D1.3 Mutation and gene editing

A gene mutation is a change to the base sequence of a gene. It is a structural change at the molecular level: the order of nucleotides in the DNA is altered. Because DNA replication copies whatever sequence is present, a mutation is passed on to every cell produced from the mutated cell. (Changes to whole chromosomes, such as an extra copy of chromosome 21, are a different kind of mutation and are not part of this topic.)

There are three types of gene mutation:

The effect of a mutation on the protein depends on how it changes the codons, the triplets of bases read during translation (see D1.2 Protein synthesis). The diagram below applies each type of mutation to the same short stretch of the β-globin gene, one of the genes for haemoglobin.

Base substitutions are the most common type. They are not all equally likely: swaps between the two purines (A ↔ G) or between the two pyrimidines (C ↔ T), called transitions, are more frequent than swaps between a purine and a pyrimidine (transversions).

Single-nucleotide polymorphisms

When a base substitution in a germ-line cell is inherited and spreads through a population, individuals end up with different bases at that position. A position like this is a single-nucleotide polymorphism (SNP, pronounced “snip”). SNPs are the most common type of genetic difference between people: two unrelated humans differ at roughly one base in every thousand, millions of positions in total. Most SNPs are in non-coding DNA and have no effect.

Why some substitutions change nothing

The genetic code is degenerate: there are 64 codons but only 20 amino acids, so most amino acids are coded for by two, four or even six codons. Codons for the same amino acid usually differ only in the third base. So a substitution in a coding sequence may or may not change an amino acid:

• Silent (same-sense) mutation: the new codon codes for the same amino acid, e.g. GAG → GAA (both glutamic acid). The polypeptide is unchanged.
• Missense mutation: the new codon codes for a different amino acid, so one amino acid in the polypeptide is different. The effect can be minor if the new amino acid has similar properties or is in a part of the protein that is not critical, or severe.
• Nonsense mutation: the codon becomes a stop codon (UAA, UAG or UGA). Translation ends early and a shortened polypeptide, which usually does not function, is produced.

Sickle-cell disease

Sickle-cell disease is caused by a missense mutation in the β-globin gene. In codon 6 the base A is replaced by T on the coding strand: GAG becomes GTG, so the mRNA codon is GUG instead of GAG and valine replaces glutamic acid. The resulting haemoglobin S is less soluble at low oxygen concentrations and forms long fibres that distort red blood cells into a sickle shape. A different substitution in the same codon, GAG → AAG (glutamic acid → lysine), gives haemoglobin C.

The consequences of an insertion or deletion depend mainly on whether the number of nucleotides added or removed is a multiple of 3.

Not a multiple of 3: frameshift

mRNA is read in non-overlapping groups of three bases from the start codon, and this grouping is called the reading frame. Inserting or deleting 1, 2, 4, 5 or any other number of bases that is not a multiple of 3 shifts the reading frame. Every codon from the mutation onwards, in the direction of translation, is read in a different set of three. This is a frameshift mutation:

• the amino acid sequence after the mutation is almost completely different;
• a stop codon is often brought into frame, so translation ends early and a truncated polypeptide is produced;
• the polypeptide almost certainly cannot fold into its normal shape or function.

The earlier in a gene the frameshift happens, the more of the polypeptide is affected.

A multiple of 3: in-frame

Inserting or deleting 3, 6, 9 … bases adds or removes whole codons, so the reading frame is kept. The polypeptide gains or loses one or more amino acids at that point, and the rest is translated normally. Even so, the change in structure can stop the protein working, especially if it is in an important region such as an active site.

Large insertions and deletions

Some mutations add or remove hundreds or thousands of bases, for example deleting a whole exon. These almost always stop the polypeptide from functioning, even when the reading frame happens to be kept.

Use the Delete 1 base, Delete 3 bases, Insert 1 base and Insert 3 bases buttons in the mutation explorer above to compare the two cases.

Errors in replication and repair

Mutations happen even without any outside cause. During DNA replication, DNA polymerase occasionally adds a base that is not complementary to the template. Proofreading by DNA polymerase and later DNA repair enzymes correct almost all of these errors, but a very small fraction remain, and repair itself sometimes makes mistakes. At the next replication the wrong base acts as a template, so the change becomes permanent in one of the daughter molecules. This gives a low background mutation rate: each human is born with roughly 60–70 new mutations that neither parent had.

Mutagens

A mutagen is an agent that increases the mutation rate above the background rate. Most mutagens are also carcinogens, because mutations can cause cancer.

