Science Wiki

DNA replication is the highly coordinated, semi-conservative biological process by which a dividing cell synthesizes an exact duplicate copy of its entire DNA genome during the synthesis (S) phase of the cell cycle. This process is the physical prerequisite for all cell division—including vegetative binary fission in prokaryotes, somatic mitosis for tissue growth and regeneration, and gametogenic meiosis for sexual reproduction—ensuring that every daughter cell inherits a complete, high-fidelity set of genetic instructions.

As demonstrated in 1958 by Matthew Meselson and Franklin Stahl using ¹⁵N/¹⁴N equilibrium density-gradient centrifugation in *Escherichia coli*, DNA replication operates via a semi-conservative mechanism. When the parental double helix is unwound, each of the two separated polynucleotide strands serves directly as a base-pairing template for the polymerization of a new complementary daughter strand. Consequently, each resulting double-stranded DNA molecule retains one conserved parental strand paired with one newly synthesized daughter strand.

Prokaryotic *oriC* vs. Eukaryotic Multi-Origin Licensing

Replication does not initiate randomly along a chromosome; it begins at defined cis-acting genomic sequences known as origins of replication, regulated to ensure that the genome is replicated once and only once per cell division:

• Prokaryotes: Bacteria with circular chromosomes typically possess a single conserved origin of replication (*oriC*, ~245 bp in *E. coli*). When ATP-bound initiator proteins (DnaA) oligomerize at 9-mer DnaA boxes within *oriC*, they introduce torsional stress that melts an adjacent AT-rich 13-mer DNA unwinding element (DUE), allowing the hexameric loader DnaC to deposit two hexameric DnaB helicase rings onto the separated single strands.
• Eukaryotes: Because eukaryotic chromosomes are massive and linear (~6.4 Gb in diploid human cells), replication initiates across 30,000 to 50,000 origins distributed throughout the cell nucleus. Eukaryotic origin control is split into two mutually exclusive cell-cycle phases: during G1 phase (Origin Licensing), the six-subunit Origin Recognition Complex (ORC1–6) binds origin DNA and collaborates with Cdc6 and Cdt1 to load head-to-head double hexamers of the inactive MCM2–7 helicase core around duplex DNA; during the G1/S transition (Origin Firing), rise in DDK (Cdc7–Dbf4) and S-CDK kinase activities recruits Cdc45 and the GINS tetramer to convert MCM2–7 into two active CMG (Cdc45–MCM2–7–GINS) replicative helicases while inhibiting re-licensing until the next mitosis.

Helicase Unwinding, SSB/RPA Coating, and Topoisomerase Relief

Once activated at the origin, the replicative helicase (DnaB in bacteria, translocating 5′→3′ on the lagging-strand template; CMG in eukaryotes, translocating 3′→5′ on the leading-strand template) hydrolyzes ATP to break the non-covalent hydrogen bonds between complementary base pairs. This unwinding creates two Y-shaped replication forks that move bidirectionally outward from the origin.

To prevent the exposed single-stranded templates from reannealing into a duplex or forming intramolecular hairpin structures (and to shield them from endonucleolytic degradation), single-strand binding proteins (SSB homotetramers in bacteria; heterotrimeric Replication Protein A, RPA, in eukaryotes) rapidly coat the unwound single-stranded DNA. Simultaneously, because the two parental strands are plectonemically intertwined (~1 helical turn every 10.5 base pairs), forceful unwinding by helicase generates severe positive supercoiling (overwinding torsional strain) ahead of the advancing replication fork. Specialized topoisomerase enzymes—specifically DNA gyrase in bacteria and Topoisomerase I and IIα in eukaryotes—continuously cleave one or both strands of the DNA backbone ahead of the fork, pass the duplex through the break to relax torsional stress, and reseal the phosphodiester bond.

Primase Initiation and 5′→3′ Polymerase Polarity

The chemical polymerization of nascent DNA strands is catalyzed by high-fidelity replicative DNA polymerase complexes (Pol III holoenzyme in bacteria; Pol ε and Pol δ in eukaryotes), which are locked onto the template by toroidal sliding clamps (β-clamp or PCNA). However, all known replicative DNA polymerases operate under two invariant biochemical constraints:

1. They cannot initiate synthesis *de novo* on a bare single-stranded template; they strictly require a pre-existing base-paired primer providing a free 3′-hydroxyl (3′-OH) group.
2. They synthesize DNA exclusively in the 5′ → 3′ direction (by catalyzing inline nucleophilic attack of the primer 3′-OH onto the α-phosphate of an incoming dNTP while reading the template strand 3′ → 5′).

To overcome the first constraint, a specialized RNA polymerase called primase (DnaG in bacteria; the Pol α–primase complex in eukaryotes) binds the single-stranded template and synthesizes a short, 8- to 12-nucleotide RNA primer *de novo*. In eukaryotes, the Pol α catalytic subunit immediately extends this RNA primer with ~20 deoxyribonucleotides to form a ~30-nt chimeric RNA-DNA initiator primer before the RFC clamp loader displaces Pol α, loads a PCNA ring, and triggers a "polymerase switch" to processive Pol ε or Pol δ.

