Unlocking the Past Read online

  Similar contrasts underpin the research developments discussed in each of the following afterwords. The data we have recovered has changed and the power of our methods for primary analysis have grown, but in addition to that, the entire landscape within which we can apply those methods to that data and handle the results has, by a parallel revolution in information technology, been transformed.

  2

  the quest for ancient DNA

  a revolution in the life sciences

  The double helix is a little younger than I am. To be more precise, I was a toddler of two years old when James Watson and Francis Crick rushed into the Eagle Pub at Cambridge to celebrate their breakthrough in understanding the structure of the molecule at the heart of all life. They had established that DNA was made up of two entwined sugar-phosphate strands, a twisted ladder whose cross-bars were made up of ‘base-pairs’. These bases were chemical units reaching out from each sugar unit in the strands, to pair up with their partner on the opposite strand. Four types of bases could be identified, in a variety of different permutations. Watson and Crick had unravelled a mechanism by which a chemical code could be passed on from cell to cell, forming the blueprint for all life on earth.

  My only involvement in all this was as a prolific factory of this remarkable molecule myself, in much the same way as any growing organism. The reason I place myself in the story is to emphasize the relatively brief history of the profound revolution in the life sciences that followed, and the pace at which our understanding of the molecular basis of life has grown. At the time of writing, we have the knowledge and power to replicate in the lab the entire process of reproduction, to transfer genes between unrelated organisms and to recover genetic information from organisms that have been dead for thousands of years. By the time this book reaches you, new discoveries will have been made. Such is the pace of DNA research. By the time I had grown from a toddler to a schoolboy, the double helix was part of the standard biology curriculum. By then, Watson and Crick’s central thesis had been fully corroborated, and the way in which their base-pair code operated had been unscrambled. We learnt how the various permutations of the four bases found along the DNA molecule–Adenine (A), Thymine (T), Guanine (G) and Cytosine (C)–were read on to shorter strands of a very similar molecule called RNA. If DNA was the master blueprint, then the working diagrams were formed of RNA. They were moved around the cell to the point where the proteins were fashioned. Each RNA strand copied bits of the master code using almost the same bases, but the place of thymine was taken by another base, uracil. These working diagrams moved to cellular workshops called ‘ribosomes’ where the code was read to build proteins. Some of these proteins were the structural materials of life itself, materials such as the collagen in bone, or the keratin in hair. Others, such as the haemoglobin in blood, and hormones such as insulin, performed the major chemical tasks of life. A still larger group comprises the enzymes, described by Francis Crick as the ‘machine tools’ of biological chemistry, the things that fashion and enable most of the subsequent chemical pathways in the living organism. In short, the determinate link between DNA, RNA and protein was at the heart of all living processes. In little more than a decade after the recognition of the DNA double helix, the key elements of the genetic blueprint for the chemistry of life had been mapped out.

  By the time I was passing the Eagle Pub myself on the way to Cambridge lectures in natural science, we were catching the first glimpses of another momentous advance. I had learnt at school about the clarity of vision of molecular scientists of how life’s processes unfolded. At that time, the scientists were still observers. By the 1970s they were taking one step further and intervening actively. What enabled this was the discovery of the molecular scissors that could chop up the DNA strand at particular points on the sequence. These ‘restriction enzymes’ could home in on a characteristic sequence of bases and break the chain at that point. So, for example, an enzyme called ‘EcoRI’cuts wherever it finds bases in the order GAATTC, and another called ‘SmaI’cuts wherever it encounters CCCGGG. With enzymes such as these that allowed the double helix to be dissected, it was becoming possible to work with targeted fragments of the DNA sequence, to separate them out, to establish the order of their bases and to move them from one organism to another. In the later 1950s and 1960s the chemical blueprint for life was being mapped out. With the 1970s came the beginnings of intervention into that process, and the basis for gene cloning and genetic engineering as well as the possibility of recovering fragments of this chemical process from the distant past.

  cloning from the past

  Gene cloning was developed for medical and agricultural purposes, to allow the manipulation of genes that were affecting either health or productivity in a negative way. Different enzymes were used to cut out a targeted stretch of DNA from one species and insert it into the DNA of a ‘host’, a separate species that was easy to work with in the laboratory, a laboratory mouse or a microbe. In the host it could be bulked up through normal growth and reproduction, modified and reinserted in either the original host or a completely new host. Quite incidentally, the procedures perfected for performing these tasks were just the ones needed to track down those few traces of ancient DNA that might be surviving in an archaeological specimen.

  In the case of ancient tissue, it could not be taken for granted that any DNA would survive at all. Archaeological preservation that appeared to be excellent quite typically involved only a selection of the molecules within the original organism, with an expected bias towards the dense structural molecules rather than those doing sensitive biochemical work. If there was any DNA surviving, then it was more than likely that it would be damaged and fragmented, and present in very small quantities. Because the gene-cloning process starts with cutting the sequence at very specific points, the ‘restriction sites’, it would be possible to design the restriction enzymes quite close together so that even short fragments with some internal damage could be isolated. Within the host organism, the normal processes of growth and propagation would transform the tiny quantities into manageable amounts for study. It was only a matter of time before cloning would be attempted on ancient tissue.

