EVIEWS ANCIENT DNA Michael hofreiter. david serre. hendrik n. poinar. melanie kuch and Syante paabo DNA that has been recovered from archaeological and palaeontological remains makes it possible to go back in time and study the genetic relationships of extinct organisms to their contemporary relatives. This provides a new perspective on the evolution of organisms and DNA sequences. However, the field is fraught with technical pitfalls and needs stringent criteria to ensure the reliability of results, particularly when human remains are studier COPROLITES Since the advent of techniques that allow dNA to be these processes create problems for the retrieval of rapidly sequenced, evolutionary studies that rely on the ancient DNA sequences. For example, a high propor- comparison of DNA sequences from living organisms tion of cytosine and thymine residues in extracts of archaeological excavations. have become commonplace. However, contemporary ancient tissues are oxidatively modified to HYDANTOINS, DNA sequences provide only indirect evidence of the which block DNA polymerases and thus the PCR historical processes that have formed them over long Furthermore, deamination products of cytosine, for pyrimidine bases(cytosine periods of time. So, in a sense, the field of molecular example, are common in ancient DNA and cause evolution istime trapped. The study of DNA from dead incorrect bases to be inserted during the PCR2(FIG.1) organisms offers a partial way out of this predicament. After a long enough time, the cumulative effects of However, many technical pitfalls need to be avoided to damage to the DNa will become so extensive that no allow the molecular evolutionist to go back in time and useful molecules remain. Assuming physiological salt effectively catch evolution red handed. In this review, concentrations, neutral pH and a temperature of 15C, we first discuss dna decay and our current under- it would take about 100,000 years for hydrolytic dam- standing of the limits to DNA preservation, before age to destroy all dna that could reasonably be reviewing several technical issues of crucial importance retrieved. Some environmental conditions, such as for the retrieval of ancient DNA. Finally, we discuss lower temperatures, will extend this time limit, where- insights that have come from DNA sequen other conditions will reduce it. However to the best retrieved from ancient animals and hominids, as well as of our understanding, to consider amplification of from the CoPRoLITEs they left behind. DNA molecules older than one million years of age is Post-mortem DNA decay When an organism dies, its DNA normally becomes Ancient DNA retrieval degraded by endogenous nucleases. Under fortunate For specimens of plants and animals that have been col circumstances, such as rapid desiccation, low tempera- lected under controlled conditions and stored in muse- tures or high salt concentrations, nucleases can ums in the past 200 years, the PCR technique has made themselves become destroyed or inactivated before all the retrieval of DNA sequences almost routine. Starting nucleic acids are reduced to mononucleotides. If this is with the study of kangaroo rats in the mojave desert the case, slower but still relentless processes start affect- that were collected at the beginning of the century ng the DNA(FIG. 1). For example, oxidation, as well as many studies have used museum collections to study the direct and indirect effects of background radiation, populations over time, as well as the phylogenetic rela- Inselstrasse 22, D-04103 will modify the nitrous bases and the sugar-phosphate tionships of recently extinct species-12. However, for backbone of the DNA. Furthermore, deamination, archaeological and palaeontological specimens, the depurination and other hydrolyte sses will lead retrieval of ancient DNA sequences is far from routine, mail:; paabo@eva. mpg.de to destabilization and breaks in DNA molecules. All as the researcher has to contend with the fact that very NATURE REVIEWS GENETICS VOLUME 2 MAY 2001 353 Mo 2001 Macmillan Magazines Ltd
Since the advent of techniques that allow DNA to be rapidly sequenced, evolutionary studies that rely on the comparison of DNA sequences from living organisms have become commonplace. However, contemporary DNA sequences provide only indirect evidence of the historical processes that have formed them over long periods of time. So, in a sense, the field of molecular evolution is ‘time trapped’. The study of DNA from dead organisms offers a partial way out of this predicament. However, many technical pitfalls need to be avoided to allow the molecular evolutionist to go back in time and effectively ‘catch evolution red handed’. In this review, we first discuss DNA decay and our current understanding of the limits to DNA preservation, before reviewing several technical issues of crucial importance for the retrieval of ancient DNA. Finally, we discuss some insights that have come from DNA sequences retrieved from ancient animals and hominids, as well as from the COPROLITES they left behind. Post-mortem DNA decay When an organism dies, its DNA normally becomes degraded by endogenous nucleases. Under fortunate circumstances, such as rapid desiccation, low temperatures or high salt concentrations, nucleases can themselves become destroyed or inactivated before all nucleic acids are reduced to mononucleotides. If this is the case, slower but still relentless processes start affecting the DNA (FIG. 1). For example, oxidation, as well as the direct and indirect effects of background radiation, will modify the nitrous bases and the sugar-phosphate backbone of the DNA. Furthermore, deamination, depurination and other hydrolytic processes will lead to destabilization and breaks in DNA molecules. All these processes create problems for the retrieval of ancient DNA sequences. For example, a high proportion of cytosine and thymine residues in extracts of ancient tissues are oxidatively modified to HYDANTOINS, which block DNA polymerases and thus the PCR1 . Furthermore, deamination products of cytosine, for example, are common in ancient DNA and cause incorrect bases to be inserted during the PCR2 (FIG. 1). After a long enough time, the cumulative effects of damage to the DNA will become so extensive that no useful molecules remain. Assuming physiological salt concentrations, neutral pH and a temperature of 15 °C, it would take about 100,000 years for hydrolytic damage to destroy all DNA that could reasonably be retrieved3,4. Some environmental conditions, such as lower temperatures5 , will extend this time limit, whereas other conditions will reduce it. However, to the best of our understanding, to consider amplification of DNA molecules older than one million years of age is overly optimistic. Ancient DNA retrieval For specimens of plants and animals that have been collected under controlled conditions and stored in museums in the past 200 years, the PCR technique has made the retrieval of DNA sequences almost routine. Starting with the study of kangaroo rats in the Mojave desert that were collected at the beginning of the century6 , many studies have used museum collections to study populations over time7,8, as well as the phylogenetic relationships of recently extinct species9–12. However, for archaeological and palaeontological specimens, the retrieval of ancient DNA sequences is far from routine, as the researcher has to contend with the fact that very ANCIENT DNA Michael Hofreiter, David Serre, Hendrik N. Poinar, Melanie Kuch and Svante Pääbo DNA that has been recovered from archaeological and palaeontological remains makes it possible to go back in time and study the genetic relationships of extinct organisms to their contemporary relatives. This provides a new perspective on the evolution of organisms and DNA sequences. However, the field is fraught with technical pitfalls and needs stringent criteria to ensure the reliability of results, particularly when human remains are studied. COPROLITES Faecal material from humans and animals found at archaeological excavations. HYDANTOINS Oxidation products of the pyrimidine bases (cytosine and thymine). NATURE REVIEWS | GENETICS VOLUME 2 | MAY 2001 | 353 Max Planck Institute for Evolutionary Anthropology, Inselstrasse 22, D-04103 Leipzig, Germany. Correspondence to S.P. e-mail: paabo@eva.mpg.de REVIEWS © 2001 Macmillan Magazines Ltd
