Photograph by Joe McNally, National Geographic
Photograph by Joe McNally, National Geographic

From Ancient Genomes to Ancient Epigenomes

Late last year, scientists unveiled the complete genome of a female Neanderthal whose 130,000-year-old toe bone had been found in a cave in Siberia. As it turned out, her sequence of some 3 billion DNA letters was not all that much different from mine or yours. The researchers identified only about 35,000 places in the genome where all modern humans differ from our ancient hominid cousins. And only 3,000 of those were changes that could impact how genes are turned on and off.

But if our DNA is so similar to Neanderthals’, why were they so…different? They were brawnier than our ancestors, with short but muscular limbs, and big noses and eyebrows. They didn’t carry certain genetic variants that put modern humans at risk of autoimmune disease and celiac disease. And although they lived alongside our ancestors as the latter migrated into Europe, for some reason the Neanderthals didn’t survive.

Part of the answer undoubtedly lies in the way the Neanderthal genome actually worked — a complex process that depends not only on the underlying DNA code, but on the way genes get turned on and off. DNA molecules are constantly interacting with chemicals that control which genes can be activated. For example, a methyl group (one carbon and three hydrogen atoms) can latch on to the genome and help switch on or off the expression of nearby genes.

This dynamic layer of genome regulation, known as the ‘epigenome’, has received a ton of scientific attention in the last few decades. Researchers have claimed that epigenetics can explain (among many other things) how the placenta works, and why some people develop autism, and why enduring a famine in childhood might affect the health of one’s  future grandchildren. A commentary in last week’s issue of Science suggests that epigenetics may also hold the key to interpreting ancient genomes, including those of the Neanderthal, a 4,000-year-old Eskimo, and an 800-year-old plant.

It’s crazy, really, that scientists can glean anything from such old, old DNA. To put together the Neanderthal genome scientists had to combine many DNA fragments painstakingly extracted from bits of bones. But these DNA sequences also carried hints of the past epigenome.

DNA methylation usually happens on DNA bases called cytosines. As it turns out, cytosines decay differently depending on whether they are methylated. The cytosines that once carried methyl groups turned into a chemical called thymine, whereas those that were not methylated turned into a different chemical, called uracil. By measuring thymine, then, researchers can estimate the amount of DNA methylation in ancient samples.

In May a team of researchers did exactly that to the Neanderthal genome, comparing its thymine profile to that of present-day people. As they published in Science, the scientists found that the overall methylation map was very similar between the two species, showing that their thymine trick was indeed a good proxy of methylation. But they also found intriguing differences. Unlike modern humans, Neanderthals carried a lot of methylation on the HOXD9 and HOXD10 genes, which are both known to be involved in limb development. This might explain some of the anatomical differences between the species, the authors say. What’s more, the genes that were methylated differently in Neanderthals and modern humans are nearly twice as likely to be linked to diseases, and particularly brain disorders.

In another recent study, researchers used similar epigenetic sleuthing on a Paleo-Eskimo found in Greenland. There are certain places in the genome, called ‘clock CpG sites’, in which methylation levels correlate predictably with age. By looking at the Eskimo’s thymine profile at these sites (gleaned from a 4,000-year-old tuft of hair), the researchers discovered that the guy likely died in his 50s.

Most of the controversy swirling around modern epigenomes relates to the question of just how readily our genes respond to changes in the environment. Somewhat amazingly, that same question can be investigated with ancient epigenomes. In a study published last month, researchers estimated the methylation levels of barley samples — ranging from 500 to 2,500 years old — found in an archaeological site in southern Egypt. The samples showed steadily decreasing levels of methylation with age (which is a clear demonstration of the aforementioned cytosine decay process, not a sign of rapidly changing methylation patterns). But there was one exception: An 800-year-old sample, which had tested positive for a killer infection called the Barley Stripe Mosaic Virus, had far higher levels of DNA methylation than an uninfected sample of the same age. It’s a neat illustration of ancient epigenomes revealing ancient exposures.

I don’t want to make too much of this approach. Scientists still don’t really know how to interpret epigenetic changes in living people (whose diet, exposures and medical history can be tracked, however crudely). What epigenetic differences say about ancient species is even more mysterious. All the same, it’s pretty incredible to think of the long biological histories that scientists manage to dig out of ice and rock.