The textbook explanation of DNA goes something like this: enzymes in our cells read a stretch of DNA and convert its code into a single-stranded RNA molecule, which is then used by ribosomes as a template for building a protein. That stretch of DNA biologists call a gene. The protein it encodes drifts off to do some job–building cell membranes, maybe, or switching off other genes, and so on.
This is a fairly accurate picture–for less than two percent of the human genome. The rest of our DNA does not encode proteins. Much of it may be made up of genetic material from viruses and disabled genes. That does not mean that this genomic dark matter is all useless to us. Scientists have known for decades that some small chunks of DNA act like switches for nearby genes. Proteins grab onto these switches to prevent our cells from making proteins from the genes, or to speed up the process.
Far more mysterious, however, are segments of DNA that our cells use to build RNA, without then turning that RNA into proteins. These RNA molecules are a shadowy network of molecules that quietly control much of the activity in a cell. They can interfere with other RNA molecules, blocking the construction of proteins. In some cases, they can act like sensors, able to grab onto particular kinds of molecules. When they sense a molecule, they may then stop a gene from being copied.
Scientists don’t know a lot about this shadow network, but new experiments on it are coming out every week. And now it turns out to have been very important in our own evolutionary history. In Nature today, scientists report that the fastest-evolving part of the human genome is an RNA gene…
Researchers have been gauging the effects of natural selection on the human genome for several years now. They’ve taken advantage of the growing supply of data on DNA from humans and other animals. By comparing those sequences, scientists can pinpoint parts of genes that have changed a lot, or changed a little. And with various statistical methods, they can figure out whether natural selection drove the changes they observe.
A lot of the studies until now have focused on good old protein-coding genes. They’re particularly well-suited to these sorts of studies. Some changes to these genes can cause significant changes to an animal, because they can change the structure of a protein. The rest of the genome is much harder to test. A change to a non-coding piece of DNA may be favored by natural selection, or it may just be neutral, spreading thanks to little more than chance.
In the new paper, an international team of scientists dove into that dark matter. They found 35,000 pieces of non-coding DNA that were very similar in chimpanzees, rats, and mice. Since these mammals are separated by 100 million years of evolution, their similarity suggests these segments may be playing an important function that has been conserved by natural selection. If they didn’t have a function, mutations would have built up in each lineage. They then looked at these segments (or rather, they had a computer look at them) in the human genome. They picked out segments in which the human versions had acquired a significant number of new mutations not found in other mammals. These mutations must have evolved since our ancestors split from chimpanzees.
The scientists found 49 candidate segments. These segments have evolved a lot in our lineage. The most drastically altered of all is a segment the scientists dubbed HAR1 (for human accelerated region). It is 118 base pairs long. Chimpanzees and chickens, separated by over 300 million years, carry versions of HAR1 that are identical except for two base pairs. In humans, on the other hand, 18 base pairs have changed since we split from chimps.
What’s HAR1 for? This is the sort of question that seems like it should be easy to answer unless you’re the scientist doing the answering. The scientists found that human cells make RNA molecules out of the HAR1 segment. Specifically, they found that brain cells do. Specifically, brain cells in the cortex, the hippocampus, and certain other regions. We do love our brains, and so it is reasonable to consider that HAR1 took on some new role in the brains of human ancestors. The sequence of HAR1 suggests that an RNA molecule produced from it would be stable enough to carry out some important job, such as regulating the activity of protein-coding genes. HAR1 probably plays several roles. It is not just active in the adult brain, but in development-guiding cells in the fetus.
In a commentary that also appears in Nature, two Oxford scientists point out that HAR1 is also active in the ovary and testis of adult humans. And it is true that genes associated with sex are fast-evolving. So they don’t want to rule out the possibility that selection has acted on HAR1 in connection with reproduction, rather than with thought. It’s a fair point, but I was struck by the fact that the expression of HAR1 is far smaller in the sex cells than in the brain.
Still, it’s a strange point that may be worth raising at your next party: we have genes that are only active in our brains and sex cells. Insert punchline here.