DNA Replication: The Molecular Machinery of Copying Life

Opening the double helix

DNA replication starts at an origin of replication, a specific sequence where initiator proteins bind and recruit the enzyme helicase. Helicase is a molecular motor that breaks the hydrogen bonds holding the two complementary strands together, unzipping the double helix and creating a Y-shaped replication fork that moves along the chromosome. Because DNA is a right-handed double helix, unwinding it ahead of the fork twists the DNA further downstream — a strain relieved by topoisomerase, which nicks and reseals a strand to let the accumulating supercoiling dissipate. Single-strand binding proteins then coat the exposed single strands so they don't snap back together before they can be copied.

The rule that shapes everything: 5' to 3' only

The enzyme that actually builds new DNA, DNA polymerase III, can only add nucleotides to the free 3' hydroxyl end of an existing strand — it can extend a chain 5' to 3', never the reverse. That single mechanical constraint forces the whole replication apparatus into an asymmetric design, because the two template strands run antiparallel to each other.

template (leading):  3' — — — — — — — — — 5'   (fork moves →)
new leading strand:  5' — — — — — — — — — 3'   continuous synthesis

template (lagging):  5' — — — — — — — — — 3'   (fork moves →)
new lagging strand:      3'←frag  3'←frag  3'←frag   (built in short
                                                        backward-facing
                                                        pieces, then joined)

On the leading strand, the template happens to run in the direction that lets DNA polymerase III follow the fork continuously, needing only a single RNA primer laid down once by primase at the start. On the lagging strand, the template runs the wrong way for continuous synthesis, so the cell copies it backward in short bursts.

Okazaki fragments and the lagging strand

Every time the fork exposes a fresh stretch of lagging-strand template, primase lays down a short RNA primer, and DNA polymerase III extends it until it runs into the RNA primer of the previous fragment — a piece called an Okazaki fragment, roughly 100 to 200 nucleotides long in human cells (and 1,000 to 2,000 in bacteria). A different enzyme, DNA polymerase I, then removes each RNA primer and fills the resulting gap with DNA, and DNA ligase seals the final nick between fragments by forming the last phosphodiester bond. The result is a lagging strand stitched together from dozens of separately started and separately finished pieces — mechanically messier than the leading strand, but chemically identical once ligase is done.

Speed, fidelity and proofreading

A human replication fork moves at roughly 50 nucleotides per second — slow compared with bacterial DNA polymerase III, which can exceed 1,000 nucleotides per second, because eukaryotic DNA is spooled around histone proteins that must be temporarily displaced ahead of the fork. Human chromosomes compensate by firing thousands of origins simultaneously rather than relying on a single fork to cover the whole genome. Fidelity comes from two layers of error correction: DNA polymerase III has a built-in 3' to 5' exonuclease that proofreads each newly added base and excises a mismatch immediately, and a separate mismatch repair system scans the finished strand afterward. Together they bring the final error rate down to about one mistake in every billion base pairs copied.

Frequently asked questions

Try it live

Everything above runs in your browser — open DNA Replication and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.