative; recognises mismatches behind the forkanother 100–1,000-fold: 10⁻⁹–10⁻¹⁰
$$\text{overall error rate} \approx \text{(insertion error)} \times \text{(fraction escaping proofreading)} \times \text{(fraction escaping MMR)}$$

Why a primer has to be RNA: a polymerase that can start chains de novo cannot proofread (it has no “correct primer” to check). Starting with RNA marks the error-prone first nucleotides for removal and replacement by a proofreading polymerase.


Extrachromosomal Replicons

RepliconModeControl
ColE1-type plasmids (pBR322, pUC)θ (theta) replication, unidirectional; RNA II is the primer, processed by RNase H; host Pol I starts synthesisRNA I (antisense) pairs with RNA II and blocks primer formation; Rop/Rom protein stabilises the RNA I–RNA II complex. pUC plasmids lack rop and carry a point mutation in the RNA region, so their copy number is higher
Iteron plasmids (F, R6K, P1)θ; plasmid Rep initiator protein binds repeated sequences (iterons)Rep titration and “handcuffing” of origins; low copy number
Rolling-circle plasmids and ssDNA phages (φX174, M13)Rep protein nicks the double-strand origin, 3’-OH is extended while the old strand is displaced as ssDNA; a separate origin is used to make the complementary strandRep is used once per round
F plasmid transfer in conjugationRolling-circle from oriT, one strand enters the recipientRelaxase nicks oriT; recipient makes the complement
Mitochondrial DNA (human, 16,569 bp, circular)Pol γ; a strand-displacement model with two origins (O_H and O_L) and a D-loop at the control regionNuclear-encoded machinery; many copies per cell
Yeast 2 µm plasmidUses host origin licensing (once per S phase)FLP recombinase-mediated amplification raises copy number

Plasmid incompatibility: two plasmids with the same replication control system cannot be stably maintained together, because the control system counts them as one population and random segregation eventually loses one.


DNA Damage and Repair

Types of damage

DamageSourceConsequence
Cyclobutane pyrimidine dimers, 6-4 photoproductsUVBlock polymerases; distort helix
Deamination: C → U, 5mC → T, A → hypoxanthineSpontaneous, nitrous acidTransitions: C·G → T·A (from C or 5mC); A·T → G·C (hypoxanthine pairs with C)
DepurinationSpontaneous (thousands per cell per day in mammals)AP (abasic) site
Alkylation: O⁶-methylguanine, 3-methyladenineEMS, MNNG, endogenous SAMO⁶-meG pairs with T → G·C → A·T
Oxidation: 8-oxoguanineReactive oxygen species8-oxoG pairs with A → G·C → T·A transversion
Strand breaks (single, double)Ionising radiation, collapsed forksDSBs are the most lethal lesion
Interstrand cross-linksCisplatin, psoralen + UVA, mitomycin CBlock both strands; Fanconi anaemia pathway

Why DNA has T and not U: cytosine deaminates spontaneously to uracil. If U were a normal base, repair enzymes could not tell a “correct” U from a deaminated C. With T in DNA, any U is damage and uracil-DNA glycosylase removes it. 5-methylcytosine deaminates to T, which is a normal base, so methylated CpG sites are mutation hotspots.

The repair pathways

graph TD
    D["DNA damage"] --> R["Direct reversal
photolyase, MGMT/Ada, AlkB"] D --> B["Base excision repair
small, non-distorting base lesions"] D --> N["Nucleotide excision repair
bulky, helix-distorting lesions"] D --> M["Mismatch repair
replication errors"] D --> DS["Double-strand break repair
HR or NHEJ"] D --> T["Damage tolerance
translesion synthesis, SOS"]

1. Direct reversal (no nucleotides removed)

  • Photolyase splits cyclobutane pyrimidine dimers using blue/near-UV light and an FADH⁻ cofactor (photoreactivation). Bacteria, plants and many animals have it; placental mammals do not have the CPD photolyase.
  • O⁶-methylguanine-DNA methyltransferase (MGMT; E. coli Ada) transfers the methyl group to one of its own Cys residues. Each molecule acts once: a suicide enzyme. Methylated Ada also activates transcription of the adaptive response genes.
  • AlkB family dioxygenases (Fe²⁺/α-ketoglutarate) oxidatively remove methyl groups from 1-methyladenine and 3-methylcytosine.