Continuous Leading-Strand vs. Discontinuous Lagging-Strand Synthesis

Because the two parental strands of the DNA double helix are antiparallel (one running 3′→5′ and the other 5′→3′ into the replication fork), the 5′→3′ restriction of DNA polymerases forces the replisome to execute two topologically distinct modes of synthesis simultaneously (semi-discontinuous replication):

• The Leading strand: Built on the parental template strand oriented 3′ → 5′ toward the advancing replication fork. Following a single initial priming event at the origin, the leading-strand polymerase (Pol III core in bacteria; DNA polymerase ε physically tethered to the CMG helicase in eukaryotes) synthesizes the nascent daughter strand continuously in the 5′ → 3′ direction, moving in the same vector as fork unwinding.
• The Lagging strand: Built on the opposing parental template strand oriented 5′ → 3′ toward the opening fork. Because DNA polymerase must synthesize in the 5′ → 3′ direction (away from the advancing fork), the lagging strand must be synthesized discontinuously in a "backstitching" trombone-loop cycle. As helicase unwinds an additional stretch of template DNA, primase repeatedly synthesizes a new RNA primer near the fork junction. The lagging-strand polymerase (Pol III core in bacteria; DNA polymerase δ bound to PCNA in eukaryotes) extends each primer in the 5′ → 3′ direction until it runs into the 5′ end of the previous fragment, generating a series of discrete DNA segments known as Okazaki fragments (1,000–2,000 nt long in bacteria; 100–200 nt long in eukaryotes, sized to the ~165-bp nucleosome repeat length).

Flap Excision (Pol I / FEN1) and DNA Ligase Nick Sealing

To convert the millions of RNA-primed Okazaki fragments on the lagging strand into a continuous genomic DNA molecule, all RNA ribonucleotides must be quantitatively removed and replaced with deoxyribonucleotides:

• In bacteria, RNase HI degrades the bulk of the RNA primer, and DNA polymerase I binds the nick, extends the upstream Okazaki fragment 5′→3′, and simultaneously uses its N-terminal 5′→3′ flap endonuclease domain to excise the remaining ribonucleotides (nick translation).
• In eukaryotes, as Pol δ finishes an Okazaki fragment and encounters the downstream RNA-DNA initiator primer synthesized by error-prone Pol α, Pol δ executes limited strand-displacement synthesis, pushing the 5′ RNA-DNA segment up into a single-stranded 5′ flap. Flap Endonuclease 1 (FEN1) (assisted by the helicase/nuclease DNA2 and RNase H2 for longer RPA-coated flaps) threads and cleaves the base of the 5′ flap to create a flush, base-paired nick.

Once the primer is replaced, a single phosphodiester bond remains missing between the 3′-OH of the upstream fragment and the 5′-phosphate of the downstream fragment. DNA ligase (LigA using NAD⁺ in bacteria; DNA ligase I tethered to PCNA and using ATP in eukaryotes) catalyzes the three-step adenylylation and nucleophilic attack that seals the nick into an intact sugar-phosphate backbone.

Replication Termination and Topoisomerase II Decatenation

Replication terminates when two converging replication forks meet. In *E. coli*, forks converge in the terminus region opposite *oriC*, where polar Tus proteins bound to *Ter* sequences block DnaB progression from one direction. In eukaryotes, when two converging CMG helicases meet, the ubiquitin ligase CRL2^Lrr1 (or SCF^Dia2 in yeast) polyubiquitinates the MCM7 subunit, recruiting the p97/VCP (Cdc48) segregase to extract the CMG complex from chromatin. Finally, because unwinding the last turns of a duplex creates topological catenanes (intertwined daughter rings or sister chromatids), Topoisomerase IV (in bacteria) or Topoisomerase IIα (in eukaryotes) executes a double-strand passage reaction to decatenate the two daughter chromosomes so they can segregate during cell division.

3′→5′ Exonucleolytic Proofreading and Mismatch Repair

Maintaining genetic fidelity across billions of base pairs is essential to prevent lethal mutations or somatic cancer. Cellular DNA replication achieves an overall error rate of approximately 1 mistake per 10⁹ to 10¹⁰ base pairs polymerized through three multiplicative fidelity filters:

