¿ìè¶ÌÊÓƵ

A palaeomtologist was asked what he thought of the film Jurassic Park.
‘Totally unrealistic!’ came the swift, certain reply. When pressed for an
explanation, he said, ‘Well, didn’t you see? A palaeontologist was offered
a drink – and he turned it down. That’s totally unrealistic!’

Palaeontologists have become used to being the butt of such jokes, ever
since Michael Crichton’s fantasy of recreating a dinosaur world reached
the silver screen. This one was told by Michael Braun, a researcher at the
Smithsonian Institution’s Laboratory of Molecular Systematics, just outside
Washington DC. The occasion was an international gathering of specialists
in ancient DNA, held in Washington in November, which Braun helped to organise.
The jokes over, the conference turned to the serious issue: that of deciding
how realistic the new field of ancient DNA research really is.

Three years ago, at the first international conference on ancient DNA,
there appeared to be few limits to the possibilities of gazing into ancient
worlds by extracting genes from fossils. News of the recovery of DNA from
17-million-year-old fossilised leaves had just hit the headlines. But now
palaeontologists are preaching caution. ‘Relatively little of the ‘exciting’
work talked about (two years ago) has found its way into the scientific
journals,’ says Bryan Sykes, a geneticist at the University of Oxford. ‘Many
of the results turned out not to be repeatable.’

Nor are published findings about ancient DNA immune from attack. The
idea that DNA can survive beyond a few millennia, let alone many millions
of years, has always been controversial. But over the past two years scepticism
has hardened in the wake of a wave of startling claims. How can the authors
be certain of their findings, ask critics, since it is no easy task to extract
DNA from organisms long dead and still harder to be sure that the DNA is
not some kind of contaminant.

COLD FUSION

One prominent sceptic is Tomas Lindahl, a biochemist with the Imperial
Cancer Research Fund in London. He likens the recent spate of results from
amber to the initial excitement over cold fusion: ‘Let’s see what the results
look like when they’ve been properly scrutinised and attempts have been
made to replicate them.’

Controversy is also threatening to flare up over the impact of ancient
DNA ‘hunting’ on fossil and skin collections. Long regarded as an unglamorous
realm of biology, museum collections are now being seen as valuable genetic
resources. This raises their value at a time when many collections are being
neglected or broken up to save money or museum space. But it might also
lead to conflicts between curators and molecular biologists.

‘Museum curators, who are conservative by nature, spend years maintaining
collections as long-term investments,’ says Richard Thomas, a geneticist
at the Natural History Museum in London. Molecular biologists, who are renowned
for their arrogance, ‘come in and knock off a nifty bit of work in a few
days, without appreciating the contribution of others’. In some cases, the
DNA research cannot be done without damaging or destroying a specimen.

But these concerns are not lost on those involved in the hunt for ancient
DNA: the criticisms are encouraging a shift in academic ambitions. Researchers
are becoming more interested in tackling real questions in evolutionary
biology and population genetics than in ‘the single flashy, splashy, 100-million-year-old
result’. ‘There has been an obsession with reconstructing the genomes of
dinosaurs,’ says Robert Wayne, who divides his time between the University
of California at Los Angeles and the Institute of Zoology, London. ‘But
it is changing.’

Ancient DNA research was born in 1984, when Allan Wilson and Russell
Higuchi, biologists at the University of California, Berkeley, extracted
fragments of DNA from museum specimens of the quagga, a zebra-like animal
that became extinct more than a century ago. The announcement was a great
surprise, because biologists knew that when an organism dies its tissues
quickly decompose. The DNA breaks down extremely rapidly under the inexorable
attack of hydrolysis, oxidation and cosmic radiation. Even under the most
favourable conditions, only small fragments are likely to remain.

A year later molecular biologist Svante Paabo, then at the University
of Uppsala but now at the University of Munich, extracted human DNA from
an Egyptian mummy almost 2500 years old. It was Paabo again who began to
expand the limits of the field when, in 1989, he reported extracting DNA
from a specimen of the extinct ground sloth, some 13 000 years old. He did
the same with a 40 000-year-old woolly mammoth, preserved in Siberian permafrost.

Early in 1990, came the first of the really extravagant claims. Biologist
Edward Golenberg and his colleagues, at the University of California, Riverside,
announced the recovery of DNA fragments from 17-million-year-old magnolia
leaves, from an unusual fossil deposit in Idaho. This was followed in 1992
and 1993 by the truly mind-boggling reports of 25-million-year-old DNA from
a bee and a termite and 120-million-year-old DNA from a weevil. Each specimen
had been entombed in amber since its death.