2. Base excision repair (BER)

  1. A DNA glycosylase cleaves the N-glycosidic bond of the damaged base (uracil-DNA glycosylase; OGG1 for 8-oxoG; MUTYH removes A mispaired with 8-oxoG) → AP site.
  2. AP endonuclease (APE1; E. coli exonuclease III/endonuclease IV) nicks 5’ to the AP site.
  3. Pol β removes the 5’-deoxyribose phosphate (dRP lyase) and inserts one nucleotide (short patch); in long-patch BER, Pol δ/ε displace 2–10 nt that FEN1 removes.
  4. DNA ligase (ligase III–XRCC1 or ligase I) seals.

3. Nucleotide excision repair (NER)

  • E. coli: UvrA₂B scans and recognises the distortion; UvrA leaves; UvrC cuts on both sides of the lesion (excinuclease), releasing a 12–13-nt oligonucleotide; UvrD (helicase II) removes it; Pol I fills; ligase seals.
  • Humans: about 30 nt are excised. XPC–RAD23B detects lesions genome-wide (global genome NER); a stalled RNA polymerase II recruits CSB/CSA (transcription-coupled NER). TFIIH (helicases XPB and XPD) opens the DNA, XPA verifies, XPG cuts 3’ and XPF–ERCC1 cuts 5'.
  • Defects: xeroderma pigmentosum (XP-A to XP-G; extreme UV sensitivity and skin cancer), Cockayne syndrome (CSA/CSB; TC-NER defect, developmental and neurological, but not the high cancer rate of XP).

4. Mismatch repair (MMR) in E. coli

  1. MutS binds the mismatch (or small insertion/deletion loop).
  2. MutL joins and activates MutH.
  3. MutH nicks the unmethylated (new) strand at a hemimethylated GATC, which can be up to about a kilobase away on either side.
  4. UvrD unwinds from the nick towards the mismatch; an exonuclease of the right polarity (3’ → 5’ if the nick is 3’ of the mismatch; 5’ → 3’ if it is 5’) degrades the new strand past the mismatch.
  5. Pol III resynthesises; ligase seals.

Strand discrimination is the key idea: the parent strand is methylated, the new strand is not yet methylated for a short window after the fork passes. Eukaryotes have MutS homologues (MSH2–MSH6, MSH2–MSH3) and MutL homologues (MLH1–PMS2) but no MutH; they use nicks and gaps in the new strand (Okazaki ends, ribonucleotides) and PCNA orientation to identify it.

5. Double-strand break repair

FeatureHomologous recombination (HR)Non-homologous end joining (NHEJ)
TemplateSister chromatid (or homologue)None
Cell-cycle phaseS and G2All phases, dominant in G1
Key proteinsMRN (Mre11–Rad50–Nbs1), CtIP, Exo1/DNA2 resect 5’ ends; RPA; BRCA2 loads Rad51; strand invasionKu70/Ku80, DNA-PKcs, Artemis, XRCC4–DNA ligase IV, XLF
AccuracyError-freeOften small insertions or deletions
Other usesMeiotic crossing over, restart of collapsed forksV(D)J recombination, class-switch recombination

The ATM kinase is activated at DSBs and phosphorylates histone H2AX on Ser139 (γH2AX), which spreads around the break and recruits repair and checkpoint proteins; ATR responds to RPA-coated single-stranded DNA at stalled forks.

6. Damage tolerance: the SOS response in E. coli

  • LexA represses a set of SOS genes (including recA, lexA itself, uvrA, uvrB, umuDC, sulA).
  • ssDNA at stalled forks activates RecA filaments (RecA*), which act as a co-protease: LexA undergoes autocleavage, and the SOS genes are derepressed. The same activity triggers autocleavage of UmuD → UmuD’ and of the λ cI repressor (prophage induction).
  • SulA inhibits FtsZ to delay division. Late in the response Pol V (UmuD’₂C) performs error-prone translesion synthesis: this is why UV mutagenesis in E. coli needs recA and umuDC.

Recombination

Homologous (general) recombination

Exchange between DNA molecules with extensive sequence identity. Uses: meiotic crossing over and correct chromosome segregation, DSB repair, restart of broken forks.

Models

  • Holliday model: single-strand nicks at the same position on both homologues, strand exchange forms a four-way Holliday junction, branch migration creates heteroduplex, and resolution by cutting either pair of strands gives crossover or non-crossover (patch) products.
  • Double-strand break repair (DSBR) model: a DSB is resected to 3’ single-stranded tails; one tail invades the homologue (D-loop), primes synthesis, the second end is captured, and a double Holliday junction forms. Resolution gives crossovers or non-crossovers. Synthesis-dependent strand annealing (SDSA) releases the invading strand early and gives only non-crossovers (the main mitotic route).

