Power, Sex, Suicide
Page 33
If, as we have seen, the deepest distinction between the sexes relates to restricting the germ-line passage of mitochondria, then the barrier between the sexes seems curiously shaky. The language one tends to read in journals or books speaks of outright conflict, along the lines of: ‘organelles from more than one parent are not tolerated in the offspring.’ In more mundane reality, the condition that forces nature to differentiate two sexes, obliging us to mate with only half the population, is constantly collapsing and reforming. Mitochondrial heteroplasmy seems to be tolerated with surprisingly few detrimental effects in many cases—there is little sign of conflict. So while the evidence suggests that mitochondria really are central to the evolution of two sexes, genomic conflict may not be all there is to it. Recent research suggests that there are other, more subtle, but probably more pervasive and fundamental reasons, too.
The area of research forcing this new thinking, ironically, is another field altogether: the study of human prehistory and population movements, by tracking human mitochondrial genes. Some of the most arresting insights into prehistory, such as our relationship with the Neanderthals, have derived from such studies of mitochondrial DNA. All these studies rest on the assumption that the inheritance of mitochondrial DNA is strictly maternal, that any mixing is simply not tolerated. In this hotbed of research, controversial data have recently raised questions about the validity of this assumption in our own case. But if some of the once iron-cast conclusions now look a bit more rickety, they do give a new insight not just into the origin of the sexes, but also into previously unexplained aspects of infertility. In the next two chapters, we’ll find out why.
14
What Human Pre-History Says About the Sexes
In 1987, Rebecca Cann, Mark Stoneking, and Allan Wilson, at Berkeley, published a celebrated paper in Nature, which (although it built on earlier work) was to revolutionize our understanding of our own past. Instead of looking to the fossil record, or to genes in the nucleus, they studied the mitochondrial DNA of 147 living people, drawn from five geographical populations. They concluded that the samples were closely related, and ultimately inherited from a single woman, who lived in Africa about 200 000 years ago. She became known as ‘African Eve’, or ‘Mitochondrial Eve’, and to the best of our knowledge everyone on earth today descends from her.
The radical nature of this conclusion needs to be placed in perspective. There has long been an unresolved controversy between two warring tribes of palaeoanthropologists—those who believe that modern man issued from Africa in the fairly recent past, displacing earlier groups of migrants, like the Neanderthals and Homo erectus; and those who believe that humans have been present in Asia as well as Africa for at least a million years. If this latter view is correct, then the evolutionary transition from archaic to anatomically modern humans must have happened in parallel in different parts of the old world.
These two views carry a potent political charge. If all modern humans came from Africa less than 200 000 years ago, then we are all the same under the skin. We have barely had time, in an evolutionary sense, to diverge, but we can perhaps be held responsible for the extinction of our closest relatives, such as the Neanderthals. This theory is known as the ‘Out of Africa’ hypothesis. On the other hand, if the human races evolved in parallel then the differences between us are not skin deep, and our unique racial and cultural identities are firmly grounded in biology, challenging our ideals of equality. Both scenarios could have been offset by interbreeding, to an unknown degree. The dilemma is exemplified by the fate of the Neanderthals. Were they a separate subspecies driven to extinction, or did they interbreed with anatomically modern Cro-Magnons, who arrived in Europe around 40 000 years ago? Bluntly, are we guilty of genocide or gratuitous sex? Today we seem distressingly capable of both, sometimes simultaneously.
The patchy fossil record has so far proved inconclusive, not least because it is extremely difficult to tell from a few scattered fossils of widely differing ages whether one population gives rise to another in the same place, or falls extinct, or is displaced by another from a different geographical region, or indeed whether two populations interbred. Numerous fossil finds over the last century—a train of missing links—have demonstrated the possible outlines of human evolution from ape-like ancestors, to all but the most unbelieving of Creationists. Brain size, for example, more than tripled in successive hominid fossils over the last four million years. But the actual line of evolution from Australopithecines, like Lucy, around three million years ago, through Homo erectus, and finally to Homo sapiens, is fraught with unresolved issues. How can we tell if a fossil discovery represents our own ancestors, or is simply a parallel species, now fallen extinct? Was Lucy really our direct ancestor, or just an extinct, upright, knuckle-dragging ape? All we can say for sure is that there are plenty of skeletons in the closet that exhibit morphologies intermediate between the apes and humans, even if it is difficult to assign their place in our own ancestry. Charting prehistory from ancient skeletal morphology alone is at best an uncertain endeavour.
In terms of our more recent ancestry, the fossil record is also dumb. Did we interbreed with the Neanderthals? If so, we might hope one day to discover a hybrid skeleton, displaying a mosaic of features intermediate between the robust Neanderthal Man and the gracile Homo sapiens. Occasionally such claims are made, but rarely convince the field. Here is the kindly Ian Tattersall, commenting on one such case: ‘the analysis… is a brave and imaginative interpretation, but it is unlikely that a majority of palaeoanthropologists will consider the case proven.’
