The Seven Daughters of Eve (6 page)

Figure 2

To see how this works we can use again the metaphor of the tribe with its pole holding black and white discs. But this time the pole is the mitochondrial DNA and the tribe that split in two is a person who has two children. Both children inherit the same mitochondrial DNA, the genetic equivalent of the same pattern of discs on the pole. When they have their own children they pass on the mitochondrial DNA to them, and so it goes on down the generations. Very occasionally, random changes, called mutations, occur in the mitochondrial DNA which alter it a little bit at a time. These occur quite by chance when the DNA is being copied as cells divide. As time passes, more random changes are added to the DNA, which are then retained and passed on to future generations. Very slowly, the mitochondrial DNA of the descendants of that first individual, their common ancestor, becomes more and more different as more random mutations are introduced one at a time.

The lines on the tree in Figure 2 are reconstructions of the relationships among these sixteen people, worked out from the differences in their mitochondrial DNA, the exact nature of which we will examine shortly. But look for the moment at the tree itself. The deep trunk at the top has four Africans at the tips, while the other deep trunk contains individuals from the rest of the world
and
one more African. Within this ‘rest of the world' trunk, close branches sometimes connect people from the same part of the world, like the Asians and Papuans at the top or the Europeans at the bottom. But they also sometimes connect individuals from different places, like the branch near the middle that links a Papuan with an Asian and two Europeans. What's going on? The deep split between the exclusively African ‘trunk' and the rest of the world is another confirmation of the antiquity of Africa which the population trees also pick up. The confusion in the ‘rest of the world' trunk is confirmation of exactly what Arthur Mourant had in mind. It is ‘the mixing that marks the history of every living species'. Small wonder, then, that this diagram threw a very large spanner in the works of the population tree
aficionados
. It shows that genetically related individuals are cropping up all over the place, in all the wrong populations. You just cannot sustain the fundamental idea of a population being a separate biological and genetic unit if individuals within one population have their closest relatives within another.

Moreover, as we shall see in greater detail later on, by using the mutation process just described we can estimate the rate at which mitochondrial DNA changes with time. This means we can work out the timescales involved. When we do that, all the branches and the trunks converge to a single point, the ‘root' of the tree, at about 150,000 years ago. This had to mean that the whole of the human species was much younger and more closely related than many people thought.

The impact of ‘Mitochondrial DNA and human evolution' was dramatic. It came down very firmly on one side of the argument about a fundamental question of human evolution. For many years there had been an intense and polarized debate on the origins of modern humans, based on different interpretations of fossil skeletons, mainly the skull. Both sides agreed that modern
Homo sapiens
, the species to which we all belong, originated in Africa. Both sides also agreed that an earlier type of human, called
Homo erectus
, was an evolutionary intermediate between ourselves and much older and more ape-like fossils.
Homo erectus
first appeared in Africa about two million years ago and by one million years ago, or perhaps even earlier, it had spread out to the warmer parts of the Old World.
Homo erectus
fossils have been found from Europe in the west to China and Indonesia in the east.

All that was – and is – agreed by both sides of the argument. What divides them is whether or not there was a much more recent spread of modern humans from Africa. The ‘Out of Africa' school think there was, about 100,000 years ago, and that these new humans, our own
Homo sapiens
, completely replaced
Homo erectus
throughout its range. The opposing school of thought, the multi-regionalists, see clues in the fossils that suggest to them that
Homo sapiens
evolved directly from their local
Homo erectus
populations. This would mean that modern Chinese, for example, are directly descended from Chinese
Homo erectus
, and modern Europeans are similarly evolved from European
Homo erectus
, rather than being descendants of
Homo sapiens
who migrated from Africa. In the multi-regional scheme a modern European and a modern Chinese would have last shared a common ancestor at least one million years ago, while in the ‘Out of Africa' scenario they would be linked very much more recently.

What the mitochondrial gene tree did was to introduce an objective time-depth measurement into the equation for the first time. It showed quite clearly that the common mitochondrial ancestor of
all
modern humans lived only about 150,000 years ago. This fitted in very well with the ‘Out of Africa' theory and was enthusiastically welcomed by its supporters. But it came as a severe shock to the multi-regionalists. If all modern humans were related back to a common ancestor as recently as 150,000 years ago, they could not possibly have evolved in different parts of the world from local populations of
Homo erectus
that had been in place for well over a million years. Though the multi-regionalists, being thoroughly modern humans themselves, have refused to accept defeat, the mitochondrial gene tree dealt a wounding blow to their theory from which it has not yet recovered.