REVIEWS plasts in plants. Only recently have the first short nuclear DNA sequences that are likely to be genuine been determined from pleiStocene animals. even if some earlier reports of nuclear sequences from human remains exist. So, out of necessity, there is an almost complete reliance on mitochondrial dNA o=P-0 (mtDNA) sequences for phylogenetic studies of extinct animals. For species that are not very closely related to each other, this is not problema hen enough time has passed between speciation events all parts of the genome are expected to show NA the same phylogeny. However, when closely related species, or population genetic questions, are studied, damage depurination) 0=P-0 it is important to remember that the mtDNA repre sents a single genetic locus that might or might not reflect the overall history of the genome. Caution in Site of hydrolytic the interpretations of the results is therefore necessary gure 1 DNA damage shown or likely to affect ancient Assessment of preservation. Only a small proportion of NA. A short segment of one strand of the DNA double helix old specimens that have been analysed contain DNA shown with the four common bases. Principal sites of damage are indicated by red arrows. Sites susceptible to that can be amplified by PCR. Therefore, rapid screen- hydrolytic attack are indicated by green arrows and those methods are crucial to identify the large fraction of prone to oxidative damage by blue arrows. G, guanine: C, badly preserved that there is no cytosine: T, thymine: A, adenine (Modified with permission eed to attempt DNA extractions. In particular, from REF4e(1993)Macmillan Magazines Ltd) amino-acid analyses that take into account the total amount of amino acids preserved in a specimen, the amino-acid composition and the extent of RACEMIZATION little and often no DNA survives in ancient tissues, of several amino acids, have in our hands proved to be whereas contemporary dna pervasive in the envi- a very useful proxy for DNA preservation".Whenever PLEISTOCENE The geological time period ronment, both inside and outside the laboratory. any of these parameters exceed certain levels, DNA from two million to 10.000 Therefore, great precautions need to be taken to avoid extractions have proved futile. In addition, amino-acid the presence of extraneous DNA in the PCR For exam- analyses can support the authenticity of dnA ple, the extraction and preparation of the PCr must be sequences retrieved from a specimen by showing that RACEMIZATION lone in a laboratory that is rigorously separated from preservation conditions have been such that retrieval The change in the three. work involving modern DNA. Treatment of the labora- of macromolecules is conceivable(BOX 1). Another acids from one form toamirror tory equipment with bleach, UV irradiation of the method that might serve the same purpose, but that entire facility, protective clothing and face shields are has not yet been extensively shown to correlate empiri- other routine precautions. In addition, several crite- cally with the presence or absence of authentic ancient PYROLYSIS GC/MS An analysis in which macromolecules are sequences are ancient(BOX D). Despite these precautions, chromatography and mass spectrometry(GCIMS)/ ria of authenticity are essential for believing that DNA sequences, is PYROLYSIS coupled contamination is a continuous threat, if for no other reason than because the specimens themselves might be Quantitation Estimations of the number of ancient chromatography followed by contaminated with modern dna template DNA molecules from which the PCr starts are The greatest difficulties are encountered in the another useful way to ensure the quality of old dNA study of human remains. The reason is that human sequences. If amplifications start from very few or even A class of repetitive DNA that DNA is particularly prevalent in the environment of single DNA strands, then errors in the first cycles will eats that are laboratories and museums, and cannot easily be dis- become incorporated into all molecules in the final PCR nucleotides in length. They tinguished from the dna endogenous to ancient products, Such errors can be numerous if the amount of human remains. In fact, because human dna damage in the DNA is large. Furthermore, if amplifica molecular markers in sequences can readily be retrieved from ancient ani- tions are carried out from, for example,a MICROSATELLITE population genetics studies. mal remains, we believe that many published studies locus on an autosome, then one of the two alleles will that report ancient human DNa sequences are unreli- fail to be amplified if the amplification starts from a sin- OMPENTVE AND REAL TIME able(BOX 2). Ancient animals pose less of a problem as gle molecule, causing the individual to be erroneously APCR analy he sequences retrieved are often testimony to their ed as a homozygote. Finally, contamination can be own authenticity if they are distinct from, yet related extremely hard to exclude if the genuine template is s that initiate the to, extant species in the same taxonomic group rare that amplification is possible only in some of the centration o L. Multicopy DNA sequences are more likely to sur- reactions. Therefore, a determination of the number of in ancient specimens, probably because they template molecules from which a PCR starts is extreme- on of product occur in larger numbers per cell. Therefore, most ly useful and can be achieved by using CoMPETTTIVEPCR, ancient DNA sequences that have been retrieved are and more recently, by using REAL-TIME QUANTTTATIVEPCR2If 354 MAY 2001 VOLUME 2 Nnature. com/reviews/genetics Mo 2001 Macmillan Magazines Ltd
354 | MAY 2001 | VOLUME 2 www.nature.com/reviews/genetics REVIEWS from mitochondria in mammals and from chloroplasts in plants. Only recently have the first short nuclear DNA sequences that are likely to be genuine been determined from PLEISTOCENE animals15, even if some earlier reports of nuclear sequences from human remains exist16. So, out of necessity, there is an almost complete reliance on mitochondrial DNA (mtDNA) sequences for phylogenetic studies of extinct animals. For species that are not very closely related to each other, this is not problematic because when enough time has passed between speciation events all parts of the genome are expected to show the same phylogeny. However, when closely related species, or population genetic questions, are studied, it is important to remember that the mtDNA represents a single genetic locus that might or might not reflect the overall history of the genome. Caution in the interpretations of the results is therefore necessary. Assessment of preservation. Only a small proportion of old specimens that have been analysed contain DNA that can be amplified by PCR. Therefore, rapid screening methods are crucial to identify the large fraction of samples that are so badly preserved that there is no need to attempt DNA extractions. In particular, amino-acid analyses that take into account the total amount of amino acids preserved in a specimen, the amino-acid composition and the extent of RACEMIZATION of several amino acids, have in our hands proved to be a very useful proxy for DNA preservation17. Whenever any of these parameters exceed certain levels, DNA extractions have proved futile. In addition, amino-acid analyses can support the authenticity of DNA sequences retrieved from a specimen by showing that preservation conditions have been such that retrieval of macromolecules is conceivable (BOX 1). Another method that might serve the same purpose, but that has not yet been extensively shown to correlate empirically with the presence or absence of authentic ancient DNA sequences, is PYROLYSIS coupled with gas chromatography and mass spectrometry (GC/MS)18. Quantitation. Estimations of the number of ancient template DNA molecules from which the PCR starts are another useful way to ensure the quality of old DNA sequences. If amplifications start from very few or even single DNA strands, then errors in the first cycles will become incorporated into all molecules in the final PCR products. Such errors can be numerous if the amount of damage in the DNA is large. Furthermore, if amplifications are carried out from, for example, a MICROSATELLITE locus on an autosome, then one of the two alleles will fail to be amplified if the amplification starts from a single molecule, causing the individual to be erroneously typed as a homozygote. Finally, contamination can be extremely hard to exclude if the genuine template is so rare that amplification is possible only in some of the reactions. Therefore, a determination of the number of template molecules from which a PCR starts is extremely useful and can be achieved by using COMPETITIVE PCR19, and more recently, by using REAL-TIME QUANTITATIVE PCR20. If little and often no DNA survives in ancient tissues, whereas contemporary DNA is pervasive in the environment, both inside and outside the laboratory. Therefore, great precautions need to be taken to avoid the presence of extraneous DNA in the PCR. For example, the extraction and preparation of the PCR must be done in a laboratory that is rigorously separated from work involving modern DNA. Treatment of the laboratory equipment with bleach, UV irradiation of the entire facility, protective clothing and face shields are other routine precautions13,14. In addition, several criteria of authenticity are essential for believing that DNA sequences are ancient (BOX 1). Despite these precautions, contamination is a continuous threat, if for no other reason than because the specimens themselves might be contaminated with modern DNA. The greatest difficulties are encountered in the study of human remains. The reason is that human DNA is particularly prevalent in the environment of laboratories and museums, and cannot easily be distinguished from the DNA endogenous to ancient human remains. In fact, because human DNA sequences can readily be retrieved from ancient animal remains, we believe that many published studies that report ancient human DNA sequences are unreliable (BOX 2). Ancient animals pose less of a problem as the sequences retrieved are often testimony to their own authenticity if they are distinct from, yet related to, extant species in the same taxonomic group. Multicopy DNA sequences are more likely to survive in ancient specimens, probably because they occur in larger numbers per cell. Therefore, most ancient DNA sequences that have been retrieved are PLEISTOCENE The geological time period from two million to 10,000 years ago. RACEMIZATION The change in the threedimensional structure of amino acids from one form to a mirror image over time. PYROLYSIS GC/MS An analysis in which macromolecules are decomposed by heat, and the products analysed by gas chromatography followed by mass spectrometry. MICROSATELLITES A class of repetitive DNA that is made up of repeats that are 2–8 nucleotides in length. They can be highly polymorphic and are frequently used as molecular markers in population genetics studies. COMPETITIVE AND REAL-TIME QUANTITATIVE PCR A PCR analysis in which the approximate number of molecules that initiate the reaction is measured either by competition with a template of known concentration or by accumulation of product throughout the PCR. O O P O– O O H N O O P O– O O H N N NH2 O O O P O– O O H HN N O O O O P O– O O H N O N N N NH N N NH2 CH3 O NH2 G C T A Principal site of damage (depurination) Site of oxidative damage Site of hydrolytic damage Figure 1 | DNA damage shown or likely to affect ancient DNA. A short segment of one strand of the DNA double helix is shown with the four common bases. Principal sites of damage are indicated by red arrows. Sites susceptible to hydrolytic attack are indicated by green arrows and those prone to oxidative damage by blue arrows. G, guanine; C, cytosine; T, thymine; A, adenine. (Modified with permission from REF. 4 © (1993) Macmillan Magazines Ltd.) © 2001 Macmillan Magazines Ltd
