Tracing human evolution 29/9/06. By Giles Newton History meets the human genome.

The story of how humans evolved, migrated and adapted to new environments is told largely by discoveries of fossils, most importantly in Africa. But the recent history of Homo sapiens is also recorded in the genome of every modern-day human. At the Wellcome Trust Sanger Institute, Dr Chris Tyler-Smith is searching the genome to find the well-hidden clues to the story.

Two hundred thousand years ago, give or take a few millennia, a new ape appeared in Africa. Walking on two legs, with a high domed forehead and large brain, Homo sapiens had arrived. Yet in the eyes of other hominids of the era – Neanderthal man, for example, and possibly the remarkably successful Homo erectus, which had spread throughout the Old World – this new species was unlikely to have looked anything special. Just another human variant.

And indeed, for 120 000 years or so, Homo sapiens was neither particularly numerous nor successful. Then, about 80 000 years ago, the species acquired an edge: modern behaviour in the form of advanced tool use, and probably social interaction and language. With its modern anatomy and behaviour, Homo sapiens spread across the globe and supplanted the other hominids around at the time.

"We are those humans who expanded out of Africa about 50 000 years ago – that's the picture we get from fossils and archaeology," says Chris Tyler-Smith, evolutionary biologist at the Wellcome Trust Sanger Institute. "We came from a relatively small initial population, so we are pretty much the same genetically all over the world. That's why there is just one species of human, why we have such low genetic diversity – we are less diverse than chimpanzees, gorillas and orang-utans, despite our greater numbers and greater geographical distribution – and why there are so few differences between people in different parts of the world."

The search for genetic variants in the human genome has been revolutionised by large-scale DNA sequencing. In the last few years, not only have the 3 billion bases in the genome been read, but millions of variants have also been catalogued through work such as the International HapMap Project (which characterised variants in 270 people – Europeans from Utah, Han Chinese, Japanese and Yoruba people in Nigeria). The vast majority of these variants are termed 'neutral' by evolutionary biologists: if you have an A in your DNA at some position in the genome, while your neighbour has a C, the chances are that it makes no difference whatsoever. Natural selection does not act on such variants, so the pattern of variation that accumulates, and the fate of any new variant that arises, depends on processes such as mutation rate, the migration of people and, most importantly, 'genetic drift' (the chance change in frequency of a variant from one generation to the next).

"In a large population, on average there will be little change; but in small populations, and human populations do seem to have been small for much of our evolutionary past, drift is likely to be significant," says Dr Tyler-Smith. "By studying these neutral variants, we can find out about migration, past changes in population size, and the extent to which populations were all one or were subdivided."

Chinese whispers

Neutral variants on the Y chromosome hold particular interest for geneticists, as this chromosome is only passed on through the male line (see box, below). Several years ago, Dr Tyler-Smith's lab found a Y variation that was specific for China; following it up, they began a fruitful collaboration with Chinese scientists including Huanming Yang, chief scientist of China's Human Genome Project, which has led to fascinating insights into the genetic history of East Asia.

"When you look at Y-chromosome variation at high resolution, you expect every Y to be different," says Dr Tyler-Smith. "But when we looked at large numbers of DNA samples from East Asia, we found large clusters of Ys that were the same. We can tell from the genetic evidence roughly where and when these lineages arose, and then look to see if history explains why they are so widespread."

The first Y-chromosome cluster arose in or around Mongolia about 1000 years ago. There is little in Mongolian history of that time to explain the cluster, but 200 years later was the era of Genghis Khan, creator of the largest land empire ever seen. He established a social system, argues Dr Tyler-Smith, that perpetuated a reproductive advantage linked to his Y chromosome: his sons and male-line descendants continued to be rulers and leaders, such as emperors of China, and highly privileged for centuries after.

Not that this means that everyone who carries this Y chromosome is a descendant of Genghis, points out Dr Tyler-Smith. "According to our estimate, this Y arose 1000 years ago, whereas he lived 800 years ago. So there would be other people around carrying it. But his military power and social system could have amplified the numbers of people with this Y over the succeeding centuries; we estimate that about 16 million people are carrying it now."

Another Y-chromosome cluster in the East Asia data was more recent – about half as old as the Genghis Y. It was also more limited geographically, being confined to north-east China and Mongolia, and was present in the Chinese minority populations but not in the majority Han population. Again, the history and genetics seem to tie up neatly: the descendants of a Manchu leader called Giocangga were very privileged during the Qing dynasty, which ruled China from about 1644 to 1912. There were 80 000 officially documented members of the nobility by the end of the Qing dynasty, and an estimated 1.6m descendants are alive today.