In all studies of ancient DNA so far, the recovered genetic material
has been limited to fragments no longer than about 800 base pairs, and more
typically 200. By comparison, DNA in living tissues exists as strings of
tens of thousands of base pairs, as components of chromosomes. Nevertheless,
even short DNA strings may contain sufficient genetic information to identify
the species and compare the DNA sequence with that of modern DNA from descendant
species or populations.

CURSE OF CONTAMINATION

A vital factor in the recent surge in ancient DNA research is the Polymerase
Chain Reaction (PCR). Developed in 1985, this biochemical method permits
vanishingly small quantities of DNA to be multiplied, producing sufficient
material for analysis. In principle, PCR can produce millions of copies
from a single DNA molecule. But this power is also a curse, because PCR
doesn’t discriminate between DNA from the specimen and rogue DNA. This might
come from bacteria or fungi that infected the specimen during life or at
death; or from the sweat of curators who have handled the specimen over
the years or another source of modern contamination.

Sykes told delegates at the Washington conference that he and his colleagues
once amplified DNA from mammoth bones only to discover its source was human,
not mammoth. ‘We eventually traced it to one of the technicians in the lab,’
said Sykes. ‘It’s unimaginable how present the danger of contamination is,’
he continued. ‘One aerosol droplet (from a sneeze or a spray from experimental
solution) can contain ten thousand copies of a DNA sequence.’

Such anecdotes are grist to the mill of sceptics like Lindahl. In the
1970s, Lindahl conducted a series of careful laboratory experiments on naked
DNA (that is, DNA devoid of its protective proteins) in water, from which
he calculated the rate at which DNA would degrade under normal conditions.
In an article in Nature in April last year, Lindahl said ‘it can be predicted
that. . . fully hydrated DNA is spontaneously degraded to short fragments
over a time period of several thousand years at moderate temperatures’.
Under the most favourable conditions, he continued, ‘it seems feasible that
useful DNA sequences aged tens of thousands of years could be recovered’
– but not DNA from older fossils.

Further on in his paper, Lindahl attacks the specific claims of two
researchers at the American Museum of Natural History in New York. Last
year Rob DeSalle and David Grimaldi reported that they extracted termite-like
DNA from a termite trapped in amber. In addition to the termite DNA, DeSalle
and Grimaldi found fruit fly DNA in their extract, which they dismiss as
contamination. But if the fruit fly DNA is a contaminant, says Lindahl,
then so too could be the termite DNA, its identity with the insect in the
amber being mere coincidence.

WAR OF WORDS

Lindahl reserves his more barbed criticism for the findings from the
magnolia leaves, which, he says, ‘don’t make any sense’. The 17-million-year-old
leaves were trapped in a deposit known to preserve fossils unusually well.
Nevertheless, Lindahl observes, it is wet, and water is the prime chemical
assailant of vulnerable DNA. ‘The DNA simply could not have survived under
those conditions,’ he insists.

Researchers in the firing line have reacted defensively. Golenberg challenges
Lindahl’s basic assumption about rates of DNA degradation under natural
conditions. ‘Lindahl fails in basic scientific technique,’ charges Golenberg.
‘You are supposed to generate hypotheses and then test them against empirical
observations. Instead, he tests the validity of the observations against
the hypothesis.’ In other words, Lindahl states that he knows DNA cannot
survive longer than a few tens of thousands of years – so any claims for
longer preservation must be wrong.

Lindahl is also criticised for approaching the issue as a chemist and
not as a molecular geneticist. ‘He ignores the fact that DNA contains information
that we can use to check our results,’ says Golenberg. Any supposed ancient
DNA should be similar to, but not exactly the same as, modern DNA from the
same type of species, thus indicating an evolutionary link between past
and present. If the termite DNA which Grimaldi and his colleagues extracted
had been a modern contaminant, the DNA would also be modern. ‘Looking for
a slightly modified sequence is both a goal and a test of our work,’ explains
Grimaldi.

Lindahl unleashed a second volley of criticisms in Nature less than
two weeks after the Washington meeting, his prime target this time being
the claims of George Poinar and his colleagues at the University of California,
Berkeley. This team has provoked a storm of publicity, most recently with
reports of DNA from the amber-trapped weevil living at the time of the dinosaurs.
Though initially incredulous of such findings, other researchers said they
were impressed when they examined the data for the first time at the Washington
conference.