REVIEWS confined to the past one million years, and more BoX 1 Authenticity criteria to determine ancient DNA sequences probably to the past 100,000 years Biochemical assay for macromolecularpreservation Amino-acid analysis should show that the state of preservation of the specimen is Applications of ancient DNA retrieval compatible with DNA preservation. Extinct animals. For the retrieval of DNA sequences Extraction controls and PCr controls from non-human remains that are up to 50,000 to per Mock extractions and PCRs without template dna are elementary controls that haps 100,000 years old, the past decade has seen much should be carried out. progress (FIG 3). Several studies have revealed the phylo- enetic relationships of extinct animals, generally on the Inverse correlation between am basis of mtDNA. For example, the mtDNA of the marsu- Generally, amplification of only short DNA pieces is possible ial wolf of australia has been shown to be related to other Australian marsupials rather than to carnivorous If the number of dNa molecules that initiate the pcr is less than -1.000. at least three marsupials in South America. This means that several of independent amplifications need to be analysed, and the products need to be cloned the morphological features that marsupial wolves share and several clones should be sequenced. with South-American marsupial carnivores have evolved Excusion ofnuclear insertions ofmitochondrialdna independently on the two continents. 0 2. By contrast, Pieces of mitochondrial DNA (mt DNA) exist in the nuclear genome and represent a the mtdNa of the extinct moas of new zealand ha potential source of false results. Sometimes they can be detected by the observation of been shown to be related to flightless birds in Australia closely related species. If overlapping amplifications using different primers detect the indicates that New Zealand was colonized twice by same sequence over an area that is highly variable it is unlikely that the sequence derives from a nuclear insertion 8,32 mtDNA from ancient remains has been taken to a new Amplification from a second extract level through the determination of the entire mtDNA The reproduction of the results from a second, independent extract should show that genomes of-16, 500 base pairs from two moa species2. the result is reliable It is even becoming possible to start addressing population genetic questions in the Late Pleistocene. the time before and during the last Ice Age. So, the When a new or unexpected result is obtained, reproduction in a second laboratory, preferably from a sample that has been independently sent from the museum to the mtDNA types of Alaskan brown bears, which are today found in distinct geographical regions, have second laboratory, ensures that a laboratory-specific contamination is not the source been shown to have coexisted in the same region determined. -30,000 years ago2. This has consequences for the received wisdom among molecular conservation a PCR starts from 1,000 or more molecules, then an geneticists who often regard the sharing of related nucleotide misincorporations do not influence the of long-term population separation, defining'conser- results9. When fewer molecules initiate a reaction, sev- vation units that need to be managed as separate eral repetitions become necessary. In addition, the groups. The fact that the bear mtDNA types loning of the PCR product and sequencing of several occurred together in one and the same area before the clones will then be essential to detect problems(FIG. 2). last Ice Age shows that much of the current distribu or example, if several different sequences occur in a tion of mtdNa types might be due to more recent mitochondrial PCR product it indicates the presence of phenomena, such as the random loss of mtDNA lin contamination or nucleotide misincorporations in early eages in small populations during the last glacial max- cycles of the PCR. imum. So, mtDNA might not always be a good metric for what constitutes a conservation unit DIAGENESIS all physical, chemical Antediluvian DNA. DNA sequences that are older In another recent project, 191 contemporary horses d biological change than about one million years were termed'antedilu- were studied and were found to be very diverse with dergone by any material vian in an insightful and timely correspondence to respect to mtDNA sequences. The same mtDNA from the time of its initial deposition in the environment either proven impossible to reproduce or have been were 12,000-28, 000 years old and preserved in Alaskan CRETACEOUS shown to derive from an identifiable source of conta- PERMAFROST, as well as from eight horses from Estonia mination. So, the DNA sequences from Miocene and Sweden that were 1,000-2,000 years old. Some of eriod from 144 to 65 (5-24 million years before present) inclusions in the mtDNA sequences of the Pleistocene horses, as well million years ago. amber cannot be reproduced, and such inclusions as all of the mtDNA sequences from horses from have been shown to be highly modified by DIAGENETIc archaeological sites in Northern Europe, were found to rocesses.DNA sequences from Miocene plant fall within the variation of modern horses. So, the compression fossils have similarly been impossible to mtDNA sequence diversity is neither a result of an time point(synchronical). reproduce and DNA sequences retrieved from a cRE- acceleration of the evolutionary rate of horse mtDNA TACEous dinosaur bone have been shown to be derived nor of a very recent introduction of wild horse mtDNAs from an insertion of mtDNA in the human into the domestic gene pool. Instead, much of the oil that never thaws in genome.So, barring some unimaginable technical mtDNA of wild horses has entered modern horses early subarctic regions advance, DIACHRONICAL studies of DNA sequences are during domestication NATURE REVIEWS GENETIC VOLUME 2 MAY 2001E Mo 2001 Macmillan Magazines Ltd
NATURE REVIEWS | GENETICS VOLUME 2 | MAY 2001 | 355 REVIEWS confined to the past one million years, and more probably to the past 100,000 years. Applications of ancient DNA retrieval Extinct animals. For the retrieval of DNA sequences from non-human remains that are up to 50,000 to perhaps 100,000 years old, the past decade has seen much progress (FIG. 3). Several studies have revealed the phylogenetic relationships of extinct animals, generally on the basis of mtDNA. For example, the mtDNA of the marsupial wolf of Australia has been shown to be related to other Australian marsupials rather than to carnivorous marsupials in South America. This means that several of the morphological features that marsupial wolves share with South-American marsupial carnivores have evolved independently on the two continents9,10,12. By contrast, the mtDNA of the extinct moas of New Zealand have been shown to be related to flightless birds in Australia rather than to the extant kiwis in New Zealand27. This indicates that New Zealand was colonized twice by flightless ostrich-like birds. Recently, the retrieval of mtDNA from ancient remains has been taken to a new level through the determination of the entire mtDNA genomes of ~16,500 base pairs from two moa species28. It is even becoming possible to start addressing population genetic questions in the Late Pleistocene, the time before and during the last Ice Age. So, the mtDNA types of Alaskan brown bears, which are today found in distinct geographical regions, have been shown to have coexisted in the same region ~30,000 years ago29. This has consequences for the received wisdom among molecular conservation geneticists who often regard the sharing of related mtDNA sequences in a group of animals as indicative of long-term population separation, defining ‘conservation units’ that need to be managed as separate groups30. The fact that the bear mtDNA types occurred together in one and the same area before the last Ice Age shows that much of the current distribution of mtDNA types might be due to more recent phenomena, such as the random loss of mtDNA lineages in small populations during the last glacial maximum. So, mtDNA might not always be a good metric for what constitutes a ‘conservation unit’. In