This month marks the 25th anniversary of the discovery of Mitochondrial Eve, a common ancestor of all humanity who helped rewrite the story of human evolution. In honor of this anniversary, we talked to the scientists who started it all.
The original paper, "Mitochondrial DNA and Human Evolution", appeared in the January 1987 issue of the scientific journal Nature. It was the work of three scientists who were then working at UC Berkeley: Rebecca Cann, now a professor of cell and molecular biology at the University of Hawaii; Mark Stoneking, currently a member of the Department of Evolutionary Genetics at Germany's Max Planck Institute for Evolutionary Anthropology; and Allan Wilson, who sadly died in 1991 at just 56. Wilson was one of the earliest - and, as such, one of the most controversial - scientists to use molecular biology and genetic markers to probe the origins of humanity, which culminated in the discovery of Mitochondrial Eve.
We recently had a chance to speak with Rebecca Cann and Mark Stoneking about their own origins, how they came to work on this landmark paper, and what they feel is the legacy of Mitochondrial Eve 25 years after it entered and forever altered the scientific discourse on the origins of humanity.
What initially drew you to this area of research?
Rebecca Cann: I did my undergraduate degree in genetics at UC Berkeley. I wanted to study human genetic diversity but the molecular tools then (1972) were so primitive that most people worked on fruit flies or molds or mice - and I wanted to study something that evolved with interesting behaviors. So I went to work for a drug company (AG Bayer) to help put an ex-husband through graduate school while I figured out what I wanted to be when I grew up. My work as a techician helped me understand that science was both repetitive and always changing as the instruments changed, so it was important to acknowledge that you would always have to be learning new things. By 1974, the use of new "DNA scissors" or restriction enzymes were showing up in publications about genes in mice, and I realized that the time had come to go back to school.
So I applied to grad school, but this time to work in molecular anthropology and human evolution with two people who were the world experts in this at Berkeley, Allan Wilson (in biochemistry) and Vince Sarich (in anthropology). Eventually I hooked up with a postdoctoral fellow of Allan's, Wes Brown, who had just finished some preliminary work for his thesis at Cal Tech on isolating and mapping restriction enzyme sites in human mitochondrial DNA (this was the first human DnA being sequenced for a large scale demonstration product by Fred Sanger's group at Cambridge in the UK). I needed more than just a blood sample to get enough tissue to purify human mitochondrial DNA (mtDNA) for my work, so I started going to Lamaze classes and talking to pregnant moms [about allowing] me to take the placenta of their babies after they gave birth. There are some kids in Berkeley and Oakland with pictures of their DNA sequences in their baby books! Now because of the invention of the polymerase chain reaction we can do the same work with a single hair or a simple cheek swab, as everyone who watches crime dramas on TV knows.
Mark Stoneking: I studied anthropology as an undergraduate, which in turn got me interested in population and evolutionary genetics. I ended up doing my master's degree on the evolutionary genetics of salmonid fish, using biochemical (protein) polymorphisms, which gave me an appreciation for the power of laboratory-based methods of assaying genetic variation to address questions about evolutionary history. When it came time to do my PhD, the first papers on mtDNA were just coming out, and it seemed to me that this could be the next big advance, in terms of being able to assay variation directly at the DNA level, instead of indirectly at the protein level. I ended up going to Allan Wilson's lab, where I just wanted to learn about mtDNA and didn't really care what organism I worked with.
But of the various mtDNA studies that were going on in various organisms – including mice, birds, frogs, etc. – the work that seemed the most interesting to me was the work another graduate student who was just finishing, Becky Cann, had been doing on human mtDNA variation. So I ended up following up on her work, and in doing so reawakened my interest in anthropology, plus I came to appreciate that for population history studies, humans are an ideal organism to work with, because there is so much other information from archaeology, linguistics, etc. that provides a rich source of hypotheses to test using genetic data, as well as a much more informed background for interpreting human genetic variation. Plus, everyone finds studies about ourselves and our past a lot more interesting than similar studies about other creatures.
What was the genesis of the paper?
RC: I purified human mtDNA from about 150 different donors, and tried to get a widespread pool of people from different areas of the world, representing different ancestral populations. This was an exciting time in anthropology, because some of the oldest protohuman fossils (Lucy) were being described from East Africa. In my grad program, I had to take human anatomy, paleontology, zoology, and primate behavior classes as well as my own training in molecular biology and genetics. So, I was exposed to a lot of new data that was giving quite a detailed picture of the earliest stages of human evolution, and was always struck by how silent the textbooks were regarding the more recent picture. Most of the experts thought and wrote that modern humans sprung up from the Middle East somewhere, and you had cave paintings, and cool blades, and suddenly they were us. I knew it had to be more complicated, new species usually arose from a single population that later spread because it had some unique adaptation.