Not so Lindahl. ‘The results from Poinar and his colleagues were even
worse than the ones I’d referred to in the first (Nature) paper,’ he says.
‘The experiments are not satisfactorily controlled and the results cannot
be accepted as they stand.’ In his latest critical paper, Lindahl says pointedly
of Poinar’s data: ‘It is hardly surprising that insect-like DNA can be
detected by PCR in experiments carried out in a department of entomology.’
Poinar works at the Department of Entomological Sciences at Berkeley.

The implication that the results are a consequence of sloppy work and
contamination could hardly be less guarded. To which Poinar replies in the
same issue of Nature: ‘None of the extraction, amplification or sequencing
was conducted in a department of entomology.’ That may be so, observes
Lindahl, but contamination can easily occur via the clothing of people who
go between the entomology department and the laboratory. And so the debate
goes on.

There is considerable sympathy for Poinar’s view. ‘As a geologist reading
Lindahl’s paper, I’d say he ignores many conditions under which rates of
DNA degradation would vary,’ says Jerre Lipps of the University of California,
Berkeley. ‘We don’t know what the rates of degradation are under all these
conditions, but you can’t ignore the likely variation.’ Just how fickle
the variation might be is obvious from the observations of Noreen Tuross,
a palaeontologist at the Smith-sonian Institution. Compare 20 bones from
a burial ground, and you’ll see different degrees of DNA preservation in
each, she says. And you will find differences in 20 samples from a single
bone.

Braun points out that DNA in nature is associated with various proteins,
such as histones, which protect it. Moreover, several environmental circumstances
exist that rapidly exclude water and oxygen, offering potential protection
against extensive DNA degradation. ‘No one can say it is impossible for
DNA to persist past twenty thousand years,’ says Braun. ‘Lindahl did us
a service showing how DNA degrades under defined conditions, but now we
need to know what happens under ill-defined conditions.’

Lindahl readily concedes there will be some variability in the rate
of DNA degradation, but is still cautious about how extensive this will
be. ‘I’d accept that you’d see variation up to two orders of magnitude,’
he says. This still puts an upper limit of some tens of thousands of years
on the oldest possible recoverable DNA, even in the most favourable circumstances
– leaves preserved in the sediments of oxygen-free swamps, insects trapped
in pine resin that turns into amber – protection is never 100 per cent.
‘Nothing is completely dry, completely oxygen-free, and completely shielded
from radiation,’ says Lindahl.

Initially amber may have the effect of removing water and ‘fixing’ the
DNA. But in the long run Lindahl suspects that amber is sufficiently permeable
to gases as to allow oxygen to seep to the deep interior, degrading the
trapped insect corpse. And damaging cosmic radiation will reach most fossil-bearing
locations. If so, then how are DeSalle, Poinar and their teams able to extract
DNA from insects many millions of years old? Are the researchers victims
of undetected contamination? Or is there something unexpected about the
chemistry of amber that allows preservation beyond theoretical limits?

It would help if researchers from different laboratories could team
up and replicate each other’s experiments on different samples of the same
fossils. But here again there is a problem: some of the specimens are extremely
rare and extracting insect DNA inevitably destroys or damages the specimen.
At the Washington meeting, Poinar was asked why his team had wanted to extract
DNA from the 120-million-year-old weevil. Are such experiments being done
simply because the specimens exist? Or is the aim to throw light on genuine
evolutionary questions?

Grimaldi believes that rare specimens, such as unique insects in amber,
should never be used for DNA analysis: ‘For work to be acceptable, it must
involve clear and important evolutionary problems; and there must be abundant
material from which to sample.’

HURT FEELINGS

Beyond professional recognition and hurt feelings, it is important to
establish fair means of access to collections. Although the major institutions
are beginning to tackle this, many smaller ones are without the resources
to do so. In private, several museum professionals speak of instances in
which molecular biologists have sampled specimens without proper permission.
The Natural History Museum in London was among the first to draw up guidelines
for molecular biologists seeking access to specimens. ‘While the technical
problems of the work are still so great, it’s important to be certain
of the competence of the individual who will do the work,’ says Thomas.
‘And when the work is done, the museum has certain rights over the products
of the research, such as the DNA extracted.’

The Smithsonian Institution has this same provision. ‘We quickly realised
that a specimen might be completely destroyed if it were repeatedly sampled,’
explains Robert Hoffman, assistant secretary for science, at the Smithsonian
Institution. ‘We therefore decided that when DNA is successfully extracted,
part of it should go into our collection, so that others may have access
to it.’ Establishing a protocol is, of course, just the first, and easiest,
step. Implementing it may be quite a different task, as Vicki Funk, a botanist
at the Smithsonian Institution, makes clear. ‘There are rules for the return
of material,’ she says. ‘But I bet the number that does come back is small.
To be honest, the system doesn’t work very well.’

