DNA Replication

DNA replication is the process by which a cell makes an exact copy of its DNA before dividing. Every time a cell divides — through mitosis or meiosis — it must first duplicate its entire genome so each daughter cell receives a complete set. The process is semiconservative (each new double helix contains one parental strand and one newly synthesized strand), bidirectional (proceeds in both directions from each origin of replication), and astonishingly accurate (roughly one error per billion base pairs copied, thanks to proofreading mechanisms). Watson and Crick’s 1953 DNA structure paper foreshadowed the replication mechanism in a single famous closing sentence; Meselson and Stahl proved it experimentally in 1958.

Semiconservative Replication

When DNA replicates, the two parental strands separate and each one templates the synthesis of a new complementary strand. The result: two double helices, each containing one parental strand and one daughter strand. This is what ‘semiconservative’ means.

Three other models were considered in the 1950s — conservative (parental helix stays together; new helix is fully new), dispersive (parental and daughter DNA is mixed throughout both new helices), and a ‘mixed’ hybrid model. Meselson and Stahl’s 1958 experiment with nitrogen-15 isotope labeling definitively proved the semiconservative model. The experiment is regarded as one of the most elegant in the history of biology.

The Replication Fork

DNA replication initiates at specific sites called origins of replication. The DNA double helix unwinds at the origin, forming a replication bubble that expands outward in both directions. Each end of the bubble is a replication fork — a Y-shaped junction where the parental DNA splits into two single strands and new DNA is being synthesized.

A single chromosome can have hundreds or thousands of origins of replication firing simultaneously. The human genome (3 billion base pairs per haploid set) replicates in about 6-8 hours during S phase because so many forks operate in parallel.

The Key Enzymes

DNA replication requires the coordinated work of several enzymes. The major players:

• Helicase — unwinds the double helix at the replication fork by breaking the hydrogen bonds between complementary bases.
• Topoisomerase — relieves the torsional strain that builds up as the helix unwinds ahead of helicase. Without it, the DNA would tangle.
• Single-strand binding proteins (SSBs) — coat the separated single strands to prevent them from re-pairing or forming secondary structure before new synthesis occurs.
• Primase — synthesizes short RNA primers that DNA polymerase needs to start synthesis (DNA polymerase can extend an existing strand but can’t start from scratch).
• DNA polymerase — the main synthesis enzyme. Reads the template strand 3′ → 5′ and synthesizes the new strand 5′ → 3′, adding one complementary nucleotide at a time. Multiple types exist; in humans, Pol α, δ, and ε are the main replication polymerases.
• DNA ligase — joins discontinuous DNA fragments on the lagging strand into a continuous strand.

Leading vs Lagging Strand

Here’s the trick that makes DNA replication mechanically interesting: DNA polymerase can only synthesize in one direction (5′ → 3′). But the two parental strands run in opposite (antiparallel) directions. So at each fork, one parental strand can be copied continuously while the other must be copied in pieces.

• Leading strand. The strand whose template runs 3′ → 5′ in the direction the fork is moving. DNA polymerase follows the fork continuously, synthesizing in a single uninterrupted run.
• Lagging strand. The strand whose template runs 5′ → 3′ in the direction the fork is moving. DNA polymerase has to work backwards, synthesizing in short fragments (Okazaki fragments) that are then joined by DNA ligase. Each Okazaki fragment requires its own RNA primer to start.

Okazaki fragments are 100-200 nucleotides long in eukaryotes and 1,000-2,000 in bacteria. Reiji Okazaki discovered them in 1968.

Proofreading and Error Correction

DNA polymerase makes mistakes — pairing the wrong base about once every 10⁵ to 10⁶ nucleotides. Most polymerases have a 3′ → 5′ exonuclease activity that immediately removes incorrectly paired bases and tries again. This proofreading reduces the error rate to about 10⁻⁷ to 10⁻⁸ per nucleotide.

After replication, a separate mismatch repair (MMR) system scans new DNA for remaining errors and corrects them. The MMR machinery distinguishes the new strand (which is unmethylated, briefly) from the old strand (which is methylated). Together, polymerase proofreading and post-replication MMR reduce the final error rate to about 1 mistake per 10⁹ base pairs copied — astonishingly accurate for any biological process.

When MMR fails, the result is hereditary cancer syndromes like Lynch syndrome (defective MMR genes cause runaway mutation accumulation and high colorectal cancer risk).

Telomeres — The End Replication Problem

Linear chromosomes face a unique problem: the lagging strand’s last RNA primer leaves a gap at the 5′ end that cannot be filled (DNA polymerase has nothing to extend from beyond it). Without compensation, chromosomes would shorten with every replication.

Telomeres are repetitive DNA sequences (TTAGGG repeats in humans) at chromosome ends that act as buffers. Each replication shortens the telomeres slightly. When telomeres get critically short, cells stop dividing (replicative senescence). The enzyme telomerase can re-extend telomeres, but most somatic cells don’t express telomerase actively, which is one of the factors limiting cell division across a lifespan.

Telomerase is highly active in stem cells, germ cells, and most cancer cells. Cancer cells’ ability to re-extend their telomeres is one of the hallmarks of cancer and a target for some cancer therapies.

Related study notes: Nucleic Acid, Mitosis, Meiosis, Cell Cycle.

Frequently Asked Questions