another recent project, 191 contemporary horses were studied and were found to be very diverse with respect to mtDNA sequences. The same mtDNA sequences were then determined from eight horses that were 12,000–28,000 years old and preserved in Alaskan PERMAFROST, as well as from eight horses from Estonia and Sweden that were 1,000–2,000 years old. Some of the mtDNA sequences of the Pleistocene horses, as well as all of the mtDNA sequences from horses from archaeological sites in Northern Europe, were found to fall within the variation of modern horses31. So, the mtDNA sequence diversity is neither a result of an acceleration of the evolutionary rate of horse mtDNA nor of a very recent introduction of wild horse mtDNAs into the domestic gene pool. Instead, much of the mtDNA of wild horses has entered modern horses early during domestication. a PCR starts from 1,000 or more molecules, then an experiment does not need to be repeated to verify that nucleotide misincorporations do not influence the results19. When fewer molecules initiate a reaction, several repetitions become necessary. In addition, the cloning of the PCR product and sequencing of several clones will then be essential to detect problems (FIG. 2). For example, if several different sequences occur in a mitochondrial PCR product it indicates the presence of contamination or nucleotide misincorporations in early cycles of the PCR. Antediluvian DNA. DNA sequences that are older than about one million years were termed ‘antediluvian’ in an insightful and timely correspondence to Nature in 1993 (REF. 21). All such sequences have since either proven impossible to reproduce or have been shown to derive from an identifiable source of contamination. So, the DNA sequences from Miocene (5–24 million years before present) inclusions in amber cannot be reproduced22,23, and such inclusions have been shown to be highly modified by DIAGENETIC processes24. DNA sequences from Miocene plant compression fossils have similarly been impossible to reproduce25 and DNA sequences retrieved from a CRETACEOUS dinosaur bone have been shown to be derived from an insertion of mtDNA in the human genome26. So, barring some unimaginable technical advance, DIACHRONICAL studies of DNA sequences are DIAGENESIS All physical, chemical and biological changes undergone by any material from the time of its initial deposition in the environment. CRETACEOUS The geological time period from 144 to 65 million years ago. DIACHRONICAL A continuous process over time rather than a process at one time point (synchronical). PERMAFROST A layer below the surface soil that never thaws in subarctic regions. Box 1 | Authenticity criteria to determine ancient DNA sequences Biochemical assay for macromolecular preservation Amino-acid analysis should show that the state of preservation of the specimen is compatible with DNA preservation. Extraction controls and PCR controls Mock extractions and PCRs without template DNA are elementary controls that should be carried out. Inverse correlation between amplicon length and amplification efficiency Generally, amplification of only short DNA pieces is possible. Quantitation of numbers of template molecules If the number of DNA molecules that initiate the PCR is less than ~1,000, at least three independent amplifications need to be analysed, and the products need to be cloned and several clones should be sequenced. Exclusion of nuclear insertions of mitochondrial DNA Pieces of mitochondrial DNA (mtDNA) exist in the nuclear genome and represent a potential source of false results. Sometimes they can be detected by the observation of several mtDNA sequences from an extract, or by finding the same sequence in another closely related species. If overlapping amplifications using different primers detect the same sequence over an area that is highly variable, it is unlikely that the sequence derives from a nuclear insertion18,32. Amplification from a second extract The reproduction of the results from a second, independent extract should show that the result is reliable. Reproduction in a second laboratory When a new or unexpected result is obtained, reproduction in a second laboratory, preferably from a sample that has been independently sent from the museum to the second laboratory, ensures that a laboratory-specific contamination is not the source of the sequence determined. © 2001 Macmillan Magazines Ltd
REVIEWS Consensus TACATATTATGCTTGGCCTTACATGAGGACCTACATTTTGAAAGTTTATCTCAAGTGTATAG the cave bears are found to have diverged from the brown bears long before divergence of mtDNA sequences in the two species"2. The diversity among cave bears was less than among modern brown bears, which Clone s is probably related to the fact that they inhabited a These and other studies in progress show that we can now begin to reconstruct the population history of cTTGCGaGt extinct animals, going back in time to well before the last Ice Age. The permafrost in Siberia and North America will certainly prove a particularly rich archive of genetic variation, because preservation of DNA is clearly better in cold environments. 8, and because permafrost deposits contain huge numbers of preserved individuals Neanderthals. NEANDERThALs were a hominid form Figure 2 Alignment of eight mitochondrial DNA clones sequenced from a single more closely related to contemporary humans than to amplification from a 26,000-year-old cave-bear bone. The consensus sequence was chimpanzees-that inhabited Europe and Western confirmed by sequencing additional amplification products. The high prevalence of C/G- T/A Asia until shortly after 30,000 years ago. When a changes is probably due to deamination of C residues in the ancient DNA templates. Note that mtDNA sequence was determined from the direct sequencing of the PCR product would lead to ambiguous results at least at two positions (arrows)(M H. et al, unpublished results. Neanderthal-type specimen found in 1856 in the Neander Valley in western Germany, it proved not to b directly related to the mtDNA of modern Europeans". Instead, the lineage leading to the Neanderthal mtDNA In another population study, DNA sequences from diverged,000 years ago, whereas the common cave bears-extinct bears that existed in Europe and ancestor of the mtDNAs of all living humans lived morphologically distinct from Western Asia until-10,000 years ago-have beenstud- about 170,000 years ago. Even if the absolute dates existed until-30,000 years ago ied from several caves in Europe. When their mtDNAs change if more is learnt about the rates and mode of in Europe and western Asia mpared to those the mtDNA. this shows that the Box 2 Ancient human DNA sequence Remains of late pleistocene humans have been found in Upper Cave, Zhokouchien, China. Two well-preserved, to amino-acid analysis. The extent of racemization of aspartic acid was 0. 11 and 0.10, respectively; that is, ratios below the value of, for example, the Neanderthal- type specimen(0.1)from which endogenous DNA has ←100b been amplified. Part of each tooth was ground to powder and extracted, following published protocols. mock extraction was processed alongside the samples NA was amplified using primers for the mitochondrial DNA(mt DNA)16SrRNA gene, as described. Both teeth yielded amplification products(lanes 3 and 4), whereas the mock extract and PCR controls were negative (lanes l and 2). A second extraction was analysed from each tooth with the same result (lanes 5-8). The amplification products were sequenced and shown to be human. Next, a piece of hypervariable region I of the mtDNA control region was amplified twice from one of the teeth(using primers H16209 and L16303) Direct sequencing, using the primer H16209, revealed reproducible 93-base-pair(bp)sequence that differed at one position from over 6,000 contemporary mt DNA sequ at present in the database when the experiment was repeated, the same sequence was obtained. At this point, most studies of human remains would be published, often in a prestigious journal, making the claim that these DNA sequences stem from the human teeth(for example, REFS 36 51-55). However, when the two control region amplification products were cloned and 20 clones from each were sequenced, a total of 20 different human sequences were found among the clones. This result is surprising in view of he fact that only one mt DNA sequence would be expected from a tooth. a further fact that proves that all the above results are due to contamination-the teeth that were analysed came from cave bears found in Upper Cave rather than from humans! We facetiously neglected to indicate this above to illustrate the fact that had these been human teeth, failure to adhere to all the criteria in BoX I would undoubtedly result in the publication of false results. Sadly the problem of contamination, as well as the criteria in BOX 1, is often ignored in published work on ancient human DNA sequences. It is particularly noteworthy that direct sequencing in this case failed to indicate the complex mixture of sequences present. 356 MAY 2001 VOLUME 2 Nnature. com/reviews/genetics Mo 2001 Macmillan Magazines Ltd