Bacterial machinery

ProteinRole
RecBCDHelicase–nuclease that enters at a DSB end; degrades DNA until it meets a Chi site (5’-GCTGGTGG-3’), then produces a 3’ tail and loads RecA
RecAForms an ATP-dependent filament on ssDNA; searches for homology and catalyses strand invasion/exchange; also SOS co-protease
RuvA–RuvBRuvA binds the Holliday junction; RuvB is the hexameric ATPase motor for branch migration
RuvCResolvase: cuts two strands of the junction symmetrically
RecFORLoads RecA at single-strand gaps

Eukaryotic equivalents: Rad51 (RecA homologue; meiotic paralogue Dmc1), Spo11 makes programmed meiotic DSBs, MRN resects, and junctions are resolved by nucleases (such as GEN1, MUS81) or dissolved by BLM helicase–topoisomerase III.

Gene conversion: heteroduplex DNA containing a mismatch is “corrected” by mismatch repair towards one allele, so a heterozygous meiosis gives a 3:1 instead of a 2:2 ratio of alleles in a tetrad (6:2 in an eight-spored ascus). If the mismatch is not repaired before the next mitosis, the two chromatids of one spore pair differ (post-meiotic segregation, 5:3). Yeast mating-type switching is a programmed gene conversion started by the HO endonuclease.

Site-specific recombination

Exchange between two short, specific sites recognised by a dedicated recombinase; no extensive homology and no DNA synthesis. The outcome depends on site orientation:

Arrangement of the two sitesOutcome
Same molecule, direct repeat (same orientation)Excision (deletion) of the segment between them as a circle
Same molecule, inverted repeatInversion of the segment
Two different moleculesIntegration (fusion), or translocation between chromosomes
FamilyChemistryExamples
Tyrosine recombinasesActive-site Tyr forms a 3’-phosphotyrosine link; one pair of strands is exchanged at a time through a Holliday junction intermediateCre–loxP (phage P1), λ integrase (attP × attB), FLP–FRT (yeast 2 µm), XerC/XerD at dif (resolves chromosome dimers in E. coli)
Serine recombinasesActive-site Ser forms a 5’-phosphoserine link; all four strands are cut and the halves rotateHin invertase (Salmonella flagellar phase variation), γδ/Tn3 resolvase, phage φC31 integrase

loxP is 34 bp: two 13-bp inverted repeats flanking an asymmetric 8-bp spacer, which gives the site its orientation. Cre needs no cofactors, so Cre–lox is the standard tool for conditional knockouts: flank (“flox”) an exon with direct loxP sites and express Cre from a tissue-specific promoter.

Phage λ integration joins attP and the bacterial attB at an identical core sequence. Integration needs Int + IHF (integration host factor); excision needs Int + IHF + Xis. Because Xis is required only for excision, the reaction is directional, which is how the prophage stays put until induction.

Transposition (for completeness)

Transposable elements move with a transposase and create a short target-site duplication where they insert.

  • Cut-and-paste (conservative): IS elements, Tn10, Tn5, the maize Ac/Ds elements studied by McClintock, Drosophila P elements.
  • Replicative: Tn3, phage Mu. A cointegrate forms and is resolved by a site-specific resolvase (an example of site-specific recombination finishing a transposition).
  • Retrotransposons move through RNA and reverse transcriptase: LTR elements (yeast Ty1, Drosophila copia) and non-LTR elements (human LINE-1; SINEs such as Alu, which borrow LINE machinery).

Memory Tricks

Don’t Confuse

ThisvsThat
Primase (DnaG): RNA primers, rifampicin-resistantvsRNA polymerase: transcripts, rifampicin-sensitive
Pol III ε: 3’ → 5’ exo (proofreading)vsPol I: 5’ → 3’ exo (primer removal, nick translation)
Klenow fragment: Pol I without 5’ → 3’ exovsHoloenzyme Pol I: all three activities
Gyrase: introduces negative supercoils, ATPvsTopo I: relaxes, no ATP, changes Lk by 1
SeqA: binds hemimethylated oriC (sequestration)vsMutH: nicks the unmethylated strand of hemimethylated GATC (MMR)
BER: glycosylase removes a basevsNER: excinuclease removes an oligonucleotide
XP: NER defect, skin cancervsCockayne: TC-NER defect, no high cancer rate
HR: accurate, needs a template, S/G2vsNHEJ: error-prone, no template, any phase
Tyrosine recombinase: Holliday junction intermediatevsSerine recombinase: all four strands cut, rotation
λ integration: Int + IHFvsλ excision: Int + IHF + Xis

Common Mistakes


Practice

All questions below are original practice questions written in CSIR NET style. They are not previous-year questions.