The new field of ancient-DNA research has transformed the value and
role of museums in the world of modern biology. It is also forcing a rethinking
of the future, as Braun notes: ‘We have a responsibility for making collections
for future research, some of which will use techniques we know about but
others will rest on innovations about which we currently know nothing.’
That’s one of the big lessons of research into ancient DNA.

* * *

Prehistory in a strand of DNA

What do Ethiopian jackals and the indigenous people of Panama have in
common with a 25-million-year-old termite? All three are having their evolutionary
pasts scrutinised through the looking glass of ancient DNA.

A decade or so ago molecular biologists were restricted to using genetic
material from living organisms to answer evolutionary questions. But that
changed with the revelation that it is possible to extract ‘ancient’ DNA
from dead organisms. Now, in principle at least, researchers can compare
genes of living organisms with those of their ancestors, looking for differences
in DNA sequences that might hold clues to evolution.

If the ancient DNA is many tens of thousands or a few million years
old, it could conceivably shed light on the evolutionary origins, or ‘family
tree’, of a group of organisms – a macroevolutionary problem. Alternatively,
DNA from organisms that died a few hundred or a few thousand years ago might
reveal how the genetic diversity of a population has been affected by environmental
events such as habitat changes – a micro-evolutionary problem.

Microevolutionary problems are likely to prove more tractable, and for
two reasons. First, the extraction of DNA fragments from relatively young
fossil material is easier. Secondly, in most instances, specimens of recently
dead organisms (in museum collections) are much more readily available and
more numerous than ancient relics.

But macroevolution will not disappear from the agenda. This became clear
in 1992 when Rob DeSalle and his colleagues, at the American Museum of Natural
History, New York, published their report on DNA extracted from a 25 to
30-million-year-old termite entombed in Dominican amber. The researchers
examined short sequences of DNA from ribosomes, tiny particles inside cells
from which proteins are synthesised. The aim was to compare DNA sequences
from the fossil termite with those from living termites and roaches. This
comparison helped to resolve certain issues in the structure of the family
tree (or phylogeny) of modern termites. It also provided a unique insight
into the genetic variation of extinct populations.

Macroevolutionary questions may sometimes be tackled with single specimens,
as in the case of the termite. But you normally need several specimens for
microevolutionary problems. The reason is that such problems concern genetic
profiles within populations of a species, not simple comparisons between
species. For instance, Robert Wayne and his colleagues at the Institute
of Zoology in London are applying ‘ancient DNA’ techniques to studying the
genetic variability of modern endangered animal populations. Their recent
research focuses on African wild dogs, which are endangered in East Africa
but numerous in southern Africa.

In Kruger National Park, South Africa, the population of three hundred
animals is known to have limited genetic variation. Was this the result
of the population being isolated? Or did some event in the recent past cause
the population to shrink, thus losing existing variation? By extracting
certain fragments of mitochondrial DNA from ten skins that had been collected
six decades ago, Wayne and his colleagues showed that genetic variability
had been three times as great then as at present. This suggests a recent
population bottleneck.

The researchers have also collected DNA from skins around the same age,
to reconstruct the recent population history in Ethiopian jackals (the most
endangered canid species), wombats in Australia and the red wolf in North
America. ‘By looking at moderately ancient DNA from skins in museum collections,
we can learn about the genetic history of modern populations that was previously
unavailable to us, except through speculation,’ says Wayne.

Anthropologists have long disputed the pattern of colonisation of the
Americas, but there are many subplots to this, too. Connie Kolman of the
Smithsonian Tropical Research Institute, Panama, is looking at an odd genetic
disparity between two linguistic groups in the west of that country. The
Chibcha and the Choco people apparently entered Panama as a unified group
more than 7000 years ago, and then split.

Modern populations of the Chibcha linguistic group have a much lower
level of genetic variation than in Choco populations. This might mean that
the Chibcha were founded by an extremely small group. Or it could mean that
they and the Choco people responded differently to the Spanish invasion,
with the Chibcha population crashing to low levels, shedding genetic variation
in the process.

The answer will come from a sampling of DNA from people who lived prior
to the Spanish conquest. If genetic variation in the Chibcha was low at
that point, then the small size of the founding population would be the
answer. If it was normal, then post-conquest events would be the explanation.