Chemical mutagens react with DNA and change the bases so that they pair incorrectly when copied, or distort the double helix:

• Benzo[a]pyrene and nitrosamines in tobacco smoke. Benzo[a]pyrene is converted in the body into a compound that bonds to guanine.
• Mustard gas, used as a chemical weapon, which adds groups to bases and can cross-link the two strands.

Mutagenic radiation carries enough energy to damage DNA:

• Ultraviolet (UV) radiation (UVB in sunlight; UVC from germicidal lamps, as the ozone layer stops it reaching the ground) is absorbed by DNA bases. It can join two adjacent pyrimidines, typically two thymines, forming a thymine dimer that causes errors when the DNA is copied or repaired. UV in sunlight is the main cause of skin cancer.
• Ionizing radiation, including X-rays, gamma rays, and alpha and beta particles from radioactive isotopes, breaks one or both DNA strands and damages bases. Faulty repair of the breaks can delete or insert bases.

Mutations can happen at any base in a genome: in genes and between them, in essential genes and unimportant ones. Which bases mutate in a particular cell is a matter of chance, so two cells exposed to the same mutagen end up with different mutations.

That does not mean every base is equally likely to change. Mutations are chemical changes, and some happen more readily than others:

• Transitions (A ↔ G, C ↔ T) are more common than transversions, because the bases involved have similar structures.
• Some positions are hotspots. In mammals, a cytosine followed by a guanine (a CG site) is often methylated, and methylated cytosine can lose an amino group and become thymine. As a result, C → T changes at CG sites happen at least ten times more often than most other substitutions.
• The neighbouring bases and the position of a sequence in the genome also affect how often it mutates.

The key point is that no natural mechanism is known that makes a deliberate change to a particular base in order to change a trait. A cell cannot “decide” to mutate the gene it needs. If bacteria are exposed to an antibiotic, resistant mutants were present before the exposure and the antibiotic selects them; the antibiotic did not cause the useful mutation. Experiments in the 1940s–50s, such as replica plating, showed this directly. Mutation provides random variation; natural selection is what makes evolution non-random.

The consequences of a mutation depend on the type of cell in which it occurs.

Somatic cells

Somatic cells are body cells that do not produce gametes, such as skin, lung or liver cells. A mutation in a somatic cell is passed by mitosis only to that cell’s descendants, forming a clone of mutant cells. It is not inherited by the organism’s offspring. Usually the effect is small, because one cell among billions is affected and it can be replaced.

The most serious exception is cancer. Mutations in genes that control the cell cycle, such as proto-oncogenes, which stimulate division, and tumour suppressor genes such as the gene for p53, which halt it, can lead to uncontrolled cell division and a tumour. Usually several mutations in the same cell line are needed, which is one reason why cancer is more common in older people and in people with long exposure to mutagens, such as smokers.

Germ cells

Germ cells are gametes and the cells that give rise to them, in the testes and ovaries. A mutation in a germ-line cell can be passed on in a gamete. The zygote formed at fertilization then carries the mutation, and because all the cells of the offspring come from the zygote by mitosis, every cell of the offspring has it, including its own germ cells. Germ-line mutations are therefore inherited and can spread through a population over many generations. This is how new alleles, including those that cause genetic diseases, enter the gene pool.

Every new allele ultimately arises by mutation. All the alleles of every gene can be traced back to changes in the base sequence. Meiosis and fertilization produce new combinations of alleles, but they cannot make a new allele. Without mutation, all the alleles that exist would eventually be lost or fixed, and variation would decline.

For an individual organism, most mutations are either:

• neutral: they are in non-coding DNA, they are silent, or the amino acid change does not affect function; or
• harmful: they reduce or destroy the function of a protein. The most harmful are lethal.

A few mutations are beneficial in a particular environment. The sickle-cell allele is harmful in people with two copies, but people with one copy are partly protected against malaria, so in regions where malaria is common the allele has risen in frequency. Over many generations, natural selection increases the frequency of beneficial alleles. Mutation is therefore essential for evolution in the long term, especially when the environment changes (see D4.1 Natural selection).

Nature of science: commercial genetic tests

Companies now sell genetic tests directly to the public. A sample of saliva is analysed for hundreds of thousands of SNPs, giving information about ancestry and about variants linked to disease, such as BRCA1 and BRCA2 alleles (breast and ovarian cancer) or the APOE ε4 allele (Alzheimer’s disease).

• Most results give a change in risk, not a diagnosis. Many people with a “high-risk” variant never develop the disease, and a “low-risk” result does not mean a person is safe.
• Without expert interpretation, such as genetic counselling, results can cause unnecessary anxiety or false reassurance, and people may make poor decisions about their health.
• There are also questions of privacy: who owns the data, and could it be used by insurers or employers?