Level 1: Foundation

  1. Which subunit of E. coli Pol III carries the proofreading exonuclease?
  2. What does E. coli DNA ligase use as the source of the adenylyl group?
  3. Which enzyme removes uracil from DNA?
Answers
  1. ε (encoded by dnaQ), a 3’ → 5’ exonuclease.
  2. NAD⁺ (eukaryotic and T4 ligases use ATP).
  3. Uracil-DNA glycosylase, the first step of base excision repair.

Level 2: Part B Style

Q1 (practice). E. coli grown for many generations in ¹⁵N medium is transferred to ¹⁴N medium. After three generations of growth, DNA is centrifuged in CsCl. The expected result is:

  • (a) one band of intermediate density
  • (b) heavy and light bands in the ratio 1:7
  • (c) intermediate and light bands in the ratio 1:3
  • (d) intermediate and light bands in the ratio 1:1
Answer and explanation

(c). The two original ¹⁵N strands are never lost; after n generations there are 2ⁿ duplexes, of which 2 are hybrid. For n = 3: 2 of 8 hybrid, 6 of 8 light → 1:3. (b) is the conservative prediction (1 heavy : 7 light), (d) is generation two.

Q2 (practice). A bacterial strain shows a mutation rate about 1,000 times above wild type. Its DNA is fully methylated at GATC sites, its polymerase has normal 3’ → 5’ exonuclease activity, and it is not UV-sensitive. The mutation most likely lies in:

  • (a) dam
  • (b) dnaQ
  • (c) mutS
  • (d) uvrA
Answer and explanation

(c) mutS. Methylation is normal (rules out dam), proofreading is normal (rules out dnaQ), and UV resistance is normal (rules out the NER gene uvrA). A mismatch-repair defect raises spontaneous mutation without UV sensitivity.

Q3 (practice). A mouse line carries an exon flanked by two loxP sites in inverted orientation. After Cre expression, the exon will be:

  • (a) deleted as a circle
  • (b) inverted, and the inversion can flip back and forth
  • (c) duplicated
  • (d) unaffected, because Cre needs a direct repeat
Answer and explanation

(b). Inverted sites give inversion. Because the product still has two inverted loxP sites, Cre can flip the segment back, giving a mixture of orientations (this is why designs such as FLEx use pairs of mutant lox sites to lock the inversion). Direct repeats would delete.

Q4 (practice). Which pairing of DNA lesion and first repair enzyme is incorrect?

  • (a) 8-oxoguanine — OGG1 glycosylase
  • (b) O⁶-methylguanine — MGMT
  • (c) Cyclobutane thymine dimer in human cells — photolyase
  • (d) Cytosine deaminated to uracil — uracil-DNA glycosylase
Answer and explanation

(c). Human (placental mammal) cells lack the CPD photolyase; thymine dimers are removed by nucleotide excision repair. The other three pairs are correct.

Level 3: Part C Style

Q1 (practice). The E. coli chromosome is 4.6 × 10⁶ bp and is replicated bidirectionally from one origin, with each fork moving at 1,000 nt s⁻¹.

(i) How long does one round of replication take? (ii) Cells in rich medium divide every 25 min. How is this possible? (iii) Assume a misincorporation rate of 10⁻⁵, of which 1% escapes proofreading, and that mismatch repair corrects 99.9% of what remains. How many errors are expected per genome replication (count both daughter strands as 9.2 × 10⁶ nucleotides synthesised)?

Answer and worked solution

(i) Each fork copies half the genome: \(t = \dfrac{4.6\times10^6/2}{1000} = 2300\ \text{s} \approx 38\ \text{min}\).

(ii) Replication time (plus the time to segregate and divide) exceeds the doubling time, so cells start new rounds at the origin before the previous round finishes (multifork replication). Daughter cells inherit chromosomes that are already partly replicated. Initiation, not fork speed, sets the division rate.

(iii) Error rate = \(10^{-5} \times 10^{-2} \times 10^{-3} = 10^{-10}\). Errors per replication = \(9.2\times10^{6} \times 10^{-10} = 9.2\times10^{-4}\), i.e. about one error per ~1,100 genome replications.

