To deliver genetic medicine it is essential to understand the potential impact of DNA sequence changes. This is a core of medical genetics. We look to see where the DNA sequence in a person is different from the reference genome and if that has caused, or has the potential to cause, medical problems.
I frequently come across incomplete or incorrect understanding about this basic knowledge. This includes researchers and medical personnel working in the area. The inconsistent and opaque terms used are part of the problem. It’s impossible to guess what a ‘nonsense’ mutation is. And it can be hard to remember, if you only come across the term occasionally. That’s why I prefer the uglier, but more explicit, ‘stop-gain’ mutation (see below for more explanation).
In fact the problems start at the very beginning, because there are multiple words in regular use in healthcare to describe a difference in DNA sequence, including ‘change’, ‘mutation’, ‘variant’, ‘alteration’, ‘mistake’, ‘error’. I am guilty here. In the clinic I tend to use ‘pathogenic mutation’ if a sequence change is disease-causing and ‘variant’ otherwise. I realise this is not easily scientifically justifiable as any change in DNA sequence is a mutation. But it greatly helps in giving clarity about which changes have medical impact and in reducing the misdiagnoses I discussed in last week’s post.
How can the DNA sequence be changed
DNA is made of four building blocks denoted by the letters A, C, T, G. There are many ways in which the sequence of letters can be altered. For example:
- One letter can be changed to a different letter.
- One or more letters can be inserted or deleted.
- The order of letters can be changed.
The effects of these changes are very variable, and mostly unknown. But we do know that across the whole genome each of us has millions of differences from the reference genome, without any ill effects.
In part this is because we have two copies of the genome entwined together in the double helix. Usually the change is only on one copy. The other copy has the normal sequence and can cover up for failings of the altered copy.
DNA sequence changes in genesGenes are one part of the genome where we have good understanding of what the sequence actually codes for: three consecutive letters (called a ‘codon‘) code for an amino acid. The amino acids are put together to make proteins.
There are 64 codons but only 20 amino acids. This means different codons can code for the same amino acid and some changes in the DNA sequence do not cause an amino acid change. For example, GCC and GCG both code for the amino acid alanine. There are also three codons for STOP, to signal where the protein ends.
What is the impact of gene sequence changes?
Changing the DNA sequence in a gene can have variable effects. Swapping one letter for a different letter may have no effect if it still codes for the same amino acid (silent or synonymous mutation), or a profound effect, for example if it causes a STOP code to be put in too early (nonsense or stop-gain mutation).
If the letter swap results in a different amino acid being put in the protein it may have no effect, a minor effect, or a substantial effect. These are called nonsynonymous or missense mutations.
Inserting or deleting one or two letters in a gene sequence has substantial impact, because it alters the reading frame and the code becomes scrambled after the insertion or deletion. This often means the function of that copy of the gene is lost. The term ‘indel’ is often used as an umbrella term to cover insertions and deletions.
A comparison with words and language
There are parallels with words and language that I have found useful in illustrating these changes. In the example below the letters are equivalent to the letters of the DNA code. Each word is equivalent to an amino acid and the sentence is the gene. You can see how different mutations (in red) have different impacts.
What is the clinical impact of gene changes?Synonymous mutations that do not alter the gene code are common. We all have 1000s of them and virtually all are harmless, as one would expect.
Nonsynonymous mutations that change one amino acid for another are also common, and most do not have major clinical impact. But if a single critical amino acid is altered it can sometimes have a profound effect. For example, achondroplasia, the commonest cause of human dwarfism, is caused by changing letter 1138 in the FGFR3 gene from G to A.
Predicting the clinical impact of nonsynonymous mutations is one of the foremost challenges in genetic medicine today. We are not very good at it, for most genes, and there is a strong tendency to over-predict the clinical impact, particularly if the nonsynonymous change is rare.
Stop-gain, insertions and deletions are much less common and much more likely to have a serious consequence. These types of mutation are the cause of many genetic conditions. For example, the majority of disease-causing mutations in the BRCA1 and BRCA2 genes are of these types. The position of the stop-gain or indel is less critical than for nonsynonymous mutations because the cell will typically remove that whole copy of a gene with a stop-gain or indel mutation, through a process called nonsense-mediated RNA decay.
Not always so simple
I have described the most common ways in which the DNA sequence of genes can be altered and the most common consequences of those changes. However, unsurprisingly, there are many other ways the function of a gene can be altered, some of which do not involve changing the sequence. There are also exceptions to the consequences I have described. For example, stop-gain or indel mutations near the end of a gene can have minimal impact on gene function.
But knowing the basics will be sufficient for most people in most situations.