by Martin Jones
Any set of data from the past sets its own limits on what we can see from past human lives. In the first archaeological books I ever encountered, it seemed that little of those lives appeared, beyond a pair of detached hands fashioning the pot or stone tool that at the time was the focus of attention. Bio-archaeology added greatly to that vista and our focus shifted from the hands to the people, the clothes they wore, the food they prepared, the villages they built and the dwellings they furnished. Once we realized that the surviving fragments of the past extended to DNA, it was as if we archaeologists might enter those dwellings, camera and microscope in hand. In that heady atmosphere, we could easily forget the preconceptions we carried with us into those dwellings, preconceptions arising from what we already thought the archaeological record was like. The ancient DNA spotlight was first trained upon the most familiar bio-archaeological tissue, human remains, and moved from there to animal bones and then to the remains of the best known food plants. So studies of ancient DNA have revealed to us something about farming communities, their principal domesticates, and their journeys and expansions across the world. Other lives that were interwoven with quite different food species, and other things that were going on in the prehistoric world, have remained in the dark. But there is more than DNA lurking amidst the dirt of an archaeological excavation, other molecules to be found and decoded. Several of them are both more abundant and more persistent than the molecule at the heart of life. If they too can reveal information in the kind of detail possible with DNA, then they can take our observations elsewhere, to other facets of the prehistoric world.
This possibility was very much in the mind of Geoff Eglinton when we set up the Ancient Biomolecules Initiative in the early 1990s. At that stage ancient DNA research was enjoying prima donna status. In a molecular drama of many parts, it was the billboard star. True to the spirit of stardom, it could be exciting and disappointing in equal measure, and it could both rise to the headlines and conspicuously flop. Geoff Eglinton knew enough about science to see we could not just work with the star performers. We needed the support cast and the production team. To see for certain what molecular information remained among the dirt and the smells of archaeological deposits, we needed to place ancient DNA in the context of a variety of other ancient molecules that were also known to be around. A few obvious candidates deserved attention.
blood, building blocks and enzymes
The great value of DNA as a source of information lies in its long-chain structure, its encoded sequence of molecular links; but it is not the only long-chain molecule in the living system. The many proteins circulating in the blood, together with those that build our soft and hard tissues and those that form the enzymes that steer the body’s chemical processes, also possess a long-chain form. They are assembled from a sequence of amino acid units whose order is determined by the DNA blueprint. Soon after the structure of the double helix was discovered, molecular biologists began to tackle the problem of how that blueprint worked. Researchers established that the DNA blueprint was read in clusters of three bases. Each triplet of bases, or ‘codons’ as they came to be known, would line up a specific amino acid to add to the protein sequence. This was achieved with the help of the cell’s ‘molecular machine tools’, the contorted molecules of RNA, DNA’s close cousin. These RNA machine tools would allow the DNA blueprint to be read, codon by codon, and a protein to be built incorporating a corresponding sequence of amino acids. So, for example, two guanine bases followed by one adenine base forms the codon that will enable a glycine amino acid unit to be added to the protein chain. As there are four potential bases at each of the three positions of each triplet of bases or ‘codon’, there are sixty-four different messages that can be conveyed to the construction process. This is more than enough to cover the twenty or so amino acids from which proteins are assembled, as well as to code the ‘start’ and ‘stop’ messages needed to define the boundaries of any particular protein-building gene. In other words, the protein sequence preserves a great deal of the information in the DNA sequence, easily translatable from one to the other, but with the potential to reach much further back in time.