Q2 (practice). A eukaryotic chromosome of 120 Mb is replicated in an S phase of 8 h. Forks move at 40 nt s⁻¹ and origins are bidirectional. (A) Calculate the minimum number of origins if all fire at the start of S phase and forks never stall. (B) Explain why the real number of licensed origins is larger. Which statements are correct?

(P) The minimum number of origins is about 53. (Q) Each origin can fire twice in one S phase if the first fork stalls. (R) More MCM complexes are loaded than are used, and the spare (dormant) origins fire when forks stall. (S) Origins are licensed in G1 when CDK activity is low and fired in S when CDK activity is high.

  • (a) P, R and S
  • (b) P and Q only
  • (c) Q, R and S
  • (d) R and S only
Answer and worked solution

(a) P, R and S.

DNA copied per origin in 8 h = 2 forks × 40 nt s⁻¹ × 28,800 s = 2.304 × 10⁶ bp. Minimum origins = 1.2 × 10⁸ / 2.304 × 10⁶ = 52.1 → 53 (round up). (P) correct.

(Q) wrong: licensing is restricted to G1, so an origin fires at most once per cycle. (R) correct: dormant origins are a backup. (S) correct: the two-state CDK control is what prevents re-replication.

Q3 (practice). Four E. coli strains are tested for survival after UV and for UV-induced mutation (reversion of a point mutation):

StrainUV survivalUV-induced mutants
Wild typeHighMany
WVery lowVery few
XLowMany more per survivor than wild type
YModerately lowNone
ZHighMany; spontaneous mutation rate 100× higher

Assign recA, uvrA, umuC and mutL to W–Z, with reasons.

Answer and worked solution
  • W = recA: no recombinational repair and no SOS induction (no LexA cleavage, no Pol V), so survival is very low and UV mutagenesis disappears.
  • X = uvrA: no NER, so lesions persist and survival is low; more lesions remain for error-prone bypass, so more mutants per survivor.
  • Y = umuC: no Pol V, so no error-prone bypass; survival is somewhat reduced and UV mutagenesis is absent, although recombination still works.
  • Z = mutL: UV repair is normal, but mismatch repair fails, so the spontaneous mutation rate is high.

Quick Revision Table

ItemFact
Replication modeSemi-conservative (Meselson–Stahl, 1958); n generations → 2 hybrid of 2ⁿ
Direction5’ → 3’ only; leading continuous, lagging as Okazaki fragments
Okazaki fragments1,000–2,000 nt bacteria; 100–200 nt eukaryotes
oriCDnaA boxes + AT-rich 13-mers + GATC sites; SeqA and RIDA block re-initiation
Helicase / loader / primaseDnaB (5’ → 3’) / DnaC / DnaG
Main replicasePol III: α polymerase, ε proofreading, β-clamp
Primer removalPol I 5’ → 3’ exo (bacteria); RNase H + FEN1 (eukaryotes)
Ligase cofactorNAD⁺ (E. coli); ATP (eukaryotes, T4)
GyraseNegative supercoils; quinolones (GyrA), novobiocin (GyrB)
TerminationTer + Tus fork traps; topo IV decatenates; XerCD at dif resolves dimers
Eukaryotic licensingORC + Cdc6 + Cdt1 load MCM2–7 in G1; CDK + DDK fire CMG in S; geminin
Eukaryotic polymerasesα prime, ε leading, δ lagging (PCNA, RFC), γ mito, β BER, η TLS
TelomeraseTERT + RNA template; TTAGGG in humans
Fidelity10⁻⁴–10⁻⁵ → proofreading → MMR → 10⁻⁹–10⁻¹⁰
PlasmidsColE1: RNA I/RNA II + Rop; rolling circle: Rep nicks dso; incompatibility
mtDNA16,569 bp human; Pol γ; D-loop
Direct reversalPhotolyase (not in placental mammals), MGMT/Ada (suicide), AlkB
BERGlycosylase → APE1 → Pol β → ligase
NERUvrABC, 12–13 nt; human ~30 nt, XPA–XPG, TFIIH; CSA/CSB for TC-NER
MMRMutS → MutL → MutH nicks unmethylated GATC; Lynch syndrome
DSBHR (MRN, BRCA2, Rad51) in S/G2; NHEJ (Ku, DNA-PKcs, ligase IV)
SOSRecA* → LexA autocleavage; Pol V error-prone; also cleaves λ cI
HR proteinsRecBCD + Chi (GCTGGTGG), RecA, RuvAB, RuvC
Site-specificTyr (Cre, λ Int, FLP, XerCD) vs Ser (Hin, resolvases); direct → deletion, inverted → inversion

Key Takeaways

What’s Next