One of the biggest problems of palaeoanthropology is its heavy reliance on morphology. This is inevitable, as little else remains. Isolating DNA would be a big help, but in most cases this is impossible. In virtually all fossil skeletons, the DNA slowly oxidizes, and very little survives beyond about 60 000 years. Even in more recent skeletons, the quantities of nuclear DNA that can be extracted are so small that sequencing is unreliable. Thus at present it seems virtually impossible to resolve our past from the fossil record alone.
Luckily, we don’t necessarily need to. In principle, we can search within ourselves to read the past. All genes accumulate mutations over time, and as they do so their sequence of ‘letters’ slowly diverges. The longer that two groups have diverged, the more differences accumulate in the sequences of their genes. Thus, if we compare the DNA sequences of a group of people, we can calculate roughly how closely related they are, at least relative to one another. People with just a few sequence differences are more closely related than people with lots of them. By the 1970s, geneticists were becoming involved in human population studies, scrutinizing the differences between genes in different races. The results implied that there is less variation between races than had been thought—as a rule of thumb, there is more variation within races than there is between races, implying that we all share a relatively recent common ancestor. Moreover, the deepest divergences are found in sub-Saharan Africa, implying that the last common ancestor of all human races was indeed African, and lived relatively recently, certainly less than a million years ago.
Unfortunately, there are various drawbacks to this approach. Genes in the nucleus accumulate mutations very slowly, over millions of years, and indeed we still share 95 to 99 per cent of our DNA sequence with chimpanzees (depending on whether we include non-coding DNA in the sequence comparison). If gene sequences can barely tell the difference between humans and chimpanzees, then clearly we need a more sensitive measure to distinguish between human races. Another problem with gene sequences is the role of natural selection. To what extent are genes free to diverge from each other at a steady pace (neutral evolutionary drift), and when does selection constrain the rate of change, by favouring particular sequences? The answer depends not just on the gene, but also on the shifting interactions of genes with each other, and with environmental factors like climate changes, diet, infection, and migration. There is rarely an easy answer.
Down the maternal line
This is where Cann, Stoneking, and Wilson stepped in with their study of mitochondrial DNA, nearly two decades ago. They pointed out that the odd mode of mitochondrial inheritance solved many of the problems associated with nuclear genes. The differences made it possible not only to trace human lineages, but also to give a tentative estimate of dates.
The first critical difference between mitochondrial and nuclear DNA is the mutation rate. On average, the mutation rate of mitochondrial DNA is nearly twenty times faster than nuclear DNA, although the actual rate varies according to the genes sampled. This fast mutation rate equates to a fast rate of evolution (but we should beware of always equating the two, as we’ll see later). The fast rate of evolution stems from the proximity of mitochondrial DNA to the free radicals generated in cellular respiration. The effect is to magnify the differences between races. While nuclear DNA can barely distinguish between chimps and humans, the mitochondrial clock ticks fast enough to reveal differences accumulating over tens of thousands of years, just the right speed for peering into human prehistory.
The second difference, said Cann, Stoneking, and Wilson, is that human mitochondrial DNA is inherited only from our mothers, by asexual reproduction. Because all our mitochondrial DNA comes from the same egg, and is replicated clonally during embryonic development, and throughout our lives, it is all (in theory) exactly the same. This means that if we take a sample of mitochondrial DNA from, say, our liver, it should be the exactly the same as a sample taken from the bone, and both should be exactly the same as a random sample taken from our mother—and hers should be exactly the same as her own mother’s, and so on back into the mists of time. In other words, mitochondrial DNA works like a matriarchal surname, linking a string of individuals together down the corridor of the centuries. Unlike nuclear genes, which are shuffled and redealt every generation, mitochondrial genes allow us to track the fate of individuals and their descendents.
The third aspect of mitochondrial DNA that the Berkeley team drew on is its steady rate of evolution: the mutation rate, though fast, remains approximately constant over thousands or millions of years. This is ascribed to neutral evolution, the assumption that there is little selective pressure on mitochondrial genes, which serve a restricted and menial purpose (so the argument goes). Sporadic mutations occur at random over generations, and as averages balance out, accumulate at a steady, metronomic speed, leading to a gradual divergence between the daughters of Eve. This assumption is perhaps open to question, and later refinements of the technique have concentrated on the ‘control region’, a string of 1000 DNA letters that does not code for proteins, and so is claimed not be subject to natural selection (we will return to this assumption later).1
So how fast does the mitochondrial clock tick? On the basis of relatively recent, approximately known colonization dates (a minimum of 30 000 years ago for New Guinea, 40 000 years ago for Australia, and 12 000 years ago for the Americas), Wilson and colleagues were able to calculate a divergence rate of about 2 per cent to 4 per cent every million years. This figure matches the rate estimated on the basis of divergence from chimpanzees, which began about six million years ago.
