Intermediatemolecular
DNA Replication in Detail
Overview
DNA replication copies the genome before every cell division with remarkable fidelity (about one error per 10^9 bases in humans, after proofreading and repair). The double helix unwinds at origins, each strand templates a new complement, and the result is two semiconservative duplexes. This lesson covers the enzymes, the leading/lagging strand asymmetry, and how the replisome coordinates it all.
Learning objectives
- Explain semiconservative replication and the Meselson-Stahl proof
- List replisome enzymes and their functions
- Contrast leading vs lagging strand synthesis
- Describe replication origins and licensing
Semiconservative: the Meselson-Stahl experiment
Meselson and Stahl (1958) grew E. coli in heavy 15N, shifted to light 14N, and found first-generation DNA at hybrid density, then half hybrid half light: exactly the semiconservative prediction (each daughter duplex keeps one parental strand). Conservative and dispersive models were ruled out in one elegant experiment.
Meselson-Stahl is routinely voted the most elegant experiment; it settled replication's mechanism definitively.
Unwinding the helix
Helicase breaks hydrogen bonds ahead of the fork; single-strand binding proteins (SSB) keep strands apart; topoisomerase (DNA gyrase in bacteria) relieves supercoiling ahead; primase synthesizes short RNA primers because DNA polymerases cannot start de novo. In eukaryotes, the MCM helicase is loaded at origins during G1 (licensing) and fired in S phase.
All DNA polymerases extend 3'-OH ends; none can initiate chains. Primase provides the starter.
Leading vs lagging strands
Polymerases synthesize only 5'->3', so one strand (leading) is made continuously toward the fork while the other (lagging) is made discontinuously away from the fork as Okazaki fragments. DNA pol III (bacteria) / pol delta and epsilon (eukaryotes) extend; pol I / FEN1 removes primers; ligase seals nicks.
The antiparallel helix plus unidirectional synthesis forces the lagging-strand workaround.
Fidelity: proofreading
Replicative polymerases have 3'->5' exonuclease proofreading: mismatched bases stall synthesis and are excised before extension continues. This improves fidelity ~100-fold over base selection alone, from ~10^-5 to ~10^-7 errors per base; mismatch repair adds another ~100-fold.
Even 10^-9 leaves ~6 new mutations per human diploid genome per cell division; repair is superb but not perfect.
Key points
- Replication is semiconservative: each daughter duplex keeps one parental strand
- Helicase, SSB, topoisomerase, primase, polymerases, ligase coordinate at the fork
- Leading strand is continuous; lagging strand is made as Okazaki fragments
- Proofreading (3'->5' exonuclease) gives ~10^-7 fidelity before repair
Common misconceptions
✓ Truth: Meselson-Stahl proved semiconservative: first-generation DNA was uniformly hybrid density, impossible under conservative or dispersive models.
✓ Truth: Only the leading strand is continuous; the lagging strand must be made in fragments because polymerases work 5'->3' only.
Practice questions
Q1. Meselson-Stahl's first-generation DNA at hybrid density proved replication is
- Conservative
- Semiconservative
- Dispersive
- Random
One parental strand per daughter duplex gives uniformly intermediate density.
Q2. The lagging strand is synthesized discontinuously because
- It has no template
- DNA polymerase synthesizes only 5'->3'
- Helicase is slow
- Primers are absent
Moving away from the fork, synthesis must restart repeatedly as new template is exposed.
Q3. Proofreading by DNA polymerase uses
- 5'->3' exonuclease
- 3'->5' exonuclease
- Helicase
- Ligase
The 3'->5' exonuclease removes mismatched 3' ends before extension continues.
Click a question to reveal the answer and explanation.
Frequently asked questions
Why can't DNA polymerase start a new chain?
It requires a pre-existing 3'-OH to extend; primase (an RNA polymerase) can initiate, providing the primer.
What limits replication speed?
Helicase unwinding, polymerase processivity (clamp proteins), and in eukaryotes the sheer number of origins (~30,000-50,000 in humans).
How is re-replication prevented?
Origin licensing: MCM helicases load only in G1 when CDK is low; high S-phase CDK blocks reloading until the next G1.