Documents and resources

Everything here describes replication in prokaryotes, specifically Escherichia coli. Eukaryotes follow the same principle and the logic is nearly identical, but they use some different proteins and add steps that don't appear here. Where no organism is named, assume prokaryote.

Before we start

The seven proteins of DNA replication, one per slide: helicase · gyrase · SSB · primase · DNA polymerase III · DNA polymerase I · ligase. From the first opening of the helix to the final seal.

In this document

The seven proteins behind DNA replication , step by step. The whole process in prokaryotes, with E. coli as the model. How a cell copies itself

Central dogma

Every error that survives all three filters is passed down to every daughter cell. Most are silent, but the ones that land in cell cycle control genes are behind a good share of tumor processes.

Why it matters

DNA polymerase on its own slips up once every 100,000 nucleotides. Its 3'→5' exonuclease activity checks each nucleotide as it lands and drops the error rate to 1 in 10⁷. The mismatch repair system cleans up whatever got through and takes it to 1 in 10⁹. That fidelity is no single enzyme's doing.

Where the precision comes from

The human genome runs to about 3.2 billion base pairs. Escherichia coli , the organism in the slides ahead, has 4.6 million — roughly 700 times fewer, and it still takes 40 minutes to copy it all. The difference is strategy: E. coli opens a single origin of replication; a human cell opens tens of thousands at once.

The real scale

And do it in a matter of hours, every time a cell divides. With an error rate of 1 in a billion nucleotides. 3.2Copy it all. Miss nothing. billion base pairs in humans

The problem

Mutations in RecQ family helicases — WRN among them — don't affect this routine opening, but the resolution of difficult structures. That's where Werner syndrome comes from.

The connection

Every cycle of opening burns ATP. On its own it moves slowly; coupled to DNA polymerase III it reaches around 1,000 base pairs per second. Helicase and polymerase don't travel one behind the other: they are bound into a single complex, the replisome, and speed each other up.

The cost

The DnaB helicase of E. coli is a ring-shaped hexamer that wraps around the lagging strand template and travels 5' →3'. It breaks only the hydrogen bonds between complementary bases; the sugar-phosphate backbone is left untouched. Don't confuse it with gyrase: helicase separates the strands, gyrase deals with the tension that separation creates up ahead. Different problems, different solutions.

The mechanism

Helicase doesn't land just anywhere. E. coli has a single origin of replication, called oriC, made of repeats rich in adenine and thymine. That pair is held by two hydrogen bonds against the three of guanine-cytosine, which makes it the easiest stretch of the whole chromosome to pry open. The initiator protein DnaA recognizes the site, separates the first bases, and loads helicase there.

Where it begins

It breaks the hydrogen bonds between bases and pulls the two strands apart. The replication fork takes shape. HelicaseOpening the helix

Step 1 of 7

Quinolones — ciprofloxacin, levofloxacin — are antibiotics that block bacterial gyrase: they trap the enzyme at the exact moment it has cut the DNA but not yet resealed it, and the genome is left in pieces. Since humans have no gyrase, the drug acts on a target that exists only in the bacterium. That's selective toxicity, not harmlessness: at high concentrations quinolones do reach human topoisomerase II, which is why they carry documented adverse effects.

The connection

It is the only topoisomerase that can introduce negative supercoils against thermodynamic equilibrium, which is why it needs ATP hydrolysis. Type I topoisomerases, by contrast, only relax existing tension and use no energy at all.

The cost

When the helix opens, the torsion doesn't simply vanish: it piles up ahead of the fork as positive supercoils. DNA gyrase — a topoisomerase absent from human cells — makes a transient cut through both strands, passes a segment of the double helix through the break, and seals it again. The net result is the introduction of negative supercoils that offset the tension.

The mechanism

Opening the helix twists the DNA up ahead. Gyrase cuts both strands, releases the torsion , and seals them back. DNA gyraseRelieving the tension

Step 2 of 7

It prevents two accidents at once: complementary strands pairing back up, and a single strand folding onto itself into internal hairpins. Either one would block the polymerase from getting through.

The connection

It uses no ATP and catalyzes no reaction: it's a structural protein, not an enzyme. Its role is purely structural, and even so, without it the fork would close almost as fast as it opens.

The cost

In E. coli it works as a tetramer and binds with no sequence specificity: it doesn't care which bases are there, only that the strand is single. Its binding is cooperative — each tetramer makes it easier for the next one to arrive, so they coat entire stretches rather than isolated points.

The mechanism

A single strand is fragile: it can zip back up or get degraded . Single-strand binding proteins (SSB) keep it open and available. SSBProtecting what’s exposed

Step 3 of 7

The limitation on DNA polymerases is structural and absolute: they can only extend an existing strand, never start one . Every bit of complexity on the lagging strand — repeated fragments, multiple primers, the cleanup afterward — exists solely because of that restriction.

Why it matters

It doesn't work alone: it physically associates with helicase to form a complex called the primosome. That partnership keeps primer placement in sync with the advancing fork, which is critical on the lagging strand, where a great many are needed.

The cost

Primase (DnaG in E. coli ) is really a specialized RNA polymerase. Unlike DNA polymerases, it can start a strand from nothing, and it lays down a short stretch of RNA — around 11 or 12 nucleotides — complementary to the template. That primer supplies the free 3'-OH end that DNA polymerase III needs to latch on. That ability to start from nothing is shared by all RNA polymerases, and it's the reason transcription needs no primer.