The secret of the ancient protein’s survival lies in some close attachments it makes with the mineral component of living organisms. For example, the collagen in bone can survive to be a very ancient protein indeed. At Newcastle, Matthew Collins has spent several years working out how ancient bones are conserved. He examined a 120,000-year-old bone with 80 percent of its original collagen in place. He estimated that in suitably cold conditions some collagen should survive for a million years or more. Before we get too excited about its genetic possibilities, we have to recognize that it is, unfortunately, a rather monotonous molecule. Other proteins could be much more informative.
bloodlines
We would not immediately expect blood proteins to be great molecular survivors; they are lodged in a fluid which is by nature chemically and biologically active. Freshly spilt blood changes colour before our eyes, and offers little resistance to pathways of transformation and decay. Blood nevertheless reaches throughout the body’s tissues, here and there providing certain blood proteins with a more secure haven from breakdown and decay, enclosed within a bone, or even a mummified muscle. What is more, the blood itself offers a highly sensitive system of protein detection that could be mimicked in the lab and thus used to track down proteins that have been reduced to tiny quantities with the passage of time.
This sensitive system within the blood involves ‘antibodies’, themselves a group of proteins whose biological function is to seek out invasive organisms by recognizing alien biomolecules on their own molecular surfaces. This they achieve by attaching a molecular ‘mould’ to a particular part of the invasive biomolecule. These couplings can be extremely precise, in that an ‘antibody’ only attaches itself to an ‘antigen’ that possesses a highly specific molecular arrangement. As our abilities to build artificial biomolecules grew, so the natural antibody/antigen coupling formed the basis for a series of artificial chemical detectives, designed to seek out molecules chosen by the researcher. These techniques of ‘immunology’ made it possible to seek out proteins, and indeed other biomolecules, in tiny quantities, and are among the earliest established techniques in biomolecular archaeology.
One series of blood proteins comprises the natural antibody systems themselves. Several of these systems have a relatively straightforward genetic basis. It was soon realized they could be used to assemble the kind of stories about origins and migrations explored in the preceding chapter. Soon after the First World War, Ludwik and Hanna Hirschfeld, a husband-and-wife team, did just that. They had assembled information on ABO blood groups from allied soldiers on the Macedonian front, and combined it with data they had gleaned from German populations. The A and B types formed a cline spreading from north-west Europe (England) by way of the eastern Mediterranean to India, leading them to suggest that ‘we should look to India for the cradle of one part of humanity’. Fourteen years later, another husband-and-wife team, William and Lyle Boyd, attempted something similar with ancient Egyptians.
Lyle Boyd first had the idea of detecting blood types in ancient mummies in 1933. She had typed blood groups from saliva and dried blood, so why not dried muscle and ancient bodies? Within four years, she and her husband had published their findings from 122 Egyptian mummies, and from a further 205 mummified bodies from America. A striking feature of their results was the repeated finding of blood group B. Among living Native Americans, blood group B is a predominantly northern phenomenon, but the Boyds were finding it in mummified bodies from the south. They were reluctant to challenge the modern evidence. This after all was well before ‘biomolecular archaeology’ was a phrase that meant anything to anyone. By the mid-1970s the Virginia pathologist Marvin Allison could be bolder. He analysed the blood types of 140 South American mummified corpses, which spanned the arrival of Columbus and the subsequent conquistadors. There was a clear contrast bet
ween earlier and later individuals.
The post-Columbian bodies displayed a range of blood types with a high proportion carrying blood group O. That was consistent with the modern distribution. In contrast, the evidence of pre-Columbian bodies mirrored the Boyds’ evidence collected four decades earlier. A, B, and AB blood groups were significantly more abundant in the pre-Columbian bodies. Allison had picked up a small evolutionary episode: some blood groups had fared better than others in the face of diseases that flourished after contact with Europe.
ancient blood: from mummies to mammoths
Around the time that Marvin Allison was re-examining blood groups in mummified humans, Jerold Lowenstein at the University of California was turning his attention to much older mammals. He took the methods a step further by building radioactive markers into his immunological reagents. Once they had sought out their target protein, they would mark it with a beacon of radioactivity. As little as a billionth of a gram of the target protein was sufficient to leave a visible radiation scar on a sheet of photographic film.