  The obvious tissues to begin with were those in which biological breakdown had at least been arrested by a shortage of either water or oxygen. Attention turned to a range of ancient ‘mummies’. Mummified bodies are those which have desiccated so speedily after death that a large part of the soft tissue remains in place on the skeleton. The process can be natural, artificial or a combination of the two. In artificial mummification, the internal organs in which breakdown begins are removed, and a variety of substances added to fix the soft tissue in various ways. One good reason to be optimistic about molecular survival was the appearance of mummified soft tissue under the microscope. As far back as 1911, attempts had been made to rehydrate the soft tissue from mummified bodies, and on a number of occasions the cell nuclei could be made out in these reconstituted tissues. If the nuclei were visible, perhaps their key components could still be found.

  In 1981, the first claim appeared in print. Two Chinese scientists, G. Wang and C. Lu, isolated and identified nucleic acids from the preserved liver of a corpse from a 2,000-year-old Han dynasty tomb from Ch’ang-sha, the capital of Hunan Province. The result was published in Chinese in a journal not widely available in the West. The find made little impact until an American group set out on a similar quest. The first focus for these new ideas was the University of California at Berkeley. Around the time of Wang and Lu’s publication, a number of scientists from around the Berkeley campus were speculating about ancient DNA. These scientists met together as the Extinct DNA Study Group, and tossed around ideas about where the science might lead. One Berkeley scientist, the geneticist Allan Wilson, had taken on a graduate student to track down ancient DNA from museum specimens. In many ways a preserved museum animal is very similar to a mummified body. Its preservation has involved rapid drying, removal of some internal orga
ns and the addition of certain preservatives. Wilson’s student, Russell Higuchi, was interested in a species that had been sighted by a number of eighteenth- and nineteenth-century travellers to South Africa. Charles Darwin made mention of the ‘quagga’ in his journal as the Beagle passed beneath the Cape of Good Hope in the 1830s. It was a timid, zebra-like animal, distinguished from the latter by the restriction of its stripes to the front of its body. Half a century after the Beagle had sailed, quaggas were extinct, surviving only as museum specimens. One such specimen was held by the Museum of Natural History at Mainz in Germany.

  The Mainz specimen was not just a skeleton; some dried skin was also attached. Higuchi took a sample of this skin and observed some dry muscle tissue attached to the underside. He sampled this muscle tissue, cleaned it up, and separated its molecular components into fractions. Then came the critical test. The fraction expected to contain any surviving DNA was mixed with DNA from a modern mountain zebra. In theory, quagga DNA should be sufficiently similar to zebra DNA for the two to bind. If any quagga DNA did survive, it should bind with the zebra DNA. Higuchi came up with a positive result.

  Having found quagga DNA, he could then go on to isolate it and sequence the bases that made up its genetic code. Two lengths, each of just over 100 base-pairs, were identified and positioned along the genetic sequence of the living horse genus Equus. One length belonged to a gene of unknown function, the other to a gene used in the energy management system of the animal’s cells. Having established what the strands of ancient DNA were, the slight differences between those sequences and the corresponding sequences in other mammals could be used to build a ‘family tree’ of species ancestry, or ‘phylogenetic tree’, to use the correct term. The tree placed the extinct quagga very close to the zebra, somewhat further from the cow and even further from humans. A century earlier, the genus Equus had been used as a key example of how fossils told the story of Darwinian evolution. Now the same genus had come to the fore again, with the quagga as the first extinct animal to reveal a small part of its genetic code.

  The quagga had long since ceased to be. Its bones no longer carried blood and its tanned dried skin no longer protected its body. Yet the very same information that brought it to life in the first place was, in part, recoverable. Its DNA had come alive again–the very process of cloning demanded that it do so. The bacterially carried strand was as much a part of a living cell as the bacterial DNA itself. Another way of looking at this strange phenomenon is that, by targeting and locating ancient DNA in a dead organism, the analysis moves beyond the distinction between life and death to the encoded information upon which that distinction is constructed. Among all molecules, ancient DNA probes more deeply into past life than any other. If Higuchi could reach his target with a well-preserved quagga, what about a well-preserved human, such as the Chinese scientists had studied?

  About the same time that Wilson and Higuchi were looking back in time from the vantage point of molecular biology, a young Swedish Egyptologist, familiar with much older specimens than quagga, was looking forward in time to the newly emerging possibilities of molecular science. Having progressed from his Egyptology studies to a PhD in molecular medicine, Svante Pääbo won a grant from the University of Uppsala Faculty of Medicine to look into the possibility of DNA surviving in ancient humans. He gathered samples from Egyptian mummies held in various European Museums, some of which dated back to the third millennium BC. A piece of skin was carefully taken from a mummified woman’s ear and a stain called ethydium bromide was applied. This stain attaches itself to DNA, and DNA alone, and is used to mark out visually its presence or absence. When Pääbo looked down his microscope he could detect nuclei within the skin cells of that ear, nuclei that carried the stain that demonstrated that her DNA had survived. This he managed to repeat with other mummies, which spurred him on to attempt to clone the DNA. He eventually achieved this with the skin of a young boy, isolating a 500-base-pair sequence containing a so-called ‘Alu-sequence’.

