DNA replication is the process by which a cell copies its DNA before it divides. It is semi-conservative: the two strands of the double helix separate, each acts as a template, and each new DNA molecule ends up with one original strand and one new strand. Helicase unwinds the helix at a replication fork by breaking the hydrogen bonds between bases. Primase lays down a short RNA primer. DNA polymerase III adds free nucleotides by complementary base pairing, always building the new strand in the 5' to 3' direction. Because the strands are antiparallel, one strand (leading) is made continuously and the other (lagging) in short Okazaki fragments. DNA polymerase I replaces the primers with DNA, and DNA ligase joins the fragments.

The enzyme names here are the prokaryotic ones, from E. coli, as used by AP Biology and most A level boards. Eukaryotic differences are at the end.

The structure you need first

Four facts about structure do almost all the work.

1. DNA is a double helix of two polynucleotide strands. Each nucleotide is a deoxyribose sugar, a phosphate group and one of four bases: adenine (A), thymine (T), guanine (G), cytosine (C).
2. The bases pair in only one way. A with T (two hydrogen bonds), G with C (three hydrogen bonds). So each strand carries all the information needed to rebuild the other.
3. The strands are antiparallel. Each strand has a 5' end (with a free phosphate) and a 3' end (with a free hydroxyl on the sugar). One strand runs 5' to 3', the other runs 3' to 5' alongside it.
4. The backbone is held by phosphodiester bonds; the strands are held together by hydrogen bonds. The backbone is strong. The hydrogen bonds between strands are individually weak, which is why the strands can be separated without breaking the chain.

Fact 3 is the reason the lagging strand exists.

What semi-conservative means

There were three ideas about how DNA might copy itself.

Watson and Crick suggested the semi-conservative model in 1953, because base pairing made it the obvious mechanism. It was proved by experiment five years later.

The Meselson and Stahl experiment

Examiners ask you to interpret the bands, not just recall the result.

Set up. In an experiment published in 1958, Matthew Meselson and Franklin Stahl grew E. coli for many generations in a medium where the only nitrogen was the heavy isotope nitrogen-15. Nitrogen is in every base, so all the bacteria's DNA became heavy. They then moved the bacteria into a medium with normal, light nitrogen-14 and took samples as the bacteria went through each generation.

Method. They extracted the DNA and spun it in a caesium chloride solution in an ultracentrifuge. The salt forms a density gradient, and DNA settles at the band matching its own density: heavy DNA lower, light DNA higher.

Results.

Only semi-conservative replication predicts both results. After one round each molecule is one heavy strand and one light strand: intermediate. After two rounds, half the molecules are hybrid and half are entirely light.

A common question gives you a third generation and asks the ratio. The answer: one quarter intermediate, three quarters light.

The enzymes, in order

Replication starts at a specific sequence called an origin of replication. E. coli has one origin on its circular chromosome. From it, two replication forks move away in opposite directions, forming a bubble.

Two rules govern DNA polymerase, and every hard question comes from them. It can only add to an existing 3' end, which is why it needs a primer to start. And it can only build in the 5' to 3' direction. It reads the template strand 3' to 5' as it goes.

The nucleotides arrive as deoxyribonucleoside triphosphates (dATP, dTTP, dGTP, dCTP). Two phosphates are cut off as each is added, and that releases the energy that drives the reaction.

Leading and lagging strands

At the fork, the two template strands run in opposite directions. On one template, the direction polymerase must build in is the same direction the fork is moving. Polymerase III starts at one primer and simply follows the fork, building continuously. That new strand is the leading strand.

On the other template, building 5' to 3' means moving away from the fork. Polymerase can only make a short stretch before it runs out of unwound template behind it. So as the fork opens up more template, primase lays down a new primer near the fork, and polymerase III builds another short stretch back towards the previous one. Each stretch is an Okazaki fragment, named after Reiji and Tsuneko Okazaki, who identified them in the 1960s. In E. coli they are about 1,000 to 2,000 nucleotides long; in eukaryotes, roughly 100 to 200.

On the lagging strand, then, the full sequence is: primer, fragment, primer, fragment. Polymerase I removes each RNA primer and replaces it with DNA, which leaves a single nick between neighbouring fragments. Ligase seals each nick. The leading strand needs just one primer per fork; the lagging strand needs one per fragment.

A trick for drawing it: put the fork on the right, moving right. Draw the top template 3' to 5' left to right, so its new strand grows rightwards, continuously. Draw the bottom template 5' to 3', so its new strand grows leftwards, in short arrows that each point away from the fork.

Accuracy

DNA polymerase III makes a mistake roughly once in every 10,000 to 100,000 nucleotides as it adds them. It also proofreads: it can step back, cut out the wrong nucleotide and try again. Together with mismatch repair after replication, the final error rate is around one in a billion. The rare errors that remain are one source of mutations.

How eukaryotes differ

Most courses want only a few lines on this.