356 | MAY 2001 | VOLUME 2 www.nature.com/reviews/genetics REVIEWS the cave bears are found to have diverged from the brown bears long before divergence of mtDNA sequences in the two species32. The diversity among cave bears was less than among modern brown bears, which is probably related to the fact that they inhabited a smaller area than their contemporary relatives. These and other studies in progress show that we can now begin to reconstruct the population history of extinct animals, going back in time to well before the last Ice Age. The permafrost in Siberia and North America will certainly prove a particularly rich archive of genetic variation, because preservation of DNA is clearly better in cold environments5,17,18, and because permafrost deposits contain huge numbers of preserved individuals. Neanderthals. NEANDERTHALS were a hominid form — more closely related to contemporary humans than to chimpanzees — that inhabited Europe and Western Asia until shortly after 30,000 years ago. When a mtDNA sequence was determined from the Neanderthal-type specimen found in 1856 in the Neander Valley in western Germany, it proved not to be directly related to the mtDNA of modern Europeans33. Instead, the lineage leading to the Neanderthal mtDNA diverged ~500,000 years ago34, whereas the common ancestor of the mtDNAs of all living humans lived about 170,000 years ago35. Even if the absolute dates change if more is learnt about the rates and mode of evolution of the mtDNA, this shows that the In another population study, DNA sequences from cave bears — extinct bears that existed in Europe and Western Asia until ~10,000 years ago — have been studied from several caves in Europe. When their mtDNAs are compared to those of contemporary brown bears, Consensus TACATATTATGCTTGGCCTTACATGAGGACCTACATTTTGAAAGTTTATCTCAAGTGTATAG Clone 1 ................T....T........................................ Clone 2 .............................................................. Clone 3 .............................................................. Clone 4 .............................................................. Clone 5 ................T....T........................................ Clone 6 ................T....T........................................ Clone 7 ......................G....................................... Clone 8 ................T....T........................................ TCTGTAAGCATGTATTTCACTTAGTCCGGGAGCTTAATCACCAGGCCTCGAGAAACCAGCAACCCTTGCGAGT ......................................................................... ......................................................................... ...........A...............AAA........................................... ......................................................................... ...................................................A......A.............. ......................................................................... ......................................................................... ........................................T................................ Figure 2 | Alignment of eight mitochondrial DNA clones sequenced from a single amplification from a 26,000-year-old cave-bear bone. The consensus sequence was confirmed by sequencing additional amplification products. The high prevalence of C/G → T/A changes is probably due to deamination of C residues in the ancient DNA templates. Note that direct sequencing of the PCR product would lead to ambiguous results at least at two positions (arrows) (M.H. et al., unpublished results). NEANDERTHAL A hominid form, morphologically distinct from contemporary humans, that existed until ~30,000 years ago in Europe and western Asia. Box 2 | Ancient human DNA sequences? Remains of Late Pleistocene humans have been found in Upper Cave, Zhokouchien, China50. Two well-preserved, ~30,000-year-old teeth from this site were subjected to amino-acid analysis. The extent of racemization of aspartic acid was 0.11 and 0.10, respectively; that is, ratios below the value of, for example, the Neanderthaltype specimen (0.12) from which endogenous DNA has been amplified33. Part of each tooth was ground to powder and extracted, following published protocols33. A mock extraction was processed alongside the samples. DNA was amplified using primers for the mitochondrial DNA (mtDNA) 16S rRNA gene, as described65. Both teeth yielded amplification products (lanes 3 and 4), whereas the mock extract and PCR controls were negative (lanes 1 and 2). A second extraction was analysed from each tooth with the same result (lanes 5–8). The amplification products were sequenced and shown to be human. Next, a piece of hypervariable region I of the mtDNA control region was amplified twice from one of the teeth (using primers H16209 and L16303). Direct sequencing, using the primer H16209, revealed a reproducible 93-base-pair (bp) sequence that differed at one position from over 6,000 contemporary mtDNA sequences at present in the database. When the experiment was repeated, the same sequence was obtained. At this point, most studies of human remains would be published, often in a prestigious journal, making the claim that these DNA sequences stem from the human teeth (for example, REFS 36,51–55). However, when the two control region amplification products were cloned and 20 clones from each were sequenced, a total of 20 different human sequences were found among the clones. This result is surprising in view of the fact that only one mtDNA sequence would be expected from a tooth. A further fact that proves that all the above results are due to contamination — the teeth that were analysed came from cave bears found in Upper Cave rather than from humans! We facetiously neglected to indicate this above to illustrate the fact that had these been human teeth, failure to adhere to all the criteria in BOX 1 would undoubtedly result in the publication of false results. Sadly, the problem of contamination, as well as the criteria in BOX 1, is often ignored in published work on ancient human DNA sequences. It is particularly noteworthy that direct sequencing in this case failed to indicate the complex mixture of sequences present. 200 bp 100 bp 1 8 2 3 4 5 6 7 © 2001 Macmillan Magazines Ltd
REVIEWS due to contamination(s)(BOX 1). This might be a pa ticular concern in this case, because the mtDNA in ques eA has previously been shown to represent a contamination in ancient DNA studies2. So, further work, and perhaps even new technical approaches, are necessary before any definitive statements about the mtDNA gene pool of early modern humans can be made In principle, another way to estimate the extent of any putative mixing between modern humans and Neanderthals would be to study nuclear genes from the Neanderthals themselves. However, even if we ignore the technical problems involved with this, it would not be very informative as we already know that many nuclear genes have coALESCENCEs that pre-date the pr tive separation of Neanderthals and anatomically mod ern humans-500,000 years ago. We can therefore rea- sonably assume that the amount of variation at most nuclear loci in Neanderthals falls within the amount of variation found among humans today ?. So, the overall genetic relationship of Neanderthals and modern humans was not very distant and might in general have resembled that of subspecies of present-day apes more y将 than that of different species. However, the diver- gence of the mtDNA pools indicates a temporal separa chromosomal sequences, should this become possibl in the future The recent determination of mt DNA sequences of a Figure 3 Some extinct organisms from which DNA sequences have been Neanderthal from the Caucasus(-30,000 years olds) determined. From bottom left to top right: quagga, marsupial wolf 0, 2, sabre-toothed and of another from Croatia(over 42,000 years old sland piopio, Stellers sea cow l (FIG 4) points to the exciting possibility of studying Neanderthal33.3 L39, 0, Aptornis defossor8 Shasta ground sloth 68, pig-footed bandicoot. aspects of the population history of this archaic group Myotragus balearics of hominids. For example, humans differ from the great apes in having much less DNA sequence variation both in mtdNA and in nuclear dnas furthermore the dis- Neanderthals went extinct without contributing tribution of mt DNA sequence variation in contempo mtDNA to contemporary humans. So, the view of rary humans indicates that they expanded from a small- modern human ori at assumes a recent African er population-50,000 years ago, whereas there is little origin, and little genetic contribution to the current or no evidence for an expansion in the great apes. It gene pool from archaic humans elsewhere is correct for would be extremely interesting to investigate mtDNA mtDNA. It does not, of course, exclude the fact that variation among Neanderthals to address the questions mixing between modern humans, which arrived in of whether they were similar to humans in having much Europe from Africa -40,000 years ago, and variation and to obtain an indication of whether an Neanderthals took place, but there is no evidence of such amixture from the molecular studies lf it becomes ssible in the future to study mtDNAs from many early modern humans. the occurrence of Neanderthal Neanderthal Caucasus(Mezmaiskaya mtDNA types among them would prove that mixing did occur. Recently, a report of mtDNA sequences from Germany(Neanderthal skeletons of early modern humans from Australia revealed a sequence that tends to fall outside the varia- Africans on of modern humans. This could represent a variant non-Africans in modern humans at the time, or alf natively a mtDNA contribution of archaic humans to Modern human the modern human gene pool that was subsequently lost Africans by drift. However, contamination with contemporary he joining of genetic lineages human DNA makes the study of human remains Figure 4 ISo ylogenetic tree. This relates they are traced backwards extremely difficult (BOX2), and the study in question did the mitocho of modern humans and also three not fulfil the criteria that in our experience are necessary Neanderthals from Croatia. Caucasusand Germany. NATURE REVIEWS GENETIC VOLUME 2 MAY 2001E Mo 2001 Macmillan Magazines Ltd
