The Mysterious World of the Human Genome

by Frank Ryan

  I asked Douka if we had any idea of the likely size of the migrating populations.

  “I really don't think we have any true idea, because we don't really have much in the way of hard data that could lead us to calculate population size. But if you look at the genetic data coming out on the Neanderthal genome, it proposes an effective population size of between 1,500 and 3,000 reproductive Neanderthal women. Based on that you could be talking about a total Neanderthal population, in the European context, of roughly 10,000 individuals at the time the fossils were collected. If you ask me, I would hazard even less than that.”

  “And contemporaneous modern humans—I presume they would also have run to small total populations?”

  “Modern humans are considered a tropical species of humans, where population size is thought to have been much greater. Plus, the idea is that as they came out of Africa and dispersed across Asia, they maintained a wide networking of groups, with associated genetic exchange. So it is not unlikely that the population of modern humans had a higher turnover and their numbers were larger.”

  In a recent paper in the journal Science, Mellars and his Cambridge-based colleague Jennifer C. French assessed relative populations of Neanderthals to modern humans using a combination of genetic and traditional archeological techniques. They compared mitochondrial DNA diversity among present-day European populations to mitochondrial DNA diversity derived from Neanderthal remains. They also analyzed various “archeological proxy evidence” for intensity of occupation over the Neanderthal to modern human transition in the well-studied southwest of France. They concluded that the numbers of modern humans settling the area showed a nine-fold increase on the numbers of Neanderthals that had previously occupied the same area. There are some obvious assumptions in such a technique, but their conclusions do pose an interesting question: What if the Neanderthal disappearance from Eurasia was brought about by simple numerical supremacy of arriving modern humans?

  As we shall discover in the succeeding chapter, this, perhaps combined with some other emerging discoveries, might answer the question that has intrigued scientists and lay public alike for more than a century. But for the moment, we shall maintain our focus on the colonization of Europe by arriving modern humans.

  I pressed Dr. Douka, “So it seems that you get modern humans arriving into Europe—or Eurasia—and from that point on there appears to be a fairly rapid cultural evolution, perhaps starting around 20,000 years ago?”

  “I wouldn't have put it at 20,000. For me it's probably earlier, more likely about 40,000 to 45,000 years ago. Some of these early modern humans would have occupied small pockets of Europe, southern Italy, and western France, but after that, between 33,000 and 30,000 years ago, we really see something completely different. This marks the arrival of the Gravettians.”

  “Another migration out of Africa?”

  “We don't know where they come from or whether their culture first develops within Europe and then expands all the way to Russia, or vice versa. These humans appear around 33,000 years ago. And there is a completely new way of doing things. They bury their dead—and they bury them with thousands of beads. Some of these are shell beads but others are small ivory beads. They also use the canines of red deer, which they make into beads. They produced wonderful sculptural figurines. So we're talking about major cultural change.”

  “We don't know whether it's an idea spreading, or whether it's a new people bringing in the new ideas?”

  “That's right.”

  “Perhaps in time the genetics will answer this?”

  “Perhaps, yes. At present we have very little in the way of DNA from these people, but this is about to change. What we know, however, is that from 33,000 onward there was a new blooming of human culture.”

  These cultural and population changes took place at a time of monumental climate and ecological change in Eurasia. These severe climatic variations were accompanied by periods of major population movement and growth in the areas affected by the Ice Age. A prolonged cold period, known in the scientific jargon as the Last Glacial Maximum, or “LGM,” occurred between 26,000 and 19,000 years ago. This cooling period is thought to have caused a massive population decline in Europe, with the survivors taking refuge in “climate sanctuaries,” or “refugia,” the four major examples of which were northern Iberia and southwest France, the Balkans, the Ukraine, and the northern coast of the Black Sea and Italy.

  The Last Glacial Maximum may well have reduced the genetic diversity of Europe, which in turn would complicate any assessment of arrival and dispersal based on the haplogroups of current Eurasian inhabitants. As the glaciers began to pull back, about 16,000 to 13,000 years ago, people began to repopulate the devastated landscape from the four geographic refuges, so that anthropological geneticists can expect to see not only genetic signatures that date from before the LGM but also the genetic signatures arising from the isolated populations of the four refuges, which would have expanded to fill the landscape as the glaciers melted.

  For example, 80–90 percent of males in Ireland, Wales, Scotland, and the Basques in northern Spain and western France share the male-specific-region Y-chromosome haplogroup, or “MSRY,” known as “R1b,” as do 40–60 percent of the male population of England, France, Germany, and most of the rest of Western Europe. This makes it likely that their patrilineal ancestors took refuge in the northern Iberian climate sanctuary. In southeastern Europe, R1b drops behind a related haplotype, R1a, in the area in and around Hungary and Serbia. Another MSRY haplogroup, labeled “I,” is found in its highest frequencies in Bosnia and Herzegovina, Serbia, Croatia, as well as Nordic countries, such as Sweden and Norway, and parts of Germany, Romania, and Moldova. The same haplogroup clade is highly European-based, making some geneticists think it could date to before the last Ice Age. These, and many other different haplogroup clades, do help to trace populations and their movements, but I should add that, as we might expect from historic sources, they also show huge population mixing.