By now, several blood proteins had been identified from bones that were a few thousand years old, and it seemed timely to push the analysis back into prehistory. In 1977, a baby mammoth christened ‘Dima’ gave Lowenstein his opportunity. Dima had been exhumed from the Siberian permafrost at Magadan, and had been dead for 40,000 years. Lowenstein designed antibodies to seek out blood albumins, which are important proteins in the colourless part of the blood, and to compare them with the albumins of a range of living mammals. He found the albumins he was looking for in Dima’s thigh. About 1 percent of the original albumins had survived, and they closely matched those from living elephants.
Like the earlier work on the ABO system, Lowenstein’s analyses of blood proteins paralleled and complemented the kind of DNA-based phylogeny that has been discussed in previous chapters of this book. The protein evidence had the limitation of relating to expressed genes only, and provided no guide to the uncoded sequences so valuable in DNA phylogenies. However, it had the advantage of longevity. Dima’s ancient DNA was just inside the limits of feasible detection, but its albumin proteins were more abundant than the quantity needed for RIA detection by a factor of a thousand. Ancient DNA analysis could just about reach into the realm of major mammalian extinctions, whereas ancient protein analysis moved comfortably into this area. Lowenstein and his colleagues studied a number of extinctions directly parallel to those explored through ancient DNA. In addition to Dima and other mammoths, they worked on the bones of a mastodon and an extinct sea cow, the soft tissues of a Tasmanian wolf and a quagga, and even the crystallized urine of an extinct pack rat. In these studies, there was a very clear and reassuring match between the two forms of molecular phylogeny, from the DNA and the protein.
dinosaur protein
Blood proteins may have some potential in the future growth in ancient biomolecule research. This may involve moving away from the plentiful blood proteins, such as haemoglobin and the albumins, to the less plentiful but highly informative proteins such as the immunoglobulins. Beyond the blood, yet other proteins surprise us by their remarkable persistence. One of these is, like collagen, tucked away in the biomineral region that can offer the greatest resistance to decay.
Collagen is the principal bone protein, but it is not the only one. A small protein called osteocalcin is the second most abundant protein in bone after collagen. It is a short molecule, made up of forty-nine amino acids, and among those forty-nine are three of an unusual type known as carboxyglutamic acid, or GLA for short. These three GLA units form a very powerful adhesive patch, which holds on tight to the mineral particles of apatite within bone. The GLA amino acid has been detected in very old bones indeed.
Dinosaur DNA may have come and gone from the scientific literature, but the detection of the GLA amino acid in 75-million-year-old bone brings these giants back into the picture. It provides strong evidence that the osteocalcin GLA patch has retained its adhesive strength, and such tightly bonded proteins have a considerable chance of persistence. There may well be other minor proteins lurking in the safe haven of the bone’s biomineral recesses, carrying key information about the distant past. Ancient DNA has reached just far enough back to give a glimpse of the most recent episodes in hominid evolution. The longevity of ancient proteins is sufficient to embrace hominid evolution in its entirety, and go beyond to much earlier evolutionary stories. We simply await their identification. Such structural integrity has already been found in another range of biomolecules noted for their extreme longevity.
the energy molecules
If DNA carries the code, and proteins do the work, then other major molecules, the energy molecules, are the ones that fuel the whole process. A consideration of this third group takes us to the most ancient biomolecules of all.
Deep in the earth’s crust, our vast reserves of petroleum indicate just how persistent some of these energy molecules can be. Their hydrocarbon backbones persist as petroleum, with many of their chemical side-limbs honed off through millions of years of exposure. In the much more recent deposits of the archaeological record, one of the more striking examples of the persistence of these fatty, oily substances is an unusual soapy substance occasionally found around a buried skeleton. Adipocere, as it is termed, has been recognized since at least the seventeenth century, when a certain Sir Thomas Browne wrote in 1658 of a ‘fat concretion, where the nitre of the Earth, and the salts and lixivious liquor of the body, had coagulated large lumps of fat, into the consistence of the hardest castle-soap’ around the exposed skeleton of an ancient churchyard burial. As is typically the case with ancient biomolecules, adipocere is not there in its original form. The body’s fat has fragmented, at a molecular level, into glycerol, which breaks down quite easily, and fatty acids that, together with alkalis in the soil, form the soapy substance which caught the attention of Browne.
