Why must DNA be copied?
DNA holds the instructions for building and running a living thing. It is stored in chromosomes in the nucleus.
When one cell divides into two, both new cells need the complete set of instructions. So before division, in the S phase of the cell cycle, the cell makes an exact copy of all its DNA. This copying is called DNA replication.
Replication happens before mitosis (body cells) and before meiosis (cells that will make gametes).
The structure that makes copying easy
DNA is a double helix: two long strands twisted around each other like a spiral ladder.
- Each strand has a backbone of sugar and phosphate.
- On each sugar sits one base: adenine (A), thymine (T), guanine (G) or cytosine (C).
- Bases on opposite strands join with weak hydrogen bonds: A pairs with T and G pairs with C. This is complementary base pairing.
Because of this rule, each strand tells you exactly what the other strand must be. If one strand reads ATGC, the other reads TACG. That is why one strand can serve as a pattern (template) for a new one.
Steps of DNA replication and the enzymes
- Unzipping: the enzyme helicase breaks the hydrogen bonds between base pairs and separates the two strands. The Y-shaped opening is called the replication fork.
- Starting: a short starter piece (primer) is placed so the copying enzyme has somewhere to begin.
- Building: DNA polymerase moves along each old strand and adds free nucleotides from the nucleus, each one matching its partner: A opposite T, G opposite C. It can only add in one direction, so one new strand is made smoothly and the other in short pieces.
- Sealing: the enzyme ligase joins the short pieces into one continuous strand.
An enzyme is a protein that speeds up a reaction in the cell without being used up.
Semi-conservative replication
At the end there are two DNA molecules. Each one has one old strand (kept, or 'conserved') and one new strand. Because half of each molecule is old, this is called semi-conservative replication.
Scientists Meselson and Stahl showed this in 1958 using bacteria grown with heavy and light nitrogen. After one round of copying, all the DNA had a weight exactly halfway between heavy and light: half old, half new.
Errors and safety: proofreading and mutations
DNA polymerase sometimes adds the wrong base (for example G opposite T). It proofreads: it checks each new pair, removes a wrong base and puts in the right one. Other repair enzymes check again later.
Thanks to this, only about 1 mistake stays in roughly every billion bases copied. A mistake that is not fixed becomes a permanent change, a mutation. Most mutations do nothing; some are harmful (some cancers); a few are useful and drive evolution. UV light, X-rays and tobacco smoke can damage DNA and increase mistakes, which is why sunscreen and not smoking protect your cells.
Replication in viruses
A virus is not a cell. It is genetic material (DNA or RNA) in a protein coat. It cannot copy itself on its own.
A virus enters a host cell and uses the cell's enzymes, nucleotides and energy to copy its own genes and build new virus particles. These burst out and infect more cells. Many antiviral medicines work by blocking the enzyme that copies the virus's genes. Viruses with RNA (such as flu) copy with less proofreading, so they mutate quickly; that is why the flu vaccine changes every year.
Try it: copy a DNA strand by hand
At home: write this strand on a strip of paper: A T G C C A T G. Cut it lengthwise from a copy of its partner strand (T A C G G T A C). Now on two new strips, write the matching partner for each half. Check: you now have two pairs, each with one old and one new strip.
In the 3D: at the last step, move the slider slowly and say each new letter before it appears. Then press Make a mistake.
Key formulas and definitions
- Base pairing: A — T, G — C
- Old strand ATGC → new partner TACG
- In double-stranded DNA: %A = %T and %G = %C
- After n rounds of replication: 1 DNA → 2ⁿ DNA molecules
- Key enzymes: helicase (unzip), DNA polymerase (build + proofread), ligase (join)
Worked examples
1. One DNA strand reads 5′-AGTCCA-3′. Write the new strand that DNA polymerase builds opposite it.
Pair each base: A→T, G→C, T→A, C→G, C→G, A→T. The new strand is TCAGGT (written 3′ to 5′ under the old one).
2. A DNA sample has 30% adenine. What percent is thymine, guanine and cytosine?
A pairs with T, so T = 30%. A + T = 60%, leaving 40% for G + C. G = C, so G = 20% and C = 20%.
3. One DNA molecule is copied 3 times in a row. How many molecules are there, and how many still contain an original old strand?
Copies double each time: 1 → 2 → 4 → 8 molecules. The 2 original strands are never broken up, so exactly 2 of the 8 molecules contain an old strand.
4. Polymerase copies 6 billion bases in a human cell and leaves about 1 error per billion. About how many mutations does one cell copy leave?
6 000 000 000 ÷ 1 000 000 000 = about 6 new mistakes per cell division.
Common mistakes
- Thinking both strands of a new DNA molecule are new. Each new molecule keeps one old strand (semi-conservative).
- Pairing A with G or C with T. Only A–T and G–C fit together.
- Mixing up helicase and polymerase. Helicase unzips; polymerase builds the new strand.
- Thinking replication happens during mitosis. It happens before, in the S phase of interphase.