Mutation is usually considered to be a random, uncontrolled process: Mistakes during replication cause random changes to the DNA sequence, which may have positive, negative or neutral effects on the organism’s survivability. Positive changes are sustained, negative mutations die off, neutral mutations just float about. Mounting evidence now suggests that mutation can also happen on a far grander scale than this, that the changes are not always randomly located, and that they may not be mistakes at all.
Evolution is not content with changing a species one base at a time: mutation can actually cause huge genetic changes over just a single generation. Whether these changes are adopted by the whole species (becoming ‘fixed’) is another matter, but it seems that rather than progressing via a gentle trickle of subtle changes, species try out all these small mutations in different combinations, shuffling their genomes about like a deck of cards.
This discovery was made by Dr Barbara McClintock, after realising through her experiments with Maize that the genome is mobile; not a static structure. She found that when she self-pollinated maize plants (fertilised them with their own genetic material) for multiple generations, they acquired an increasing number of mutations, including entire genes being removed or switching positions – something entirely new at the time, which eventually won her the Nobel Peace Prize for Physiology/Medicine.
Since McClintock’s experiments we have found multiple instances of genome mobility in humans. For instance, there are six different genes that make haemoglobin subunits, which come together to make different kinds of haemoglobin molecules. These genes are almost identical to one another, and are now widely agreed to have arisen from multiple copies of one original gene. Over time, they developed their own individual functions and are all used at different stages during early life.
Detrimental occurrences of genome mobility are also common: many genetic diseases are caused by the complete deletion of large segments of DNA, sometimes removing thousands of base pairs of sequence and multiple genes. Instances of genome mobility have been found that cause chunks of DNA to delete, duplicate, invert and even translocate (move from one chromosome to somewhere completely different).
How often do these events happen? They have clearly occurred many times in our past, judging by the number of genes we have that appear to be close copies of one another. It’s impossible to know their frequency of new mutations in the general population – many events may be ‘silent’ and have no clinical consequences that would cause us to do genetic testing on the affected person. Even losing multiple genes is tolerable: one of the reasons we have two copies of each chromosome is to provide a backup copy of any genes that are missing or broken. On top of this, mutations that are really bad may cause death at an early stage of development and result in miscarriage and never be recorded. Some studies observing recombination rates in sperm cells have begun to redress this, but still do not provide a complete picture of recombination in the population as a whole, which may be different in egg cells, and also vary between ethnic groups.
While it’s still challenging, computational techniques for identifying rearrangements is slowly improving. What they have revealed is quite surprising: big rearrangements don’t strike the genome at random, rather, they always seem to start and end in regions with similar properties. These ‘hyper-mutable’ sequences crop up frequently and some are even common to multiple species. Could it be that what we have thought of as ‘mistakes’ in DNA replication are actually a built in feature of our genome?
It seems that, really, evolution through mutation occurs on two different scales – the micro scale: small changes (such as a single base being added, removed or changed within a sequence) which can gradually improve and adapt genes, and the macro scale, swapping big chunks of DNA all in one go. This is called ‘recombination’ and primarily happens during meiosis (cell division which forms egg and sperm cells with 50% of the genetic material required to produce a new life). During meiosis, the two copies of each chromosome in your cells line up next to each other, before being pulled apart into two separate cells. Pieces of DNA from the two different chromosomes can get tangled during this and be swapped into the wrong cell. When two identical pieces of sequence are swapped between chromosomes this is known as ‘homologous recombination’, which enormously accelerated the rate of evolution in humans, and every other species that enjoys sexual reproduction.
The purpose of recombination is to see whether different combinations of genes work well together. If on one chromosome an advantageous mutation has occurred in gene A, but a debilitating, disease causing mutation has occurred in gene B, without recombination the species would never reap the benefits of the improved version of gene A, because it would always be co-inherited with the damaged gene B. Certain regions of each chromosome are designed in a way that make them likely to recombine with their counterparts, and they are mostly positioned outside of genes, so if a little bit of sequence is lost in the process it doesn’t have any functional consequences.
What qualities of these regions makes them hotspots for recombination? The sequence of bases in a chromosome have another function beyond holding the blueprints for genes: it actually influences the physical structure of the chromosome. This is often the role of ‘repeat’ sequences formerly considered to be ‘junk’ DNA because they don’t code for any proteins and look like nonsense: the same small pattern of bases repeated again and again hundreds of times. When these sticky regions stick to their counterpart on another chromosome, they can cause recombination.
It all breaks down when the recombination doesn’t result in an equal exchange between two matching chromosomes. The results of non-homologous recombination are the odd rearrangements which can cause diseases: inversions, translocations, deletions and duplications.
A recent finding in the study of genome architecture is the growing association between big rearrangements and palindromic sequences. A DNA palindrome is like a language palindrome in that it reads the same forwards as it does backwards, however, it is more complicated in that it reads the same backwards, but on the opposite strand of the sequence…
These are computationally challenging to detect, especially because they don’t necessarily completely perfect. More and more newly discovered mutations, as well as those which we have known the locations of for ages, are being found to start or end in palindromic sequences. The current consensus on this is that palindromic sequences can fold back on themselves, creating ‘hairpin’ loops of DNA which can end up reversed or cut out of their normal positions.
This is just another way in which DNA is a completely mind-blowing and fascinating molecule: through properties no more sophisticated than the order in which four bases are laid out along a chromosome DNA can not only instruct a single cell on how to grow into an entire freaking multicellular organism, it can also rewrite itself and modify it’s own physical structure. Learning how DNA governs itself will improve our understanding of the genetics of diseases, as well as our comprehension of life, nature and everything.