NATURE REVIEWS | GENETICS VOLUME 2 | MAY 2001 | 357 REVIEWS to exclude (to the greatest extent possible) that the results are due to contamination(s) (BOX 1). This might be a particular concern in this case, because the mtDNA in question shows similarity to a nuclear insertion sequence that has previously been shown to represent a contamination in ancient DNA studies26. So, further work, and perhaps even new technical approaches, are necessary before any definitive statements about the mtDNA gene pool of early modern humans can be made. In principle, another way to estimate the extent of any putative mixing between modern humans and Neanderthals would be to study nuclear genes from the Neanderthals themselves. However, even if we ignore the technical problems involved with this, it would not be very informative as we already know that many nuclear genes have COALESCENCES that pre-date the putative separation of Neanderthals and anatomically modern humans ~500,000 years ago. We can therefore reasonably assume that the amount of variation at most nuclear loci in Neanderthals falls within the amount of variation found among humans today37. So, the overall genetic relationship of Neanderthals and modern humans was not very distant and might in general have resembled that of subspecies of present-day apes more than that of different species34,38. However, the divergence of the mtDNA pools indicates a temporal separation that could be verified by studying, for example, Ychromosomal sequences, should this become possible in the future. The recent determination of mtDNA sequences of a Neanderthal from the Caucasus (~30,000 years old39) and of another from Croatia (over 42,000 years old40) (FIG. 4) points to the exciting possibility of studying aspects of the population history of this archaic group of hominids. For example, humans differ from the great apes in having much less DNA sequence variation both in mtDNA and in nuclear DNA38. Furthermore, the distribution of mtDNA sequence variation in contemporary humans indicates that they expanded from a smaller population ~50,000 years ago41, whereas there is little or no evidence for an expansion in the great apes. It would be extremely interesting to investigate mtDNA variation among Neanderthals to address the questions of whether they were similar to humans in having much variation and to obtain an indication of whether an Neanderthals went extinct without contributing mtDNA to contemporary humans. So, the view of modern human origins that assumes a recent African origin, and little genetic contribution to the current gene pool from archaic humans elsewhere is correct for mtDNA. It does not, of course, exclude the fact that mixing between modern humans, which arrived in Europe from Africa ~40,000 years ago, and Neanderthals took place, but there is no evidence of such a mixture from the molecular studies. If it becomes possible in the future to study mtDNAs from many early modern humans, the occurrence of Neanderthal mtDNA types among them would prove that mixing did occur. Recently, a report of mtDNA sequences from skeletons of early modern humans from Australia revealed a sequence that tends to fall outside the variation of modern humans36. This could represent a variant that was present in modern humans at the time, or alternatively a mtDNA contribution of archaic humans to the modern human gene pool that was subsequently lost by drift. However, contamination with contemporary human DNA makes the study of human remains extremely difficult (BOX 2), and the study in question did not fulfil the criteria that in our experience are necessary COALESCENCE The joining of genetic lineages to common ancestors when they are traced backwards in time. 1985 1990 1995 2000 Figure 3 | Some extinct organisms from which DNA sequences have been determined. From bottom left to top right: quagga56, marsupial wolf9,10,12, sabre-toothed cat57, moa27, mammoth58–62, cave bear63, blue antelope64, giant ground sloth65, Aurochs66, mastodon60, New Zealand coot67, South Island piopio11, Steller’s sea cow61, Neanderthal33,34,39,40, Aptornis defossor68, Shasta ground sloth46,48, pig-footed bandicoot69, moa-nalo70 and Myotragus balearicus71. Caucasus (Mezmaiskaya) Croatia (Vindija) Germany (Neanderthal) Neanderthal Modern human Africans & non-Africans Africans Africans Figure 4 | Schematic phylogenetic tree. This relates the mitochondrial DNAs of modern humans and also three Neanderthals from Croatia40, Caucasus39 and Germany33. © 2001 Macmillan Magazines Ltd
REVIEWS Using coprolite deposits, it is possible to follow the change in diet of defecating animals through time. For example, the analysis of faecal DNA from ground sloths has shown that pine forests in southern Nevada were present -28, 500 years ago. However, sloths living in the same cave almost 20,000 years later, at the end of the last glaciation and shortly before their extinction, were found to have a diet more reminiscent of the cur- rent flora in the region. Recently, DNA has also been retrieved from faeces obtained from three 2,000-year old Native Americans". Eight different plants were shown to have formed part of the diet of these individ- uals. One striking observation was the presence of meat from large mammals, such as pronghorn ante- lope and bighorn sheep, in these samples-a dietary component that is not possible to detect by a morpho- Figure 5 20,000-year-old ground sloth coprolite DNA coprolites are likely to prove a rich source of informa- sequences from both the defecator and its diet have been tion about the genetics and dietary habits of ancient determined from this sample nimals and humans Conclusions nsion took place, and if so, when. Although three Studies of ancient DNA from museum collections derthal mt DNA sequences are obviously too few to and Late Pleistocene animals have provided several versity, it is interesting to note that the new insights into the evolutionary history of both variation seen so far falls in the middle of the human specis a o nr DNA sequences has provided a molecu- sand populations, and the determination of distribution, whereas variation of the all the great ape Neand species is much larger lar perspective on the genetic relationship of current humans with an extinct hominid form. A particularly Coprolites. Faecal material is commonly found at exca- exciting development is the analysis of DNA from vations of human settlements and of caves that are fre- ancient coprolites because it yields information about quented by animals. Because contemporary scats have the diet and thus the behaviour of ancient organisms proved a useful source of DNA from both the defeca- However, technical pitfalls make the field liable to tor and its diet"-, it was frustrating that well-pre- dubious results unless many precautions and experi- served Late Pleistocene coprolites (FIG. 5)had failed to mental controls are implemented. This is particularly yield DNA. However, pyrolysis GC/MS studies showed true for human remains. Future developments that at coprolites, as well as other ancient remains, con- remove or repair chemical damage in the ancient tain large amounts of crosslinks between reducing sug- DNA, as well as use the presence of such damage to nd amino groups, presumably in proteins as well verify that DNA sequences are indeed ancient, would as in DNA. Therefore, it was a logical step to attempt to be welcome additions to the field. remove such crosslinks from coprolite extracts when PTB(N-phenacyl thiazolium bromide)-a reagent O) Links that breaks sugar crosslinks-became available 5 This FURTHER INFORMATION gas chromatography and mass made the amplification of coprolite DNA from cave spectrometry Miocene Pleistocene I quagga sites in the southwestern United States possible. marsupial wolf sabre-toothed cat I moa I mammoth Many of these caves contain layers of faeces that span cave bear blue antelopeIgiant ound sloth I ens of thousands of years". PTB is also helpful in mastodon Steller's sea cow Neanderthal bandicoots I animals extractions from some pleistocene bones o Myotragus balearicus 1. Hoss, M, Jaruga, P, Zastawny. T. Nature in the press) uceic Acids res 0701996 都5 X&Wison, A.C. M时d IFication. Proc. Naf J.Ma"Ea31.101-11201990 nan. M. Were ige, J.J., Jones, C. G, Bruton Nichols fauna? Proc. Nat 3. Paabo, S& Wilson, A C Miocene DNA sequences- R.A. 'Ghost alleles of the mauritius kestrel Nature 403 ad. Sci USA93.3898-39011996: era.um93.14992 Na re 362 y n- 15 ag g the primary structure oper, A et al. Andent DNA and island endemics. Nature 12. Krajewski, C-. Buckley, L& Westerman, M. DNA iew of DNA damage 9. Thomas, R. H, Schane W. Wilson, A C.& Paabo, s NA phyogeny of the extinct marsupial wolf Nature 340, 13. Handt, O, Hoss, M, Krings, M. S. Ancient DNA. 465-46701989 358 MAY 2001 VOLUME 2 Nnature. com/reviews/genetics Mo 2001 Macmillan Magazines Ltd