If this speed is correctly calibrated, then the actual measured differences between the 147 mitochondrial DNA samples give a date for their last common ancestor of about 200 000 years ago. Furthermore, in agreement with nuclear DNA studies, the deepest divergences were found among African populations, implying that our last common ancestor was indeed African. A third important conclusion of the 1987 paper related to migration patterns. Most populations from outside Africa had ‘multiple origins’, in other words, peoples living in the same place had different mitochondrial DNA sequences, implying that many areas were colonized repeatedly. In sum, Wilson’s group concluded that Mitochondrial Eve lived fairly recently in Africa, and the rest of the world was populated by repeated waves of migration from that continent, lending support to the ‘Out of Africa’ hypothesis.
Not surprisingly these unprecedented findings gave birth to a dynamic new field, which dominated genealogy in the 1990s. The unresolved questions raised by skeletal morphology, by linguistic and cultural studies, by anthropology and population genetics, could at last all be answered with ‘hard’ scientific objectivity. Many technical refinements have been introduced, and calibrated dates modified (Mitochondrial Eve is now dated to about 170 000 years ago), but the basic tenets presented by Wilson and his colleagues underpinned the entire edifice. Wilson himself, an inspiring figure, sadly died of leukaemia at the height of his powers, at the age of 56, in 1991.
Surely Wilson would have been proud of the achievements of the field he helped found. Mitochondrial DNA has answered many questions that had seemed eternally controversial. One such question is the identity of the people living in the remote Pacific archipelagos of Polynesia. According to the famous Norwegian explorer Thor Heyerdahl, the Polynesian islands were populated from South America. To prove it he built a traditional balsa-wood raft, the Kon-Tiki, and set sail with five companions from Peru in 1947, arriving in the Tuamotus archipelago, 8000 km away, after 101 days. Of course, proving that a feat is possible is not the same as proving that it actually happened. Mitochondrial DNA sequences speak otherwise, corroborating earlier linguistic studies. The results suggest that the Polynesians originated from the west in at least three waves of migration. About 94 per cent of the people tested had DNA sequences similar to the peoples of Indonesia and Taiwan; 3.5 per cent seemed to have come from Vanuatu and Papua New Guinea; and 0.6 per cent from the Philippines. Interestingly, 0.3 per cent had mitochondrial DNA matching some tribes of South American Indians, so there is still a remote chance that there could have been some prehistoric contact.
Another difficult question apparently resolved was the identity of the Neanderthals. Mitochondrial DNA taken from a mummified Neanderthal corpse (found in 1856 near Düsseldorf) showed that their sequence is distinct from modern humans, and no traces of Neanderthal sequence have been found in Homo sapiens. This implies that the Neanderthals were a separate subspecies, which fell extinct without ever interbreeding with humans. In fact, the last common ancestor of Neanderthals and humans probably lived about 500 to 600 thousand years ago.
These findings are just two of the many fascinating insights into human prehistory afforded by mitochondrial DNA studies. But every silver lining has a cloud. A rather simplistic view of mitochondria has become the mantra, which is repeated ever more succinctly, and ever more misleadingly; the provisos are lost in the telling. We are told that mitochondrial DNA is inherited exclusively down the maternal line. There is no recombination. It is not subject to much selection because it codes for only a handful of menial genes. The mutation rate is roughly constant. The mitochondrial genes represent the true phylogeny of people and peoples because they reflect individual inheritance, not a kaleidoscope of genes.
This mantra generated unease in some quarters from the beginning, but only recently have these misgivings found substance. In particular, we now have evidence of genetic recombination between maternal and paternal mitochondria
, of discrepancies in the ticking of the mitochondrial ‘clock’, and of strong selection on some mitochondrial genes (including the supposedly ‘neutral’ control region). These exceptions, while raising some doubts about the validity of our inferences into the past, sharpen our ideas of mitochondrial inheritance, and help us to grasp the real difference between the sexes.
Mitochondrial recombination
If mitochondria are passed down the maternal line exclusively, then there would seem to be little possibility for recombination. Sexual recombination refers to the random swapping of DNA between two equivalent chromosomes, to make up two new chromosomes, each of which contains a mixture of genes from both sources. Clearly DNA from two distinct sources—two parents—is needed to make recombination possible, or at least meaningful: swapping genes from two identical chromosomes makes little sense, unless one of the two chromosomes is damaged; and this, as we shall see, does raise a spectre. In general, however, during sexual reproduction, the pairs of chromosomes in the nucleus are recombined to generate new groupings of genes, mixing up genes from different parents or grandparents, but this does not happen with mitochondrial DNA, as all the mitochondrial genes are derived from the mother. Thus, according to orthodoxy, mitochondrial DNA does not recombine: we don’t see a mixture of mitochondrial DNA from both the father and the mother.