DNA SEQUENCES AND CANCER
Human DNA contains a large number of short stretches of repeated sequence – so much so that these makeup about half of our genome. These sequences are important in cancer because some are the cause of another type of genetic instability.
There are two main categories of repeat sequences: simple-sequence DNA (or satellite DNA) and interspersed repeats. In satellite DNA the length of the repeat unit is from one to 500 base pairs. Because these repeated frequencies of nucleotides (A, T, C and G) they have a slightly different density to bulk DNA so that, when they are separated on a density gradient, they show up as ‘satellite’ bands. There are two types: microsatellites (repeated sequences of up to 6 base pairs) and minisatellites (between 14 and 100 bases long). Microsatellites may have arisen during DNA replication if the daughter strand ‘slips’ on the template strand so that the same sequence is copied twice.
Microsatellite instability (MIN) is the second form of genetic instability. It is caused by mutations in genes that carry out a specific type of DNA repair – necessary when the machinery that replicates DNA makes a mistake and an incorrect base is incorporated in newly synthesised DNA. That machinery is pretty efficient – it makes such a mistake only about once every 100,000 joins together. When it does so, however, there is a brief window in which the proteins of the mismatch repair system can detect an error (before the newly synthesised DNA strand becomes methylated), cut out the offending base and replace it with a correctly matched one. This gives an overall error rate of one in 109 bases. Defects in DNA repair proteins produce a ‘mutator phenotype’ that shows up particularly as errors in microsatellites (these normally have a very low mutation rate). The question of whether a mutator phenotype is essential for cancer cells to evolve or whether the normal mutation rate will suffice is controversial but it is one into which, fortunately, we need not be drawn here.
In terms of DNA sequence, human beings are ~99.9% identical but the 0.1% shortfall means that about one in every 1,000 bases differs among individuals. Many of these ‘single nucleotide polymorphisms’ (SNPs pronounced ‘snips’) effects but, overall, their effects are what distinguishes us from one another and some subtly modulate our susceptibility not only to also to many other diseases (Fig. 1).
1. Single nucleotide polymorphism (SNP).
These minor variants may explain why some individuals with healthy lifestyles (non-smoking, good diet, low alcohol intake, etc.) develop cancer, while others with unhealthy lifestyles do not.
Because SNPs are randomly distributed most do not occur in coding regions, so that the effects they exert are subtle, for example, slightly altering the rate at which an mRNA is made. It seems likely that the cumulative effect of many such minor variants accounts for the missing cause of hereditary breast cancer, mentioned earlier, and indeed, several SNPs associated with a two-fold increase in breast cancer risk have been identified (these affect the genes TGFB1, FGFR2, TOX3, MAP3K1 and LSP1). The study of monozygous (genetically identical) twins is a powerful way of determining the overall genetic contribution to disease susceptibility. Monozygotic twins show a strong correlation in their risk of developing breast cancer, indicating that a significant proportion, perhaps as high as 50%, arise in a genetically susceptible minority of women.