The mechanism

No DNA polymerase can start from nothing: they all need a free 3' end to pull from. Primase builds that starting point — an RNA primer. PrimaseThe starting point

Step 4 of 7

Both strands are synthesized at the same time, by the same enzyme complex. To pull that off, the lagging strand forms a loop that temporarily reorients it, so both polymerases can move together despite reading in opposite directions.

Why it matters

It reaches around 1,000 nucleotides per second, the same pace as helicase because the two are coupled in the replisome. But on its own it would slip off the template every few dozen bases. The β sliding clamp — a protein ring that locks it onto the DNA — is what keeps it attached. Without that clamp, its processivity would fall to a few dozen nucleotides per binding event.

The cost

DNA polymerase III only adds nucleotides in the 5' →3' direction. Because the strands are antiparallel, that single direction means opposite things on each one: on the leading strand it lines up with the advancing fork and synthesis runs uninterrupted; on the lagging strand it points the other way, so the enzyme has to let go and start over each time fresh template is exposed. Those stretches run 1,000 to 2,000 nucleotides in E. coli.

The mechanism

Continuous DNA synthesis. One primer. Moves in the same direction the fork opens. Discontinuous DNA synthesis. One primer per Okazaki fragment. Moves against the opening. Leading strand Lagging strandDNA polymerase IIITwo strands, two strategies

Step 5 of 7

Leaving RNA inside DNA would compromise the molecule on two fronts. Ribose is chemically less stable than deoxyribose. And uracil, inside DNA, means damage: cytosine deaminates spontaneously into uracil, so the cell patrols the genome with an enzyme that excises any uracil it finds, on the assumption that it used to be a broken cytosine. The cleanup isn't cosmetic, it's structural.

Why it matters

It is far slower than polymerase III — around 20 nucleotides per second — but it only has to cover the short stretches where primers sit, not the whole genome. Each enzyme is tuned to its own job.

The cost

DNA polymerase I has 5' →3' exonuclease activity, which lets it chew through the RNA primer at the front while filling the gap with DNA nucleotides at the back. That coordinated advance is called nick translation. It is the only polymerase in the process that degrades and synthesizes at once.

The mechanism

Primers are made of RNA: they carry uracil, not thymine. DNA polymerase I strips them out and fills the gap with DNA. DNA polymerase IRemoving the RNAS

Tep 6 of 7

T4 DNA ligase is a cornerstone tool of molecular cloning: it's the enzyme that joins insert to vector when building recombinant DNA. The same mechanism that closes Okazaki fragments is what lets us build plasmids in the lab.

Why it matters

E. coli ligase runs on NAD ⁺ rather than ATP. It's a notable difference: eukaryotic ligase and T4 phage ligase — the one used in molecular biology — both depend on ATP.

The cost

After polymerase I finishes filling in, a nick is left behind: the nucleotides are complete and correctly paired, but the covalent bond between the 5' phosphate of one and the 3' hydroxyl of the next is missing. Ligase catalyzes that join and turns two neighboring fragments into a single continuous strand.

The mechanism

The nucleotides are all in place, but the backbone is still broken. Ligase forms the phosphodiester bond that closes the nick. DNA ligaseSealing the strand

Step 7 of 7

It's what guarantees genetic information passes down generation after generation without degrading: every copy is built on an original template, never on a copy of a copy.

Why it matters

Meselson and Stahl settled it in 1958 by growing E. coli in heavy nitrogen (¹⁵N) and then switching it to ordinary nitrogen (¹⁴N). After one generation, all the DNA came out at intermediate density, which ruled out the conservative model. After two generations, two bands appeared — one intermediate, one light — which ruled out the dispersive one. Only the semiconservative model predicted that exact outcome.

How it was proven

Each daughter molecule keeps one of the original molecule's two strands intact and builds a brand-new complementary partner. That's why the model is called semiconservative: half of it is conserved. The alternatives on the table at the time were the conservative model — the original molecule stays whole and an entirely new copy appears — and the dispersive one, with old and new fragments mixed throughout both strands.

The mechanism

Each with one old strand and one new one. That's what makes replication semiconservative. Two identical molecules

The result

The whole process, animated and explained in full, in the Central Dogma video. And this is happening right now - Nelson, D. L., & Cox, M. M. (2021). Lehninger principles of biochemistry (8th ed.). Macmillan Learning. - Alberts, B., Heald, R., Johnson, A., Morgan, D., Raff, M., Roberts, K., & Walter, P. (2022). Molecular biology of the cell (7th ed.). W. W. Norton & Company. - Meselson, M., & Stahl, F. W. (1958). The replication of DNA in Escherichia coli. Proceedings of the National Academy of Sciences of the United States of America, 44(7), 671–682. https://doi.org/10.1073/pnas.44.7.671 - Kunkel, T. A. (2004). DNA replication fidelity. Journal of Biological Chemistry, 279(17), 16895–16898. https://doi.org/10.1074/jbc. R400006200 - O'Donnell, M., Langston, L., & Stillman, B. (2013). Principles and concepts of DNA replication in bacteria, archaea, and eukarya. Cold Spring Harbor Perspectives in Biology, 5(7), a010108. https://doi.org/10.1101/cshperspect.a010108

References

The interactive Central Dogma minigame is at aequorea.net, alongside the rest of the series' downloadable resources. The full video walks through every one of these steps with 3D animation.

Keep studying

With two identical DNA molecules, the cell is ready to divide. But having the information copied is no use if it can't be read: that's the job of transcription, folder 3 in this series.