For us, it was great news. Mitochondrial DNA was catapulted by this controversy into its position as the prime molecular interpreter of the human past. A surge of research effort was bound to follow in laboratories all over the world. And that meant there would be lots of data with which we could compare our own results. If we were going to put the results from the old bones into a modern context, then we could not do better than use mitochondrial DNA.

Mitochondria are tiny structures that exist within every cell. They are not in the cell nucleus, the tiny bag in the middle of the cell which contains the chromosomes, but outside it in what is called the cytoplasm. Their job is to help cells use oxygen to produce energy. The more vigorous the cell, the more energy it needs and so the more mitochondria it contains. Cells from active tissues like muscle, nerve and brain contain up to one thousand mitochondria each.

Each mitochondrion is enclosed within a membrane. Arranged in an elaborate structure within the membrane are all the enzymes required for the final stage of aerobic metabolism. This is the part where the fuel we take in as food is burnt in a sea of oxygen. There are no flames and all the oxygen is dissolved, but it is as much a piece of combustion as what happens in a gas fire or a car engine. Fuel and oxygen combine to produce energy. Fires and engines produce their energy as heat and light. Mitochondria do not give off light when they burn fuel but they do heat up – it is partly the heat given off by mitochondria that keeps us warm. However, the main output is a high-energy molecule called ATP, which is used by the body to run virtually everything, from the contraction of heart muscles, to the nerves in your retina that is reading this page, to the cells in your brain that are interpreting it.

Buried right in the middle of each mitochondrion is a tiny piece of DNA, a mini-chromosome only sixteen and a half thousand bases in length. This is minuscule compared to the total of three thousand million bases in the chromosomes of the nucleus. Finding DNA in mitochondria at all was a big surprise. And it is very peculiar stuff. For a start, the double helix of this DNA is formed into a circle. Bacteria and other micro-organisms have circular chromosomes, but not complex multi-cellular organisms and certainly not humans. The next surprise was that the genetic code in mitochondrial DNA is slightly different from the one that is used in the nuclear chromosomes. Mitochondrial genes hold the code for the oxygen-capturing enzymes that do the work in mitochondria. However, many genes that govern the workings of the mitochondria are firmly embedded within the chromosomes of the nucleus.

How did this all come about? The current explanation is stunning. It is thought that mitochondria were once free-living bacteria that, hundreds of millions of years ago, invaded more advanced cells and took up residence there. You could call them parasites, or you could call their relationship with the cells symbiotic, with both cells and mitochondria doing something for each other. Cells got a great boost from being able to use oxygen. A cell can create much more high-energy ATP from the same amount of fuel using oxygen than it can without it. For their part, the mitochondria evidently found life within the cell more comfortable than outside. Very slowly, over millions of years, some of the mitochondrial genes were transferred to the nucleus, where they remain. This means mitochondria are now trapped within cells and could not return to the outside world even if they wanted to. They have become genetically institutionalized. Even now you can see the evidence of gene transfers between mitochondria and nucleus that didn't work out. The nuclear chromosomes are littered with broken fragments of mitochondrial genes that have moved across to the nucleus over the course of evolution. They can't do anything because they are not intact. So they just sit there, as molecular fossils, a reminder of failed transfers in the past.

There is something else which is unique to mitochondria. Unlike the DNA in the chromosomes of the nucleus, which is inherited from both parents, everyone gets their mitochondria from only one parent – their mother. The cytoplasm of a human egg cell is stuffed with a quarter of a million mitochondria. In comparison, sperm have very few mitochondria, just enough to provide the energy for swimming up the uterus as they home in on the egg. After the successful sperm enters the egg to deliver its package of nuclear chromosomes it has no further use for the mitochondria, and they are jettisoned along with the tail. Only the sperm-head with its package of nuclear DNA enters the egg. The plump, fertilized egg now has nuclear DNA from both parents, but its only mitochondria are the ones that were in the cytoplasm all along – and they all come from the mother. For that simple reason, mitochondrial DNA is always maternally inherited.