1. Active-Site Geometric Selection (~10⁻⁴ to 10⁻⁵ error rate): The polymerase Fingers domain closes efficiently only when an incoming dNTP forms a canonical Watson-Crick base pair (adenine with thymine; guanine with cytosine) with a ~10.6 Å C1′–C1′ span.
2. 3′ → 5′ Exonucleolytic Proofreading (~10⁻² escape rate, yielding ~10⁻⁷ cumulative error rate): Replicative polymerases (Pol III ε-subunit in bacteria; intrinsic exonuclease domains of Pol ε and Pol δ in eukaryotes) possess a 3′ → 5′ proofreading exonuclease site located ~30–40 Å from the polymerase catalytic cleft. If a non-complementary nucleobase is misincorporated, the frayed 3′ mismatch stalls forward extension, allowing the primer 3′ terminus to melt and partition into the exonuclease site, where the incorrect nucleotide is hydrolytically excised before the primer returns to the polymerase cleft.
3. Post-Replicative DNA Mismatch Repair (MMR) (~10⁻² to 10⁻³ escape rate, yielding ~10⁻⁹ to 10⁻¹⁰ final error rate): Immediately behind the replication fork, the MutS/MutL complex (MSH2–MSH6 and MLH1–PMS2 in humans) scans the nascent duplex, detects any unproofread base mismatches or slippage loops, discriminates the newly synthesized daughter strand via transient Okazaki nicks or PCNA orientation (or GATC hemimethylation via MutH in *E. coli*), excises the error-containing tract, and directs high-fidelity resynthesis.

Telomere Attrition and Telomerase Reverse Transcription

Whereas circular bacterial chromosomes can be completely replicated around their entire circumference, the linear architecture of eukaryotic chromosomes introduces an inescapable mechanical challenge at the extreme 3′ ends of the lagging-strand template—formulated independently by James Watson and Alexey Olovnikov in 1971–1972 as the end-replication problem. When the terminal RNA primer at the very tip of a linear chromosome is excised (and when 5′→3′ exonucleases resect the blunt leading-strand end to create a protective 3′ overhang for shelterin binding), there is no upstream 3′-OH primer available for DNA polymerase to fill the terminal gap. Consequently, somatic chromosomes shorten by 50 to 200 base pairs per cell division.

To prevent this progressive terminal erosion from deleting essential protein-coding genes, the ends of linear eukaryotic chromosomes are buffered by multi-kilobase tracts of non-coding hexanucleotide repeats (5′-TTAGGG-3′ in vertebrates) known as telomeres. In germline cells, embryonic stem cells, hematopoietic progenitors (and reactivated in ~85–90% of malignant human tumors), telomere shortening is counteracted by the ribonucleoprotein enzyme telomerase. Telomerase consists of a catalytic protein subunit (TERT, Telomerase Reverse Transcriptase) and an intrinsic RNA subunit (TERC / hTR, containing the template sequence 3′-CAAUCCCAAUC-5′). Acting as a specialized reverse transcriptase, telomerase binds the single-stranded 3′ G-rich telomeric overhang and iteratively synthesizes de novo 5′-TTAGGG-3′ DNA repeats onto the chromosome 3′ terminus, providing extended template length for Pol α–primase to synthesize the complementary C-rich strand.

How did the 1958 Meselson–Stahl experiment prove that DNA replication is semi-conservative rather than conservative or dispersive?

Matthew Meselson and Franklin Stahl grew *Escherichia coli* for many generations in medium containing the heavy nitrogen isotope ¹⁵N (so all parental DNA had a high buoyant density), then shifted the bacteria to light ¹⁴N medium and centrifuged extracted DNA in cesium chloride (CsCl) density gradients. After exactly one generation, 100% of the DNA formed a single intermediate hybrid band (¹⁵N/¹⁴N), ruling out the conservative model (which predicted one heavy and one light band). After two generations, the DNA split into equal 50% intermediate hybrid (¹⁵N/¹⁴N) and 50% light (¹⁴N/¹⁴N) bands, ruling out the dispersive model and proving semi-conservative strand separation.

How does a human cell prevent its 30,000 to 50,000 replication origins from firing more than once during a single S phase?

Eukaryotes enforce strict "once-and-only-once" replication per cell cycle by temporally separating origin licensing (loading inactive MCM2–7 helicase double hexamers onto DNA) from origin firing (activating MCM2–7 into CMG helicases): licensing is permitted only during G1 phase when Cyclin-Dependent Kinase (CDK) activity is low and Geminin is degraded by APC/C. As soon as S phase begins, rising S-CDK and DDK kinase levels trigger origin firing while simultaneously phosphorylating and inactivating ORC, Cdc6, and Cdt1 (and stabilizing Geminin in metazoans), making it biochemically impossible to load new MCM2–7 helicases onto already-replicated DNA until the cell completes mitosis.

Why are eukaryotic Okazaki fragments (~100–200 nucleotides) roughly ten times shorter than bacterial Okazaki fragments (~1,000–2,000 nucleotides)?

In bacteria, lagging-strand synthesis proceeds across naked or loosely NAP-bound DNA, allowingPol III to extend each Okazaki fragment for 1,000 to 2,000 nucleotides before terminating. In eukaryotes, newly synthesized lagging-strand DNA behind the replication fork is immediately wrapped into nucleosomes by histone chaperones (such as CAF-1) every ~165–200 base pairs. When the advancing lagging-strand DNA polymerase δ encounters a newly assembled nucleosome positioned on the downstream Okazaki fragment, steric hindrance halts strand-displacement synthesis and triggers FEN1 flap cleavage and DNA ligase I sealing right at the ~165-bp internucleosomal linker boundary.