• Many origins. A human chromosome is far longer than a bacterial one and polymerase is slower in eukaryotes, so replication starts at many origins along each chromosome, tens of thousands across the human genome.
• Different polymerases. Eukaryotes have their own named polymerases. Polymerase alpha (with primase) starts each strand, polymerase epsilon mainly builds the leading strand, and polymerase delta mainly builds the lagging strand. The I and III names apply to prokaryotes.
• Linear chromosomes and telomeres. Because the lagging strand needs a primer, the very end of a linear chromosome cannot be fully copied, so chromosomes shorten slightly with each division. Telomeres, repeated non-coding sequences at the ends, take that loss. The enzyme telomerase can extend them in some cells, such as stem cells.
• Chromatin. Eukaryotic DNA is wound round histone proteins, which must be moved and replaced as the fork passes.
• When. In eukaryotes replication happens in the S phase of interphase, before mitosis or meiosis.

The mistakes that lose marks

• Saying helicase "unzips" by breaking the backbone. It breaks hydrogen bonds between bases, not phosphodiester bonds.
• Saying ligase joins bases. Ligase forms phosphodiester bonds in the backbone. Hydrogen bonds between bases form on their own.
• Getting the direction backwards. New strands grow 5' to 3'. The template is read 3' to 5'.
• Saying the lagging strand is made later. Both strands are made at the same time at each fork; the lagging strand is just made in pieces.
• Misreading the Meselson and Stahl bands. Generation one rules out conservative. You need generation two to rule out dispersive.
• Saying replication happens during mitosis. It happens in the S phase of interphase, before division.

A scene plan for a four minute video

Replication is a process with a moving centre, the fork, so the video should keep the fork in the same place on screen and let the strands move through it. How to explain a process in a video covers that technique.

1. A cell about to divide, and the question: how does each new cell get a full, correct copy?
2. The double helix flattened into a ladder: sugar-phosphate rails, base-pair rungs, 5' and 3' labelled at each end of each rail.
3. The three models side by side as coloured diagrams: old strands blue, new strands red.
4. Meselson and Stahl: three centrifuge tubes, the bands appearing one generation at a time.
5. The origin opening into a bubble with two forks.
6. One fork, close up: helicase splitting the rungs, single-strand binding proteins, gyrase ahead.
7. The leading strand: one primer, polymerase III following the fork.
8. The lagging strand: primer, fragment, primer, fragment, each arrow pointing away from the fork.
9. Polymerase I swapping RNA for DNA, ligase sealing the nicks.
10. Two finished molecules, each half blue and half red, back in two daughter cells.

The blue and red colouring, carried from scene 3 to scene 10, shows semi-conservative replication without a word. For more on building a biology video around one colour-coded idea, see how to make a biology explainer video for a school assignment, and photosynthesis, explained visually uses the same trick with two oxygens.

Make your own version

1. 1

   Check your specification

   AQA, OCR, Edexcel and AP Biology differ on whether you need primase, polymerase I, Okazaki fragments and eukaryote polymerases. Cut what your course does not test.

2. 2

   Fix a colour code and a fork position

   Old strands one colour, new strands another, the fork in the same spot in every scene. Decide this before you write a word.

3. 3

   Write one paragraph per scene

   Follow the scene plan above, using your course's exact enzyme names. About 600 words is four minutes at 150 spoken words a minute.

4. 4

   Paste the script into openCanviz

   Choose Keep my wording so every enzyme name is said as written, set a target length, and pick the whiteboard style so the fork builds up as you explain it.

5. 5

   Check every 5' and 3' label

   Pause on each scene. Drafted diagrams can flip strand directions, draw Okazaki fragments on the leading strand or label ligase wrongly. Fix any scene in the editor.

If you are using this for an exam, how to make an AP exam review video covers turning a topic like this into practice. Watching is not revising: pause before each scene and say which enzyme comes next.

Common questions

Why is DNA replication called semi-conservative? Because each new DNA molecule conserves one of the two original strands and has one newly made strand. Half of the original is kept in each copy.

What is the difference between the leading and lagging strands? The leading strand is made continuously in the same direction the fork moves. The lagging strand is made in short Okazaki fragments, each built away from the fork, because DNA polymerase can only add nucleotides in the 5' to 3' direction.

Why does DNA polymerase need a primer? It can only add nucleotides to an existing 3' end; it cannot start a strand from nothing. Primase makes a short RNA primer that provides that starting 3' end.

What did Meselson and Stahl prove? That DNA replication is semi-conservative. Bacteria grown on heavy nitrogen-15 and then moved to light nitrogen-14 produced only intermediate-density DNA after one generation, and equal intermediate and light DNA after two.

What is the difference between DNA polymerase I and III? In prokaryotes, DNA polymerase III does most of the copying, adding nucleotides to both new strands. DNA polymerase I removes the RNA primers and replaces them with DNA.