The fertilized egg divides again and again, forming first an embryo, then a foetus, which in turn becomes a new-born baby and, eventually, an adult. Throughout this process, the only mitochondria to be found are copies of the originals from the mother's egg. Though both males and females have mitochondria in all their cells, only women pass theirs on to their offspring because only women produce eggs. Fathers pass on nuclear DNA to the next generation, but their mitochondrial DNA gets no further.

Changes to DNA, both in the mitochondria and in the nucleus, arise spontaneously as simple mistakes during the copying that accompanies cell division. Cells have error-checking mechanisms which correct most mistakes, but a few escape this surveillance and get through. If these mutations occur in cells that go on to produce eggs or sperm, known as the
germline cells
, then they can be passed on to the next generation. Mutations that occur in the other body cells, called
somatic cells
– the ones that aren't going to produce germline cells – will not be passed on. Most DNA mutations have no effect at all. Only very occasionally, when they strike and disable a particularly important gene, will mutations be noticed. In the worst cases, these mutations can produce serious genetic diseases, some of which we shall encounter in a later chapter, but most of the time they are harmless.

The rate at which mutations occur in nuclear DNA is extremely low – roughly, only one nucleotide base in one thousand million will mutate at every cell division. Mitochondria, on the other hand, are not quite so vigilant with their error-checking and allow through about twenty times as many mutations. This means that many more changes are to be found in mitochondrial DNA than in the equivalent stretch of nuclear DNA. In other words, the ‘molecular clock' by which we can calculate the passage of time through DNA is ticking much faster in the mitochondria than in the nucleus. This makes mitochondria even more attractive as a tool in investigating human evolution. If the mutation rate were very low, then too many people would have exactly the same mitochondrial DNA and there wouldn't be enough variety to tell us anything much about developments over time.

There is yet another bonus. Although mutations are found all round the mitochondrial DNA circle, and this whole range was used by Allan Wilson and his students in ‘Mitochondrial DNA and human evolution', there is a short stretch of DNA where mutations are especially frequent. This section, about five hundred bases in length, is called the
control region
. It has managed to accumulate so many mutations because, unlike the rest of the mitochondrial DNA, it does not carry the codes for anything in particular. If it did, then many of the mutations would affect the performance of the mitochondrial enzymes. This does sometimes happen when mutations hit other parts of the mitochondrial DNA outside the control region; there are some rare neurological diseases which are caused by mutations in genes that disable essential parts of the mitochondrial machinery. Because they are so damaged, these mitochondria do not survive well and are only very rarely passed on to the next generation. So these mutations gradually die out. The control region mutations, on the other hand, are not eliminated, precisely because the control region has no specific function. They are neutral. It appears that this stretch of DNA has to be there in order for mitochondria to divide properly, but that its own precise sequence does not matter very much.

So here we have the perfect situation for our research: a short stretch of DNA that is crammed full of neutral mutations. It would be much quicker and cheaper to read the sequence of the control region, just five hundred bases, than the entire mitochondrial DNA sequence at over sixteen thousand bases. But was the control region going to be stable enough to be useful in examining human evolution? If the control region were mutating back and forth at a great rate at every generation, then it would be extremely difficult to make out any consistent patterns over the course of longer time spans. We knew already from the work of Allan Wilson that if we were going to dig down deep into the genetic history of our species,
Homo sapiens
, using mitochondrial DNA, we needed to cover at least 150,000 years of human evolution – say 6,000 generations at twenty-five years per generation. If mutation in the control region were too frantic or erratic, it would be very hard, if not impossible, to distinguish the important signals from all the incidental, irrelevant changes after a few generations. We needed a way of testing this before embarking on the time-consuming and expensive commitment of a large study of human populations. How could we best do this?

Ideally, I wanted to find a large number of living people that could be proved to be descended through the female line from a single woman. In the course of my medical genetics research on inherited bone disease, I had worked with several large families; so now I took out the charts on which I had recorded their pedigrees. Although these went back several generations, there were depressingly few continuous maternal lines connecting the living members of these families. I could ask for the families' help to put me in touch with relatives who were not shown on the charts; but it would be a long business. Still, there seemed nothing else for it, and I began to dig out their names and addresses. On my way back home that night, while I was thinking about something else, I experienced one of those rare moments when an idea suddenly arrives from the recesses of the mind, goodness knows how, and you know within a millisecond that it is the answer to your problem, even though you haven't had time to work out why. I suddenly remembered the golden hamster.