  Curiously, when we look at the matrilineal-based mitochondrial haplogroups in Europe, we see what appears to be a very different pattern from what we saw in the male-associated Y chromosome. When compared to the males, the female-associated haplogroups show much less geographic patterning, which seems to indicate that European women share a more common ancestry. How fascinating if this might reflect different socio-cultural traditions affecting the mobility of men and women.

  Some 99 percent of all European mitochondrial haplotypes fall within the categories H, I, J, K, M, T, U, V, and W or X. Haplotype H is the most common, being found in no less than 50 percent of all Europeans, and six of the above haplogroups, H, I, J, K, T, and W are only found in European populations. The latter suggests that these haplogroups arose in the ancestral Caucasoid populations after they had separated from the ancestors of modern Africans and Asians. As historical records show, these ancestral genomes will have been blended with many different gene flows from eastern Asia, southern Siberia, and Africa, but the ancestral patterns are still readily detectable even in modern genomes. Geneticists are currently assessing whether Europeans are mostly descended from Paleolithic or Neolithic ancestors by looking at more and more genomes that date from 15,000 years ago or older.

  These early ancestors were very similar to us, but they weren't the same. The use of genetic data to discern aspects of human prehistory is known as “archeogenetics.” One of the intriguing insights that is coming out of such archeogenetic studies in recent years is the fact that significant evolutionary change has taken place in our human genome in the last 50,000 years—a key time in the human migrations out of Africa and the colonization by modern humans of Europe, Asia, Australasia, and the Americas. Paleoanthropologists have raised the possibility that this has been driven by the evolution of culture.

  Perhaps we shouldn't be too surprised at this. Culture is a quintessential facet of human life and experience. Once again, the geneticists have looked to mutational change,
and in particular to the acquisition of new Snips that seem to mark out specific cultural populations. The theory is that some of these Snips in the autosomal chromosomes are conserved because they happen to be close to a particular variant of a gene (in the jargon, a specific “allele”) that is already present, even in just a small number of individuals, perhaps dating back to a single ancestor, that has been favored by natural selection because it gives a survival advantage.

  Robert Moyzis and his colleagues at the University of California at Irvine searched for such evolutionary change among some 1.6 million Snips scattered throughout the entire human genome. They concluded that roughly 1,800 genes had been influenced in this way over the last 10,000 to 40,000 years—or, to put it another way, some 7 percent of the genome had been specifically targeted by evolutionary forces during the expansion and settlement of humans in this phase of human history. The populations studied included Americans of European origins, Americans of African origins, and Americans of Asian (Han Chinese) origins. Key areas of the genome that appeared to be under intense selective pressure included some of the most important aspects of human internal chemistry. These included our ability to fight off infectious diseases, sexual reproduction, DNA chemistry and copying as part of the cell cycle, protein metabolism, and the function of the nerve cells that form our brain and central nervous system.

  The authors also concluded that we have undergone evolutionary change affecting many different physical attributes, very likely extending beyond those that they tested for. This study suggested that human evolution can involve widespread physiological and physical change in what, from an evolutionary perspective, would be a relatively short space of time. How likely is it that the same rapidly moving evolutionary selection pressures are still operating on us today?

  As our journey progresses, we become increasingly amazed at how the exploration of our human genome reveals so much that was formerly mysterious in our personal and cultural history. The potential for such exploration has undergone a sea change in recent years, presenting us with possibilities that would have seemed impossible a generation ago. As we shall now discover, it has become possible to learn from the study of genomes that diverged from our human ancestral line more than half a million years ago.

  It's the questions and not the answers that are interesting, because these are questions that have no answers. But they are interesting questions to think about because they somehow reflect how we think about differences between us and our ancient ancestors.

  SVANTE PÄÄBO

  Svante Pääbo is a Swedish evolutionary geneticist who works at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany. The son of biochemist Karl Sune Detlof Bergström, who shared the Nobel Prize in Physiology or Medicine in 1982 for discoveries related to the prostaglandins, Pääbo founded the new investigative scientific discipline known as paleogenetics. Until recently, geneticists had assumed that reading the genomes of extinct species of animals and plants, as popularized by Michael Crichton in his novel Jurassic Park, would be practically impossible because DNA degrades over time. The older the fossilized bones, the more degraded the DNA. Thus ancient fossils, dating back tens or even hundreds of thousands of years, were assumed to contain little or no residual DNA and were thus thought to be an unlikely source of useful genetic information. But over the decades since the 1980s, Pääbo and his colleagues at the Max Planck Institute began to make inroads into what had hitherto seemed impossible.

