PRDM9 is my favourite gene. Why? Because it is the strongest driver of speciation identified to date. Thanks to the activity of PRDM9 (and probably some other similar genes we haven’t recognised yet), we live in a world full of awesome metazoans such as hedgehogs, dragonflies, narwhals and axylotils. The gene was tricky to find and it’s function is still not completely understood. Here, I will explain the story of it’s discovery, what we think it does, and why that’s awesome.
PRDM9 was only identified quite recently by scientists trying to understand the process of genetic recombination. [Recap paragraph!]: In my last post I spoke at length about how chromosomes can swap pieces of DNA with one another during cell division. I mostly talked about ‘non-homologous recombination’, where two chromosomes swap non-matching pieces of DNA with one another, one chromosome often completely losing vital genes and it’s counterpart gaining extras. Non-homologous recombination often causes disease, so why haven’t we evolved out of it? The reason is that homologous recombination – where two chromosomes swap like-for-like stretches of DNA – is an integral part of evolution, as it allows species to ‘shuffle’ different variations of genes and see which combinations work best together. The question is, what controls recombination? How does it happen?
Recombination isn’t random. Certain places in the genome recombine frequently, while in other places it almost never happens. We know this by looking at ‘haplotype maps‘ of people from lots of different ancestral groups, which show us which combinations of small genetic changes are always inherited together in blocks (no recombination happens between them), and which have been shuffled around over time. The places where recombination happens a lot are called ‘recombination hotspots’.
What makes a recombination hotspot?
Maybe it’s the DNA sequence? Hotspot regions are about 1-2 thousand DNA bases in length. Many of them share common features like having a large % of G and C bases, or being repetitive, but there’s no one feature that unites them. 40% of them do share a common piece of sequence, 13 basepairs in length, but it also crops up in other places that aren’t hotspots.
Ok, maybe they all have something else in common…
The recombination rate of DNA is different in men and women, even across identical bits of sequence. This could be explained by differences in methylation. Methylation of DNA controls how chromosomes are folded up inside a cell. Men fold their DNA differently to women – there is more methylation at the ends of the chromosome, curling it into lots tight loops, than in the middle where the loops are fewer and larger. Women methylate their DNA at equal intervals and create even sized loops across the chromosome. Women also have a pretty even amount of recombination across the chromosome, while men have more at the ends and less in the middle. It seems like the bits on the outsides of loops are more likely to get tangled with another chromosome, and that the region on the insides of loops – which are nearest their sister chromosomes during cell division – would have the highest rates of recombination.
BUT when scientists compared the recombination rate across a 10,000 basepair region of a chromosome….between all men, all with the exact same sequence in the region, they found the recombination rate was STILL different between them!
Ok, so the sequence sometimes but not always has something to do with it, and methylation sometimes but not always has something to do with it?
“Bloody hell!” scientists exclaimed, in unison, “It must be a trans-acting factor!”
A trans-acting factor is a gene which has the function of regulating what another bit of DNA does. Two groups came across the gene simultaneously: it bound to the 13bp sequence common to 40% of hotspots, and it altered DNA methylation. The gene was PRDM9.
What is a PRDM9 when it’s at home?
PRDM9 makes a protein with three domains: a protein/protein binding domain, a domain that trimethylates histones (that means it influences how the chromosome folds by altering methylation) and a series of ‘zinc-fingers’. Each of zinc-finger can bind to a specific sequence of three base pairs of DNA. In humans the array of zinc-fingers bind to the 13bp hotspot motif. The evidence stacks up to make this the first (possibly of several) gene known to direct recombination in humans.
The plot thickens: PRDM9 is common to all species, except that it also isn’t
The PRDM9 gene was also identified in mice, chimps and other animals. Its function appears to have remained unchanged across all animal species. The zinc-finger domain, however, is wildly different between species: even in humans and chimps the gene’s sequence is completely different in this region, which may explain why humans and chimps don’t share any recombination hotspots found to date. This element of the gene had a high level of mutation even between humans: the 13bp motif appears to be gradually getting less and less common in our own genome.
The drastic change in this part of the protein is explained by the ‘hotspot paradox’. If DNA keeps recombining at the same places all the time, one or two DNA bases is bound to be lost occasionally. This means that over time, the sequence of the hotspot will change and trans-acting factors will no longer bind to it: the hotspot will die. To combat this, PRDM9 allows frequent mutation of its zinc-finger region, changing what sequence it binds to, evolving to make new hotspots as the old ones disappear. The fact that we see different sequences here emerging even within our species explains why we don’t see the same sequences in any other animals.
PRDM9 actually drives the evolution of different species
PRDM9 not only encourages recombination across the chromosome, it also specifically encourages it in the DNA around the centromere. Centromeres are usually located in the middle of the chromosome, and are the anchor used to pull different chromosomes apart during cell division. This function is shared by all centromeres in all multicellular organisms. The DNA sequences of centromeres is utterly different between even closely related animals. It seems that PRDM9 drives changes to centromeric sequence more than anywhere else. The reason for this is that different centromere sequences are what stop different species breeding with one another: if an organism doesn’t receive chromosomes with similar centromeres from its mother and father, its cells won’t divide properly and it can’t create a terrifying chimera baby. Hybrid animals, like mules (donkey/horse hybrids) are often sterile, because their chromosomes match enough to make an animal, but not enough for that animal to make viable egg or sperm cells, because they can’t be paired up correctly during meiosis. By making centromeres mutate really fast, PRDM9 speeds up the rate at which changing groups of animals become sexually incompatible and ‘speciate’ from one another down new divergent evolutionary paths.
In summary, PRDM9 is a badass gene which allows us to mix up our genes through recombination. It speeds up the divergence of animals into new, awesome and probably cute different species. Most importantly, PRDM9 prevents these awesome species breeding with one another and creating potentially terrifying new apex predators.