When I was a small boy, I read in a children's encyclopaedia that all the pet golden hamsters in the world were the descendants of just one female. I can definitely say that I had not thought about this again over the intervening decades. And yet the idea surfaced now. I do remember thinking at the time that the story couldn't possibly be true. But what if it were? This would be the ideal way to test out the stability of the control region. All the golden hamsters in the world would have a direct maternal line back to this ‘Mother of all Hamsters'. It follows that they would also have inherited their mitochondrial DNA from her, since it is passed down the female line in hamsters just as it is in humans. All I had to do was collect DNA from a sample of living hamsters and compare their control region sequences. I didn't need to have an accurate pedigree, because if there really had been only one female to start with they all had to trace back to her anyway. If the control region was going to be stable enough to be any use to us, then its sequence should be the same, or very similar, in all living hamsters.

I asked Chris Tomkins, an undergraduate student who, in the summer of 1990, had just started his final year genetics project in my laboratory, to see what he could find out about the golden hamster. The first thing he discovered is that, properly speaking, they are not called golden hamsters at all but Syrian hamsters. Chris went straight down to the Oxford public library and came back with some good news: he had found out that there was a National Syrian Hamster Council of Great Britain. He called the secretary and next day we were on our way to an address in Ealing, west London. Here we were greeted, with no little suspicion, by the secretary of the Syrian Hamster Club of Great Britain – Roy Robinson (now sadly deceased).

The late Mr Robinson was the product of a vanished age, a self-taught amateur scientist of great distinction. His dimly lit study was full of books on animal genetics, many of them written by himself. He pulled out his book on the Syrian hamster. His eyesight was very poor, and even with the help of very thick spectacles he needed to hold the text right up close to his face. He confirmed the story I had read as a boy. Apparently, in 1930 a zoological expedition to the hills around Aleppo (now Halab) in north-west Syria had captured four unusual small golden-brown rodents, one female and three males, and taken them back to the Hebrew University in Jerusalem. They were kept together, and the female soon became pregnant and gave birth to a litter. There was clearly going to be no difficulty in breeding them in captivity. The university began to distribute them to medical research institutes around the world, where they became popular as an alternative to the more usual rats and mice – though they were tricky lab animals, active only at night, bad-tempered and prone to bite their handlers (good for them!). The first recipient was the Medical Research Council institute at Mill Hill in north London, which passed some on to London Zoo. By 1938 the first golden hamsters had reached the United States.

Sometimes, lab animals that are no longer required are taken home by staff and kept as pets rather than being killed. Over time, hamsters spread from one household to another and, as their popularity increased, commercial breeders added them to their catalogues and groups of hamster enthusiasts started up. In 1947 a piebald hamster appeared in one breeding colony – the first of many coat colour varieties, caused by spontaneous mutations in the coat colour genes, it showed itself because of the inbreeding within the colony. It wasn't difficult to mate the mutants with each other and produce a pure-bred strain. Breeders became ever keener to find new coat colours, and over the next few years many different such mutants were discovered and pure-bred strains established – cream, cinnamon, satin, tortoiseshell and many more. Hamsters made good pets and the availability of strains with different coats only added to the interest. Thus began the population explosion: today there are over three million hamsters kept as pets all over the world.

Mr Robinson lived in an old horticultural nursery, which at the time we visited was quite run down. A long, rectangular plot enclosed by walls of beautiful old brick contained overgrown flower beds and a handful of greenhouses with cracked and broken panes. There were also two substantial sheds, and we made our way to the first one on the left, where Mr Robinson unlocked the door to let us in. We could not believe our eyes. Inside were rack upon rack of cages, all labelled and numbered, within each of which nestled a family of hamsters. Mr Robinson had collected an example of every single coat variety that had ever been produced, and was interbreeding them to unravel the genetics. There were pure-white hamsters, lilac hamsters, hamsters with short dark fur and hamsters with long fine coats like an angora goat. So eminent was Mr Robinson in the world of Syrian hamsters that each time a new coat mutant was discovered, a pair would be sent to Ealing. We were looking at the world reference collection. To cap it all, he opened an old ‘Quality Street' sweet tin and there inside, neatly stacked, were the dried skins of the original animals that had been sent to him. Martin Richards, who had made the trip along with Chris and myself, was so taken that he bought two hamsters from a pet shop in Ealing on the way home. He kept them in his flat for two years until they passed away. Of more immediate significance, we took away from Mr Robinson's collection a few hairs taken from each strain.