358 | MAY 2001 | VOLUME 2 www.nature.com/reviews/genetics REVIEWS Using coprolite deposits, it is possible to follow the change in diet of defecating animals through time. For example, the analysis of faecal DNA from ground sloths has shown that pine forests in southern Nevada were present ~28,500 years ago. However, sloths living in the same cave almost 20,000 years later, at the end of the last glaciation and shortly before their extinction, were found to have a diet more reminiscent of the current flora in the region48. Recently, DNA has also been retrieved from faeces obtained from three 2,000-yearold Native Americans49. Eight different plants were shown to have formed part of the diet of these individuals. One striking observation was the presence of meat from large mammals, such as pronghorn antelope and bighorn sheep, in these samples — a dietary component that is not possible to detect by a morphological analysis of the faeces. So, molecular analyses of coprolites are likely to prove a rich source of information about the genetics and dietary habits of ancient animals and humans. Conclusions Studies of ancient DNA from museum collections and Late Pleistocene animals have provided several new insights into the evolutionary history of both species and populations, and the determination of Neanderthal DNA sequences has provided a molecular perspective on the genetic relationship of current humans with an extinct hominid form. A particularly exciting development is the analysis of DNA from ancient coprolites because it yields information about the diet and thus the behaviour of ancient organisms. However, technical pitfalls make the field liable to dubious results unless many precautions and experimental controls are implemented. This is particularly true for human remains. Future developments that remove or repair chemical damage in the ancient DNA, as well as use the presence of such damage to verify that DNA sequences are indeed ancient, would be welcome additions to the field. expansion took place, and if so, when. Although three Neanderthal mtDNA sequences are obviously too few to gauge their diversity, it is interesting to note that the variation seen so far falls in the middle of the human distribution, whereas variation of the all the great ape species is much larger40. Coprolites. Faecal material is commonly found at excavations of human settlements and of caves that are frequented by animals. Because contemporary SCATS have proved a useful source of DNA from both the defecator and its diet42–44, it was frustrating that well-preserved Late Pleistocene coprolites (FIG. 5) had failed to yield DNA. However, pyrolysis GC/MS studies showed that coprolites, as well as other ancient remains, contain large amounts of crosslinks between reducing sugars and amino groups, presumably in proteins as well as in DNA. Therefore, it was a logical step to attempt to remove such crosslinks from coprolite extracts when PTB (N-phenacyl thiazolium bromide) — a reagent that breaks sugar crosslinks — became available45. This made the amplification of coprolite DNA from cave sites in the southwestern United States possible46. Many of these caves contain layers of faeces that span tens of thousands of years47. PTB is also helpful in extractions from some Pleistocene bones40. 1. Höss, M., Jaruga, P., Zastawny, T. H., Dizdaroglu, M. & Pääbo, S. DNA damage and DNA sequence retrieval from ancient tissues. Nucleic Acids Res. 24, 1304–1307 (1996). 2. Pääbo, S. Ancient DNA: extraction, characterization, molecular cloning, and enzymatic amplification. Proc. Natl Acad. Sci. USA 86, 1939–1943 (1989). 3. Pääbo, S. & Wilson, A. C. Miocene DNA sequences — a dream come true? Curr. Biol. 1, 45–46 (1991). 4. Lindahl, T. Instability and decay of the primary structure of DNA. Nature 362, 709–715 (1993). Provides a comprehensive review of DNA damage that includes what is expected to occur in archaeological and palaeontological specimens. 5. Collins, M. Neanderthal DNA: not just old but old and cold. Nature (in the press). 6. Thomas, W. K., Pääbo, S., Villablanca, F. X. & Wilson, A. C. Spatial and temporal continuity of kangaroo rat populations shown by sequencing mitochondrial DNA from museum specimens. J. Mol. Evol. 31, 101–112 (1990). 7. Groombridge, J. J., Jones, C. G., Bruford, M. W. & Nichols, R. A. ‘Ghost’ alleles of the Mauritius kestrel. Nature 403, 616 (2000). 8. Cooper, A. et al. Ancient DNA and island endemics. Nature 381, 484 (1996). 9. Thomas, R. H., Schaffner, W., Wilson, A. C. & Pääbo, S. DNA phylogeny of the extinct marsupial wolf. Nature 340, 465–467 (1989). 10. Krajewski, C., Driskell, A. C., Baverstock, P. R. & Braun, M. J. Phylogenetic relationships of the thylacine (Mammalia: Thylacinidae) among dasyuroid marsupials: evidence from cytochrome b DNA sequences. Proc. R. Soc. Lond. B 250, 19–27 (1992). 11. Christidis, L., Leeton, P. R. & Westerman, M. Were bowerbirds part of the New Zealand fauna? Proc. Natl Acad. Sci. USA 93, 3898–3901 (1996); erratum 93, 14992 (1996). 12. Krajewski, C., Buckley, L. & Westerman, M. DNA phylogeny of the marsupial wolf resolved. Proc. R. Soc. Lond. B 264, 911–917 (1997). 13. Handt, O., Höss, M., Krings, M. & Pääbo, S. Ancient DNA: methodological challenges. Experientia 50, 524–529 (1994). Figure 5 | 20,000-year-old ground sloth coprolite. DNA sequences from both the defecator and its diet have been determined from this sample46. Links FURTHER INFORMATION gas chromatography and mass spectrometry | Miocene | Pleistocene | quagga | marsupial wolf | sabre-toothed cat | moa | mammoth | cave bear | blue antelope | giant ground sloth | mastodon | Steller’s sea cow | Neanderthal | bandicoots | Myotragus balearicus SCATS Faecal material left behind by animals. © 2001 Macmillan Magazines Ltd
REVIEWS set of controls to support the authenticity of DNA 55. Wang. L. et al Genetic structure of a 2, 500-year-old human 34. Krings, M, Gel population in China and its spatiotemporal changes. Mal .Ew17.1396-140002000 Wood, A, Capeli. C, Possnert, G& Paabo, s. 56. Higuchi, R, Bowman, B, Freiberger, M, Ryder, O A& . Mo. BioL Ew.16.1466-1473(1999 av ien h: A Decks icm e horse family. Hauswirth, w.w.&Parham, P. 35. hgman, M, Kaessmann, H, Paabo, s& Gylensten, U 57. Janczewski, D N, Yuhki, N, Gibert, D. A, Jefferson, G. T. iochondial genome variation and the origin of modern O'Brien, s.J. Molecuar phylogeney La Brea. Proc. Natf 17. Poinar H. N. Hos Bada, J.L. aabo, sAmino acid 36. Adcock, G ef al Mitochondrial DNA sequences in ancient ad. SCi USA89.9769-977301 272.864-866(1996 Nat Acad. Sci USA 98, 537-542(2001 Paabo, S. Human evolution Trends Cel BioL 9, M13-M16 59. Hoss, M, Paabo, S& Vereshchagin, N. K Mammoth DNA nnot) survive 38 Kaess S Great 60 18 H N.& Stankiewicz, B. A Protein preservation sues. Proc NaU Acad. sci innikow, L. V. et al Molecuar analys mericanum as an outgroup. Proc Natl Acad. Sci USA 19. Handt, O Krings, M, Ward, R H& NA from the northern caucasus. Nature 404 490-493 J. Hum Genet 5( S& Mkhebson, V M Phylogenetic 40. Krings, M. et al A view of netic diversity 20. Morin, P. uantitative DNA analysis 41. Rogers, A.R.& Harpending H Population growth makes 物大签1p asuda R. Dubrovo L A. Yoshida. M. C.& 365. 700 42. Hoss. M. Kohn. M. Paabo, s. KI (1993] t analysis by PCR. Nature 359, 199(199 sequences of mitochondrial cytochrome band 12s te 22. Austin, J.J., Ross, A.J., Smith, A B, Fortey, R.A.& 43. Kohn, M. Knauer F. S:offella, A. Schroder. W.& Paabo s. 23. Austin, /. J Smith. A B. Thomas, R H. Paleontology in 44 Sequences. Mol Eco. 4. 95-103a199 end mitochondris 63. Hanni, C, Laudet, V, Stehen, D.& Taberlet, P. Trar Thomas, R H Problems of reproducibility-doe Conservation genetics of the European brown bear-a faeces. Proc R Soc. Lond. B 266, 657-663(1999) Halanych, K M& Herzig chondrial DNA sequence relationships of the extinct 24. Stankiewicz, B, Poinar, H, Briggs, D, Evershed. R& and in vivo Nature 382, 275-278(1996) 46. Poinar, H N ef a Molecular coproscopy: dung and diet of 99 othrotheriops shafer arkia fossis. Phil. Trans. R So B33 3 d sloth food habits, Rampart afl Acad. Sci USA93, 181-185(1 1991) DNA suggests a recent expansion 26 s B253146 editor species. A nuckar fossil' of the mitochondrial D-loop and the origin 27. Codope A ef ac naepeadenst uss Bg xleaz44d moas for mo eera stustrasing the usetuiness of coproit ealand region 52.429-446(1996 68. and,A E. Montano three archaic Native Americans. Proc. Nat Acad Sci. USA Systems equences xtinct moas clarify ratite evolution. 98.4317-4322 edn(Chicago The first determination of complete mitochondrial V, Cardenas, F, Guhl, F, Delaney, A D& 29. Leonard, J A, Wayne. R.K.& Cooper, A From the cover: vine, D V. Phylogenetic analysis of mtDNA lineages in 30. Paabo, s o bears, conservation genetics, and the value of 52. izaguirre, N& ne travel Proc. Nat/ Acad Sci. USA 97 Basque populations: implications for haplogroup A.Po.R. Soc Lond. B266.2187-2193(1999 najor paleolithic expansion from southwestern 71. Laueza-Fox Shader N.& 31. vila, C. et al Widespread origins of domestic horse Genet65,199-2070199 53. oota alearicus, an extinct bovid from the Balearic Islands.J. zoL288.56-622000. a.cmCm112020300 odern Japanese population. Am J. Hum. Genet. 64 mples; and humans. Cel 90, 19-30(199 Herrmann, B. Ampification of Y- The first determination of a Neanderthal noent skeletal material. Hum. undesministerurm fur Baldung und Forschung for financal mitochondrial DNA sequence, including an extensive Genet 104 161999 NATURE REVIEWS GENETIC VOLUME 2 MAY 2001 359 Mo 2001 Macmillan Magazines Ltd