So I started looking for geographic signals in my data, something that would help pinpoint a place and time. Our computer tools for analyzing the kind of data I had were still pretty crude, I had to use a mainframe and a punch-card deck for the first rounds of analysis, and the programs were so bad they could only take a small amount of my data, I had to run and rerun batches of 20-40 people at a time, trying to randomize their input. So, when the first computer program that could take what we call a matrix of 50 x 50 - meaning you could compare 50 people against each other - it was a big improvement. Then eventually it could do 125 x 125, and we were off and running.
And it started to show that if you had a female African ancestor, you would likely come out at the base of the human gene tree, and all other people would be linked, but not in a simple geographic way. People on different branches of the tree were mixed together, so if you had a picture of "race" in your head that said all Asians were more similar to each other than any one Asian was to any one European, you were wrong. So we started trying to understand how the mutation rate might be used to assess the antiquity of this variation - which populations were the oldest, which were the youngest, and how it was that they became mixed up in lots of different places in the world.
How did you come to work with your collaborators, and what how did the research process lead you to the eventual conclusions about Mitochondrial Eve?
RC: My thesis work ended when I filed for my PhD in 1982. I had been waiting for more donors from Australia, representing the Aboriginal peoples there, but finally felt I had to stop somewhere and wrap up. So a new grad student in the Allan Wilson lab, Mark Stoneking, began to work on this material as it trickled in, and eventually he was able to get donors from Papua New Guinea as well.
These were important samples because in my human trees that I was producing, after some African samples, it appeared that the most mutationally divergent people were Aboriginal Australians and Papua New Guinea donors. Since we knew how long Australia and Papua New Guinea had been isolated from southeast Asia by the global drops in sea level, this gave us an internal calibration point for the molecular clock to say how many mutations had accumulated uniquely in one place or the other over X thousands of years. And when you put the big picture together, you understood that the very base of the tree might be as much as 200,000 years old, and it was overwhelmingly African. The inference was the one woman at the base of that mtDnA tree was the universal lucky mother of all modern people, the one woman with an unbroken line of maternally inherited genes that came down to us today.
MS: The bulk of the data came from Becky Cann's work – for a variety of reasons I ended up analyzing mtDNA variation in aboriginal Australians and New Guineans for my PhD, and the first data I obtained were also part of the study. My main contribution was to assist with the analyses, which were quite daunting in those days – this was considered a huge dataset, and computerized methods to deal with such data were just then being developed. For example, to get a copy of the program that we used to generate the phylogenetic tree, I had to send a computer tape to the author of the program, who then copied the program to the tape and sent it back to me, which I then took to the computer center to upload onto the mainframe.
We did have a remote terminal in the lab to run the program, so at least we didn't have to use punch cards, but we then had to go to the computer center to pick up and make sense of the pages and pages of output. From the branching pattern in the phylogenetic tree, as well as from analyses of how much sequence variation there was in different populations, we came to the conclusion that there was an African origin of human mtDNA – basically, all of the variation outside Africa was a subset of the variation within Africa. And to figure out how much time it took to generate all of the variation in human mtDNA, we used a molecular clock approach to estimate that it would have taken about 140,000 – 280,000 years.
How much, if at all, did the debate on the nature of human origins (multiregional hypothesis vs. Recent African Origin) inform your work?
RC: A lot. As a geneticist, I realized that humans spread over 12,000 kilometers could not represent a single gene pool with mutational equilibrium. So multiregionalism theoretically was unlikely to work. Also, it just didn't fit with the gene pools I was seeing at a molecular level. If humans were so isolated for 500,000 to 1 million years, why was it that 2 humans were molecularly so similar? Only 7% of the entire human genetic diversity can be accounted for by geography. We knew that already, from blood protein work. It is the same at the DNA level, as you might expect. So now we know that from Neanderthals and Denisovans that some isolated bits of earlier human genomes have survived in some specific populations, but that is a tiny fraction, a REALLY tiny fraction, of the total diversity of human genes.
MS: When we started the work I was very ignorant about the debates concerning human origins, and in fact the debate had really just started with suggestions by Chris Stringer and Gunter Brauer about the potential central role of Africa in the origin of our species. And in fact, while we were analyzing the data, we weren't sure there would be much of anything meaningful in human mtDNA variation or not. But when we got the first trees, as well as the variation estimates, then it was clear to us that we were seeing a signal of a recent African origin in human mtDNA.
Were you aware of how big a reaction your paper would provoke?
RC: As soon as I started talking about the preliminary analysis of data at scientific meetings I got immediate hostile reactions, especially from some paleontologists. Also some human geneticists that didn't understand population genetics.