  This ability to reach back in time and recover the molecule at the very heart of life was attracting increasing interest, but the discovery would remain a novelty unless there was real information that could be gleaned from it. When Pääbo published his findings of DNA from the Egyptian mummies in Nature, he speculated optimistically about the possibilities of the new research (Pääbo, 1985). Perhaps we could start looking at gene frequencies; virus evolution; descent of the Nile valley population; Pharaonic and interfamily relations. It was one thing raising these possibilities, but Pääbo knew that he had to find some real information from ancient DNA to maintain the interest his work had attracted. Around this time, a further step forward in methodology helped him significantly on his way.

  mimicking nature: the polymerase chain reaction

  One of the challenges of ancient DNA analysis was to make sense of tiny quantities of genetic material. This is a challenge that living organisms meet with every new generation. Each adult individual is only formed as a consequence of the cell system’s ability to make sense of two individual sets of genetic code, one from the egg, the other from the sperm. The natural process of doubling up at the heart of cell proliferation makes it possible to work with DNA information that starts in tiny quantities. The double helix unwinds, each strand forming the template to construct a new partner. Where one double helix once was, now there are two, then four, then eight and so on. The gene-cloning methods that Pääbo had adopted used a fast-growing organism to perform that doubling up. It was not long before someone realized that we could go one step further and automate the process.

  Our growing understanding of the DNA molecule was spawning an entirely new profession, that of the now familiar ‘genetic engineer’. One of the first generation of these molecular mechanics was Kary Mullis, who was as familiar with the curves and bends in the double helix as a car mechanic might be with engines and carburettors. Beyond this, he was a lateral thinker extraordinaire, whose achievements included the invention of a light-sensitive plastic, and the publication in Nature of a consideration of the cosmological consequences of time reversal. The next major step in DNA manipulation he worked out, not in a laboratory surrounded by equipment and reagents, but in his head while on a weekend drive through the Californian redwoods and, incidentally, while chewing over a problem that was quite different from the one he was about to address with astounding success.

  Soon after the structure of the double helix itself had been established, a key enzyme in the process, called polymerase, was identified. This enzyme built new double helices, and repaired slightly damaged ones, by lining up the links in the chain, the nucleotides, in the right order. It was in action more or less continuously in the growing organism, taking care of DNA replication and repair. Kary Mullis was toying with how he could use the enzyme to establish the sequence of any DNA chain. He played with the molecule in his mind, imagining how a raised temperature would decouple and unpeel the two strands of the double helix. He then thought of lowering the temperature so that the strands once again annealed, but not with each other. They annealed instead to added ‘primers’–very short strings of nucleotide that bound with particular short stretches of the separated DNA strands. These primers attached themselves to either end of the stretch of DNA, one on one strand of the disaggregated helix, one on the other. They acted something like bookends, within which the polymerase enzyme could stack the nucleotides in the right order. Pretty soon he had two stretches of DNA, where before there had been one, but corresponding only to the section bounded by the bookends, the primers. It was a question of controlling the temperature, adding primers, and then polymerase and enough free nucleotides for the enzyme to set to work.

  At this point Mullis’s imagination drifted from the sequencing problem to something else. If it was possible in this way to get from one DNA molecule to two, fairly quickly and outside a living cell, the cycle could be repeated to get from two to four, four to eight, and so on. With the right sort of equipment
, it should be possible to double up again and again, and in the space of an afternoon turn one single DNA molecule into a billion identical molecules. In a lateral thinker’s head on a moonlit California drive, the Polymerase Chain Reaction was born.

  The PCR, as it is known, fortunately transferred well from imagination to reality. It is now a key molecular tool that has transformed all aspects of DNA research, not least the quest for ancient DNA.

  making sense of the fragments: PCR in action

  Svante Pääbo was one of those who quickly picked up on PCR and applied it to ancient DNA. Armed with the new technique, he managed to move forward from simply recognizing the DNA to working with informative sequences. He had by now joined Allan Wilson’s lab in Berkeley and was exploring the possibilities of North American material. A group of archaeologists from Florida State University had been digging in a swampy pond in Brevard County, Central Florida.

  The Windover Pond is a flooded sinkhole, a valuable source of drinking water for humans and animals alike. Humans had used such sinkholes ever since they first spread across the New World, and that is why sinkholes excite the interest of archaeologists. A sinkhole such as this one could contain any number of remnants of people’s visits to quench their thirst, in sediments accumulating over thousands of years. As the archaeologists would discover, it could also contain the human bodies themselves. Beneath the shallow water that remains in the sinkhole lay several metres of peat that had built up during 12,000 years of visits. As the archaeologists dug down into the peat, they found the remains of a middle-aged woman about three metres down. There are quite a few bog bodies in the archaeological record, but one particular thing was striking about her. Within her skull she had what looked like an intact brain.