molecular overcoats
Fats are more remote from the genetic blueprints than proteins, and they pass through food in a more intact state. In other words, their variation may not tell us a great deal about the ancient organism we are looking at. Nevertheless, some lipids can tell us a great deal. Take, for example, the shiny waxy surface on the leaf of a plant, its waterproof protection against the elements. Within this waxy coat are lipids that are both highly characteristic of their plant origin and among the most persistent organic molecules of all, albeit in a slightly modified state. These are the cutins, long-chain molecules composed of a sequence of fatty acid units linked together. They are the molecules that endow a leaf cuticle with much of its strength.
At London University, Margaret Collinson and Andrew Scott have been looking at these waxy cuticles on leaves that are up to 300 million years old. Even in these ancient specimens, it is possible to lift off the leaf’s waterproof overcoat, a smooth thin sheet looking remarkably similar to the modern cuticle and retaining a fine cast of the cellular structure beneath. None of the cutin structures survives intact in this deceptively familiar skin. Neither do they break down to the extent of becoming unrecognizable. A small proportion is formed of a series of closely related molecules called cutans. These, as far as we know, can remain intact indefinitely, or at least until the very slow cycles of reworking of the Earth’s crust carry them down into the molten levels below. The DNA from Edward Golenberg’s Miocene Magnolia leaf may have been an illusion, but for some of the lipids in the leaf’s cuticle the Miocene epoch around 20 million years back was merely a transitory episode of their molecular youth.
plant skeletons?
Plants do not have mineral backbones like our own skeleton. For that kind of support they rely on organic molecules such as cellulose and lignin. They are, however, not without their own mineral inclusions, and these can linger on long after the cells around them have decayed. One of the most plentiful minerals within plant tissue is silica, absorbed in solution by the roots and then distributed around the plant’s cells. These tiny particles of plant si
lica can survive in the ash of a fire, or as a residue of decayed plant tissue. They are not easily discernible from routine excavation of layers that appear relatively uniform to the naked eye. However, if small pieces of the excavated profile are set in resin and examined under the microscope, the layers of plant silica within ancient stems and leaves come into view. In the ancient settlement mounds of the Near East, they are especially plentiful. What seems at first sight like a uniform heap of collapsed mud-brick may prove, under the microscope, to be dominated by the ashy remains of plant silica.
These particles of silica, known as ‘phytoliths’, meaning ‘plant stones’, are in archaeological terms the plant parallel to the apatite in bones, the major mineral produced by the living body and the most durable element of it. Phytoliths are sometimes truly skeletal in that they have a strengthening function, sharpening the edges and honing the barbs along the edges of our more vicious grasses. At other times they are simply deposits of surplus dissolved silica, drawn into the plant from the soil water. They can survive longer than any other plant tissue. In Palaeolithic sites they may be the only plant fragments that persist. The Amud Cave in Israel is the latest Neanderthal site in the Levant, and is dated to 50,000 years ago. The cave sediments contain marked concentrates of grass phytoliths, with a significant number coming from the seed head. This may prove to be the earliest evidence of the collecting of grass seed that foreshadowed the development of cereal exploitation in south-west Asia several thousand years later.
Much can be learnt from the shapes of phytoliths. They are sometimes characteristic of particular plant genera, such as rice or maize. Sometimes they characterize particular environments, irrigated and dry-farmed crops differing in their range of phytolith types. Not much has been done on chemical variations within the silica particles, but in due course these variations may have much to reveal. Silica is not the only mineral residue from plant cells. Many crystalline and semi-crystalline substances deposited within the cell retain a characteristic form, and are sometimes extremely durable after death. One example is the layer of phytine crystals within a seed that acts as a trace element bank for the seed’s future germination. Even after the seed has decayed, the microscopic phytine crystals can linger for thousands of years, free within an active soil.