  Through their work, we have discovered that ancient DNA, though degraded, will sometimes survive the ravages of time. In pioneering this revolutionary new science of paleogenetics, Pääbo has perfected methods of amplifying and then extracting genetic information from fossil bones and other ancient remains, making it possible to explore the genomes of long-dead animals and plants and, most intriguing of all, of ancient species of humans. In an interview with the online website, Edge, Pääbo admitted that he started out naïvely thinking it would be easy to study the genomes of long-dead individuals. To begin with his interest was the more recent past, and in particular the mummified bodies of ancient Egyptians, assuming these would be more amenable than the vastly older extinct hominins. He would subsequently confess: “I was driven by delusions of grandeur.” But he persisted in his dream, and, though it proved to be far from easy, he eventually succeeded.

  Perhaps the simplest way to explain how he did so would be to climb back on board our train and pay a brief visit to the genomic landscape of such ancient fossils. Bone and tooth have turned out to be two of the best sources of ancient DNA in humans, so this will determine our choice of destination—the fossilized bone or tooth from a long-dead human. Here we enter a landscape very different from anything we visited before. We arrive at no neat railway line running away into the distance, east and west, but a confusion of fragments that, from a distance, resemble an explosion in a spaghetti factory. Pulling in closer, it still takes us a little while to recognize that what we are looking at is myriad haphazard shards of decayed genome. We are dismayed at the sight of these broken sections and pieces, which appear to be meaningless in terms of reading the original code in the pattern of sleepers. It seems to confirm what geneticists had long thought about the state of affairs in such ancient fossils…

  But all is not quite as it seems. Each individual fragment of spaghetti contains a small but genuine fragment of DNA sequence—in our analogy, a broken piece of railway track. And now, as we look at the vast scatter of fragments more closely, we see that each fragment contains anything from a few sleepers to perhaps a few hundred. Hardly reassuring, one might still be inclined to think, especially when the assembled human genome comprises 6.4 billion sleepers. Indeed, if all there were in this fossil bone was the fragmented remains of a single copy of the genome, the task of reading that genome would be hopeless. The key to understanding is that these myriad fragments are not bits and pieces of a single copy of the genome, these are the fragmented remains of vast numbers of the same genome, the residue of the billions of individual cells that made up the bone during life.

  All of those different copies of the genome will have broken up into different fragments. And just as in the original Human Genome Project, when the genome was deliberately fragmented to reduce it to more manageable genetic bites, this huge variety of fragments, with breaks in different places, will contain DNA sequences that overlap. The question now is just how much of an overlap do we need to be sure that the overlap hasn't just resulted from chance? In fact, we can easily work out the mathematics with a simple calculation. What's the likelihood of, say, three or four or six or eight nucleotides following an identical sequence in a row? Since there are just four nucleotides, the chance of the nucleotide matching gives us the random possibility of one in four—G, A, C, or T. For the first two in sequence to be the same gives us a random possibility of 4 x 4—a one in 16 chance. With each subsequent nucleotide, we multiply by an additional 4. By the time we have calculated the mathematics for eight sleepers following identical sequences, the chance of this happening by accident is one in 65,536—in other words, extremely unlikely. We now have a system that works.

  The first step is obvious. We need to sequence every fragment and bank this information in a computer with a very large memory. The second step is to search the banked sequences for matching sections that will pick out areas of overlap between the different fragments. From there we can begin the laborious process of stitching sequences together using the overlaps to identify contiguous areas. In fact, we are reproducing pretty much the way the first draft of the human genomic sequence was calibrated, except we are dealing with a vast collection of much smaller sequences that must now be painstakingly knit together.

  To this breakthrough, Pääbo enlisted two others. The first of these was Kary Mullis's Nobel Prize-winning discovery of the polymerase chain reaction. This ensured that even if the individual fragments were present in very few copies, too tiny by far to be detectable by the automated
sequencing machines, they would be duplicated over and over until they became identifiable. Next came the brainwave of teaming up with an innovative American biotechnology company, 454 Life Sciences, which had been set up by Jonathan Rothberg to develop machines that were capable of automated high-throughput DNA sequencing. The company had been taken over by the pharmaceutical giant Roche in 1997. The entire process could now be automated. The machines could automate the extraction of ancient DNA, amplify it using PCR, and thus generate a soup of small bits and pieces of the entire genome, from which they could bank and then knit it all together by aligning the overlaps.

  Of course, there were other problems to be overcome, such as contamination of the test samples by unwanted sources—human, animal, and bacterial. They found ways of removing contaminating DNA with enzymes. They also avoided contaminating specimens with their own DNA by working in a designated “clean room” and by taking strict biosafety precautions. During the actual sequencing they discovered that one of the DNA nucleotides, cytosine, sometimes degraded to uracil, the nucleotide that replaces thymine in RNA. Then it became even more complicated; sometimes cytosine that had been epigenetically tagged with a methyl group degraded not to uracil but to thymine. This created confusion until they realized that, by accident, they had made a significant discovery. In 2009, one of the team, Adrian Briggs, figured out a method for distinguishing thymines derived from methylated cytosines from original thymines. Now they could read off some of the original epigenetic programming of the genome. So it happened that, little by little, the scientists advanced and perfected their DNA extraction techniques and improved the quality of the readouts.