MS: I was completely taken by surprise – I was still a naïve graduate student, and this was my first experience dealing with this sort of attention (a lot of which fell on my shoulders because by that time Becky had moved to Hawaii and Allan was on a sabbatical leave in Cambridge, England, so by default I was easiest for the press to reach). It was, let us say, a real eye-opener, and I learned some valuable lessons.
How would you characterize your experience with both the academic and the public reaction to your breakthrough?
RC: People tended to have one of two reactions - they either thought they knew this answer all along, or, they thought I couldn't possibly be right. Some were very dismissive, and you may know that only about 40% of Americans even believe in biological evolution, and a lot of these people even think that it doesn't apply to humans. So there were some religious objections, especially when the Mitochondrial Eve story took off and you had to say that Mitochondrial Eve wasn't the same woman as the Eve in the Judeo-Christian Bible's story of genesis.
It took about five years before i could drink a strawberry margarita, because it looked so much like the placental purification steps for mtDNA preparations. (I used a waring blender to pulverize tissue in the lab). I also had a lot of fun with people who, when we switched to collecting hair samples from donors, wanted to give us pubic hairs instead of head hairs because that had to be a better source of learning about human genealogy.
MS: On the academic side, one would like to think that scientists would focus their criticisms on potential weaknesses in the evidence and interpretations, and while that was true for most, there were a minority who chose to attack the study using any and every means, no matter how preposterous the allegation, simply because they didn't like the conclusions. I was also amazed by the public reaction, in particular to the concept that all of the mtDNA variation traces back to a single ancestor (the so-called "mtEve"). As an evolutionary geneticist, to me this seemed patently obvious – given a single origin of life from which all living things on this planet are descended, then it has to be the case that the variation in any gene traces back to a single ancestor at some point in the past. The only alternative would be to suppose that some of us got some of our genes from someplace else, like Mars! For mtDNA, because it is maternally inherited, this ancestor had to be female. And yet this concept really resonated with the public – and was even misunderstood by a few colleagues – perhaps because this was the first time we really had the data to say something about when and where an ancestor of one of our genes lived.
Twenty years on, what would you say has been the biggest legacy of the paper?
RC: Humans are remarkably alike, once we get past our seven or so genes that influence skin pigmentation and 100 or so genes that help model the shapes of our faces, noses, eyes out of 20,000+ total. I tell my students they should all celebrate black history month, since they are all Africans genetically. Fossils are an important way to understand the past, and incredible records of evolutionary change. But so is our DNA. And it survives so we can pass it on and learn from our studies. We are each uniquely endowed and challenged. We are a new species, we went through a period where we were like an endangered species, a very small isolated population, and now we are 7 billion. We came very close to going extinct. We could do it again, unless we take better care of the planet.
MS: Despite all the criticisms of this and subsequent studies of human mtDNA variation, the central conclusion of the study remains unchanged – we have mtDNA data from hundreds of thousands of individuals, complete mtDNA genome sequences from thousands of individuals, and the current view is that the human mtDNA ancestor lived in Africa some 150,000-200,000 years ago. So 25 years on, our view of human mtDNA variation is remarkably similar to what we published. Moreover, our paper brought genetics into the debate over human origins, opening a new line of evidence which ultimately led to the resolution of the debate in favor of a recent African origin of modern humans followed by some assimilation of non-African hominins (e.g., Neandertals and Denisovans) – i.e., "leaky replacement" is currently the prevailing model for human origins.
Are there any common misconceptions about your work that you would like to correct?
MS: There remains much confusion over what is meant by all of our mtDNA variation tracing back to a single female ancestor who lived in Africa some 150,000 – 200,000 years ago. Just to clarify, she was not the only female alive at that time; she was a member of a population who also had mtDNA types, but these other mtDNA types ultimately went extinct because the descendants of these other females either left no offspring or had only male offspring. This random extinction of lineages, coupled with a single origin of DNA, is sufficient to guarantee that all of the variation in all of our genes has to trace back to a single common ancestor at some point in the past. And even though all of our genes have ancestors, our mtDNA ancestor was not the ancestor of all of our other genes – they trace their ancestry back to different individuals (and even different species), living at different times in different places. This is why the "mtEve" designation is incorrect – Allan Wilson favored the term "lucky mother", to emphasize the role of chance in the survival of mtDNA lineages over time, but I guess it just wasn't as catchy a term as Mitochondrial Eve.
Our huge thanks go to Rebecca Cann and Mark Stoneking for participating in this interview.
Top image by zentilia, via Shutterstock.
Skulls by Ryan Somma on Flickr.
Punch cards by Marcin Wichary on Flickr.
Illustration of mtEve family tree via Wikimedia.
Early humans mural by Karen Carr.
Strawberry margarita by rian0306 on Flickr.