News: News: Manchu Y chromosomes

Taking a rather different tack, a separate study is looking at the links between the genetics, linguistics and geography of the Himalayas. "There are a lot of parallels between language and genetics," says Dr Tyler-Smith. "Language tends to be passed on from one generation to another, so it often correlates with genetics, but we know that people can learn a new language and countries can change language that they speak."

The project, a collaboration between several European labs and people from Nepal and Bhutan, is the brainchild of George van Driem from Leiden University in The Netherlands, who has been studying the complex set of languages in the Himalayan region for several decades. "To simplify, you have Tibeto-Burman languages to the north and east, and Indo-European-type languages to the south and west," says Dr Tyler-Smith. "They meet in the Himalayas, and we know from other work that the Himalayas look like a genetic boundary between India and China. So we're looking in a fine scale at these boundaries and at the interplay between language and genetics."

Form and function

Neutral genetic variants provide many interesting stories about the history of human populations, but it is the functional variants - changes in our DNA that in some way affect human biology - that get biologists really excited. Such variants can give an insight into key events in the evolution of early humans (such as a variant in the ASPM gene that may have allowed brains to increase both in size and complexity) or in the evolution of modern behaviour (such as a variant in the FOXP2 that has been linked to the evolution of language).^{1 ,2}

More common traits also raise fascinating questions about the history of Homo sapiens. We vary in our hair and skin colour, our ability to drink fresh milk as adults and tolerance for alcohol, and our susceptibility to different drugs and infectious diseases. Have such traits evolved recently, or do they hark back to Homo sapiens migrating out of Africa, adapting to environments with different temperatures and new animals and plants? Or to the birth of farming and domestication, 7000–10 000 years ago, which would have brought new foodstuffs but increased opportunities for disease?

Unfortunately, functional variants are both rare and difficult to find; ASPM and FOXP2 were uncovered through studies of families with specific genetic abnormalities. "In the last few years, it has been possible to develop a different kind of approach," says Dr Tyler-Smith. "You scan the entire genome, look for signs of evolutionary selection, and then ask which genes show the signatures and what they are doing. This procedure is also very difficult – the signatures of selection are not easy to distinguish – and relies on clever statistics and huge amounts of data, but this is where much of the current excitement currently lies."

One idea being explored by Dr Tyler-Smith's team is the 'less is more' theory of Maynard Olsen – that the loss of genes may be particularly important in human evolution. In general, damage to genes is a bad thing, its downside most often associated with genetic diseases such as cystic fibrosis. But, in certain circumstances, if it were ever advantageous to lose the function of a particular gene, the mutation could be selected and spread in the population.

"We were fortunate in that the first example we chose for a pilot study was the caspase-12 gene," says Dr Tyler-Smith. "Most mammals, including chimps, have an active copy of this gene, but when it was first characterised in humans, it was thought to be defunct. A more careful look showed that it was indeed inactive in most humans, but some people retain the active form of the gene. By resequencing the gene, we found a strong signal of positive selection for the inactive form."

Caspase-12 is involved in the immune response and, intriguingly, people with inactive genes are more likely to avoid severe sepsis (the growth of bacteria in the bloodstream), and more likely to survive if they get it. When population densities increased, and infectious diseases became more common and more likely to be transmitted between people, losing the gene and becoming more resistant to sepsis could therefore have been a decided advantage – one that outweighed the original function of caspase-12.

Such stories about genes and human history are likely to become ever more frequent, given the terabytes of genetic data that are being generated by DNA sequencers. "The field is moving very quickly," says Dr Tyler-Smith. "The HapMap results were published less than a year ago and already the scale of the HapMap project has increased enormously – HapMap 2 will be out soon. Soon we'll see the first drafts of the entire genomes of humans being resequenced, which will be fabulous for evolutionary biologists.

"Of course, genetics will be the easy part of these studies; I'm sure that in a few years we will have a good picture of the regions of the genome that have been selected. But understanding the biology of what has gone on, and the causes of that selection, will be much more difficult."

References

1 Itzhaki J. Feature: The FOXP2 story: A tale of genes, language and human origins .

2 Bailey P. Feature: A series of fortunate events .

Further reading

• Jobling MA, Tyler-Smith C. The human Y chromosome: an evolutionary marker comes of age . Nat Rev Genet 2003;4(8):598–612.
• Xue Y et al. Recent spread of a Y-chromosomal lineage in northern China and Mongolia . Am J Hum Genet 2005;77(6):1112–6.
• Xue Y et al. Spread of an inactive form of caspase-12 in humans is due to recent positive selection . Am J Hum Genet 2006;78(4):659–70.
• Zerjal T et al. The genetic legacy of the Mongols . Am J Hum Genet 2003;72(3):717-21.

Links

• Feature: Lands of our fathers - Y-chromosome diversity and the histories of human populations .
• Feature: Mitochondrial DNA and human history .