NATURE REVIEWS | GENETICS VOLUME 2 | MAY 2001 | 359 REVIEWS 14. Höss, M., Handt, O. & Pääbo, S. in The Polymerase Chain Reaction (eds Mullis, K., Ferre, F. & Gibbs, R.) 257–264 (Birkhauser, Boston, Massachusetts, 1994). 15. Greenwood, A., Capelli, C., Possnert, G. & Pääbo, S. Nuclear DNA sequences from Late Pleistocene megafauna. Mol. Biol. Evol. 16, 1466–1473 (1999). 16. Lawlor, D. A., Dickel, C. D., Hauswirth, W. W. & Parham, P. Ancient HLA genes from 7,500-year-old archaeological remains. Nature 349, 785–788 (1991). 17. Poinar, H. N., Höss, M., Bada, J. L. & Pääbo, S. Amino acid racemization and the preservation of ancient DNA. Science 272, 864–866 (1996). Presents amino-acid analysis as a tool to substantiate claims that DNA can (or cannot) survive in ancient organic remains. 18. Poinar, H. N. & Stankiewicz, B. A. Protein preservation and DNA retrieval from ancient tissues. Proc. Natl Acad. Sci. USA 96, 8426–8431 (1999). 19. Handt, O., Krings, M., Ward, R. H. & Pääbo, S. The retrieval of ancient human DNA sequences. Am. J. Hum. Genet. 59, 368–376 (1996). 20. Morin, P., Chambers, K., Boesch, C. & Vigilant, L. Quantitative DNA analysis from noninvasive samples for accurate microsatellite genotyping of wild chimpanzees (Pan troglodytes verus). Mol. Ecol. 10 (in the press). 21. Lindahl, T. Recovery of antediluvian DNA. Nature 365, 700 (1993). 22. Austin, J. J., Ross, A. J., Smith, A. B., Fortey, R. A. & Thomas, R. H. Problems of reproducibility — does geologically ancient DNA survive in amber-preserved insects? Proc. R. Soc. Lond. B 264, 467–474 (1997). 23. Austin, J. J., Smith, A. B. & Thomas, R. H. Paleontology in a molecular world: the search for authentic ancient DNA. Trends Ecol. Evol. 12, 303–306 (1997). 24. Stankiewicz, B., Poinar, H., Briggs, D., Evershed, R. & Poinar, G. Chemical preservation of plants and insects in natural resins. Proc. R. Soc. Lond. B 265, 641–647 (1998). 25. Sidow, A., Wilson, A. C. & Pääbo, S. Bacterial DNA in Clarkia fossils. Phil. Trans. R. Soc. Lond. B 333, 429–433 (1991). 26. Zischler, H., Geisert, H., von Haeseler, A. & Pääbo, S. A nuclear ‘fossil’ of the mitochondrial D-loop and the origin of modern humans. Nature 378, 489–492 (1995). 27. Cooper, A. et al. Independent origins of New Zealand moas and kiwis. Proc. Natl Acad. Sci. USA 89, 8741–8744 (1992). 28. Cooper, A. et al. Complete mitochondrial genome sequences of two extinct moas clarify ratite evolution. Nature 409, 704–707 (2001). The first determination of complete mitochondrial DNAs from fossil remains. 29. Leonard, J. A., Wayne, R. K. & Cooper, A. From the cover: population genetics of ice age brown bears. Proc. Natl Acad. Sci. USA 97, 1651–1654 (2000). 30. Pääbo, S. Of bears, conservation genetics, and the value of time travel. Proc. Natl Acad. Sci. USA 97, 1320–1321 (2000). 31. Vila, C. et al. Widespread origins of domestic horse lineages. Science 291, 474–477 (2001). 32. Loreille, O. et al. Ancient DNA analysis reveals divergence of the cave bear, Ursus spelaeus, and brown bear, Ursus arctos, lineages. Curr. Biol. 11, 200–203 (2001). 33. Krings, M. et al. Neandertal DNA sequences and the origin of modern humans. Cell 90, 19–30 (1997). The first determination of a Neanderthal mitochondrial DNA sequence, including an extensive set of controls to support the authenticity of DNA sequence. 34. Krings, M., Geisert, H., Schmitz, R. W., Krainitzki, H. & Pääbo, S. DNA sequence of the mitochondrial hypervariable region II from the neandertal type specimen. Proc. Natl Acad. Sci. USA 96, 5581–5585 (1999). 35. Ingman, M., Kaessmann, H., Pääbo, S. & Gyllensten, U. Mitochondrial genome variation and the origin of modern humans. Nature 408, 708–713 (2000). 36. Adcock, G. et al. Mitochondrial DNA sequences in ancient Australians: implications for modern human origins. Proc. Natl Acad. Sci. USA 98, 537–542 (2001). 37. Pääbo, S. Human evolution. Trends Cell Biol. 9, M13–M16 (1999). 38. Kaessmann, H., Wiebe, V., Weiss, G. & Pääbo, S. Great ape DNA sequences reveal a reduced diversity and an expansion in humans. Nature Genet. 27, 155–156 (2001). 39. Ovchinnikov, I. V. et al. Molecular analysis of Neanderthal DNA from the northern Caucasus. Nature 404, 490–493 (2000). 40. Krings, M. et al. A view of Neandertal genetic diversity. Nature Genet. 26, 144–146 (2000). 41. Rogers, A. R. & Harpending, H. Population growth makes waves in the distribution of pairwise genetic differences. Mol. Biol. Evol. 9, 552–569 (1992). 42. Höss, M., Kohn, M., Pääbo, S., Knauer, F. & Schröder, W. Excrement analysis by PCR. Nature 359, 199 (1992). 43. Kohn, M., Knauer, F., Stoffella, A., Schröder, W. & Pääbo, S. Conservation genetics of the European brown bear — a study using excremental PCR of nuclear and mitochondrial sequences. Mol. Ecol. 4, 95–103 (1995). 44. Kohn, M. H. et al. Estimating population size by genotyping faeces. Proc. R. Soc. Lond. B 266, 657–663 (1999). 45. Vasan, S. et al. An agent cleaving glucose-derived protein crosslinks in vitro and in vivo. Nature 382, 275–278 (1996). 46. Poinar, H. N. et al. Molecular coproscopy: dung and diet of the extinct ground sloth Nothrotheriops shastensis. Science 281, 402–406 (1998). 47. Hansen, R. M. Shasta ground sloth food habits, Rampart Cave, Arizona. Paleobiology 4, 302–319 (1978). 48. Hofreiter, M. et al. A molecular analysis of ground sloth diet through the last glaciation. Mol. Ecol. 9, 1975–1984 (2000). A diachronical study of ground sloth diet over almost 20,000 years, illustrating the usefulness of coprolites for molecular studies. 49. Poinar, H. et al. A molecular analysis of dietary diversity for three archaic Native Americans. Proc. Natl Acad. Sci. USA 98, 4317–4322. 50. Klein, R. The Human Career 2nd edn (Chicago Univ. Press, Chicago, 1999). 51. Monsalve, M. V., Cardenas, F., Guhl, F., Delaney, A. D. & Devine, D. V. Phylogenetic analysis of mtDNA lineages in South American mummies. Ann. Hum. Genet. 60, 293–303 (1996). 52. Izagirre, N. & de la Rua, C. An mtDNA analysis in ancient Basque populations: implications for haplogroup V as a marker for a major paleolithic expansion from southwestern Europe. Am. J. Hum. Genet. 65, 199–207 (1999). 53. Oota, H., Saitou, N., Matsushita, T. & Ueda, S. Molecular genetic analysis of remains of a 2,000-year-old human population in China — and its relevance for the origin of the modern Japanese population. Am. J. Hum. Genet. 64, 250–258 (1999). 54. Schultes, T., Hummel, S. & Herrmann, B. Amplification of Ychromosomal STRs from ancient skeletal material. Hum. Genet. 104, 164–166 (1999). 55. Wang, L. et al. Genetic structure of a 2,500-year-old human population in China and its spatiotemporal changes. Mol. Biol. Evol. 17, 1396–1400 (2000). 56. Higuchi, R., Bowman, B., Freiberger, M., Ryder, O. A. & Wilson, A. C. DNA sequences from the quagga, an extinct member of the horse family. Nature 312, 282–284 (1984). 57. Janczewski, D. N., Yuhki, N., Gilbert, D. A., Jefferson, G. T. & O’Brien, S. J. Molecular phylogenetic inference from saber-toothed cat fossils of Rancho La Brea. Proc. Natl Acad. Sci. USA 89, 9769–9773 (1992). 58. Hagelberg, E. et al. DNA from ancient mammoth bones. Nature 370, 333–334 (1994). 59. Höss, M., Pääbo, S. & Vereshchagin, N. K. Mammoth DNA sequences. Nature 370, 333 (1994). 60. Yang, H., Golenberg, E. M. & Shsoshani, J. Phylogenetic resolution within the Elephantidae using fossil DNA sequences from the American mastodon (Mammut americanum) as an outgroup. Proc. Natl Acad. Sci. USA 93, 1190–1194 (1996). 61. Ozawa, T., Hayashi, S. & Mikhelson, V. M. Phylogenetic position of mammoth and Steller’s sea cow within Tethytheria demonstrated by mitochondrial DNA sequences. J. Mol. Evol. 44, 406–413 (1997). 62. Noro, M., Masuda, R., Dubrovo, I. A., Yoshida, M. C. & Kato, M. Molecular phylogenetic inference of the woolly mammoth Mammuthus primigenius, based on complete sequences of mitochondrial cytochrome b and 12S ribosomal RNA genes. J. Mol. Evol. 46, 314–326 (1998). 63. Hänni, C., Laudet, V., Stehelin, D. & Taberlet, P. Tracking the origins of the cave bear (Ursus spelaeus) by mitochondrial DNA sequencing. Proc. Natl Acad. Sci. USA 91, 12336–12340 (1994). 64. Robinson, T. J., Bastos, A. D., Halanych, K. M. & Herzig, B. Mitochondrial DNA sequence relationships of the extinct blue antelope Hippotragus leucophaeus. Naturwissenschaften 83, 178–182 (1996). 65. Höss, M., Dilling, A., Currant, A. & Pääbo, S. Molecular phylogeny of the extinct ground sloth Mylodon darwinii. Proc. Natl Acad. Sci. USA 93, 181–185 (1996). 66. Bailey, J. F. et al. Ancient DNA suggests a recent expansion of European cattle from a diverse wild progenitor species. Proc. R. Soc. Lond. B 263, 1467–1473 (1996). 67. Trewick, S. A. Flightlessness and phylogeny amongst endemic rails (aves: Rallidae) of the New Zealand region. Phil. Trans. R. Soc. Lond. B 352, 429–446 (1996). 68. Houde, P., Cooper, A., Leslie, E., Strand, A. E. & Montano, G. A. in Avian Molecular Evolution Systems and Systematics (ed. Mindell, D. P.) 121–158 (Academic, London, 1997). 69. Westerman, M., Springer, M. S., Dixon, J. & Krajewski, C. Molecular relationships of the extinct pig-footed bandicoot Chaeropus ecaudatus (Marsupialia: Perameloidea) using 12S rRNA sequences. J. Mamm. Evol. 6, 271–288 (1999). 70. Sorenson, M. D. et al. Relationships of the extinct moanalos, flightless Hawaiian waterfowl, based on ancient DNA. Proc. R. Soc. Lond. B 266, 2187–2193 (1999). 71. Lalueza-Fox, C., Bertranpetit, J., Alcover, J. A., Shailer, N. & Hagelberg, E. Mitochondrial DNA from Myotragus balearicus, an extinct bovid from the Balearic Islands. J. Exp. Zool. 288, 56–62 (2000). Acknowledgements We thank Qin Zhan-Xiang for providing the cave bear samples; and the Max Planck Society, the Deutsche Forschungsgemeinschaft and the Bundesministerium für Bildung und Forschung for financial support. © 2001 Macmillan Magazines Ltd