Ancient DNA makes pre-history an open book

My Saturday essay in the Wall Street Journal:

Imagine what it must have been like to look through the first
telescopes or the first microscopes, or to see the bottom of the
sea as clearly as if the water were gin. This is how students of
human prehistory are starting to feel, thanks to a new ability to
study ancient DNA extracted from bodies and bones in archaeological
sites.

Low-cost, high-throughput DNA sequencing—a technique in which
millions of DNA base-pairs are automatically read in
parallel—appeared on the scene less than a decade ago. It has
already transformed our ability to see just how the genes of human
beings, their domestic animals and their diseases have changed over
thousands or tens of thousands of years.

The result is a crop of new insights into precisely what
happened to our ancestors: when and where they migrated, how much
they intermarried with those they met along the way and how their
natures changed as a result of evolutionary pressures. DNA from
living people has already shed some light on these questions.
Ancient DNA has now dramatically deepened—and sometimes
contradicted—those answers, providing a much more dynamic view of
the past.

It turns out that, in the prehistory of our species, almost all
of us were invaders and usurpers and miscegenators. This scientific
revelation is interesting in its own right, but it may have the
added benefit of encouraging people today to worry a bit less about
cultural change, racial mixing and immigration.

Consider two startling examples of how ancient DNA has solved
long-standing scientific enigmas. Tuberculosis in the Americas
today is derived from a genetic strain of the disease brought by
European settlers. That is no great surprise. But there’s a twist:
1,000-year-old mummies found in Peru show symptoms of TB as well.
How can this be—500 years before any Europeans set foot in the
Americas?

In a study published late last
year in the journal Nature, Johannes Krause of the Max Planck
Institute for the Science of Human History in Jena, Germany, and
his colleagues found that all human strains of tuberculosis share a
common ancestor in Africa about 6,000 years ago. The implication is
that this is when and where human beings first picked up TB. It is
much later than other scientists had thought, but Dr. Krause’s
finding only deepened the mystery of the Peruvian mummies, since by
then, their ancestors had long since left Africa.

Modern DNA cannot help with this problem, but reading the DNA of
the tuberculosis bacteria in the mummies allowed Dr. Krause to
suggest an extraordinary explanation. The TB DNA in the mummies
most resembles the DNA of TB in seals, which resembles that of TB
in goats in Africa, which resembles that of the earliest strains in
African people. So perhaps Africans gave tuberculosis to their
goats, which gave it to seals, which crossed the Atlantic and gave
it to native Americans.

Another genetic puzzle has been the fact that most modern
Europeans have certain DNA sequences that are similar to those of
some American Indians but different from those of most Asians,
including natives of Siberia. How can this be, since American
Indians are supposedly descended from Asians who migrated across
the Bering land bridge from Siberia to Alaska about 14,000 years
ago? Were there ancient seafarers in the Atlantic? Or is it simply
from mating between European settlers and American Indians after
Columbus? Neither, as it happens.

Modern DNA could not resolve these issues, but ancient DNA
provides answers. Eske Willerslev’s research group at the
University of Copenhagen, working with Russian scientists, read the
genomes of two bits of human remains found near Lake Baikal in
Siberia; one of these individuals lived 24,000 years ago, the other
17,000.

Both had genes similar to modern Europeans and modern American
Indians but distinct from modern Siberians or other East Asians. As
the researchers say in a paper published early last year in
Nature, this implies that a population of hunter-gatherers
lived in northern Eurasia in the last ice age and partly gave rise
to the first Americans in the East and to Europeans in the West,
before they themselves died out in Siberia and were replaced by
immigrants from elsewhere in Asia.

This may help to explain the enigma known as Kennewick Man, a
9,000-year-old skeleton from Washington state, which seems to have
features more like those of a modern European than of a modern
American Indian. The earliest inhabitants of the Americas seem to
have been distant cousins of Europeans, connected through Siberia,
with their genes later diluted by other Asians migrating through
Alaska.

As this example shows, one of the common themes of research on
ancient DNA is that the mixing of native and immigrant populations
happened much more often than previously suspected. The new
research allows us to identify the many different elements of that
complex history. It is like watching a cake being
reverse-engineered into flour, sugar, eggs, milk and its other
ingredients. The familiar textbook notion that, for most of human
existence, people native to one region developed in isolation from
those native to a different region no longer makes sense.

A long-running debate in archaeology revolves around how to
explain such key events as the advent of agriculture or the
replacement of a certain type of tool by another. The key divide is
over what caused the change: Did hunter-gatherers take up farming,
or did farmers move in and replace hunter-gatherers? This is
sometimes called the “pots versus people” debate.

Geneticists studying the genes of people alive today have leaned
toward theories based on “serial founder effects” rather than on
mass migrations. The idea is that while most people stayed put,
small groups of farmers would have moved short distances and
started new colonies, which would then have expanded. This would
account for the fact that the further from Africa a population
lies, the lower is its genetic diversity: The populations had been
through a series of genetic bottlenecks caused by small numbers of
founders.

The study of ancient DNA has challenged this view. We now know
that mass migrations occurred repeatedly, overwhelming natives
while absorbing some of their genes. In a study published in 2009
in the journal Science, analysis of ancient DNA by Joachim Burger
and Barbara Bramanti of Johannes Gutenberg University in Mainz,
Germany, and Mark Thomas at University College London, showed that
the first farmers of central Europe could not have been descended
solely from their hunter-gatherer forerunners.

In response to such research and to their own findings, Joseph
Pickrell of Columbia University and David Reich of
Harvard University argue that “major upheavals” of human population
have been “overwriting” the genetic history of the past 50,000
years. The result, they say, is that “present-day inhabitants of
many places in the world are rarely related in a simple manner to
the more ancient peoples of the same region.” In short, we are none
of us natives or purebred.

Perhaps the most striking example of this is a discovery
announced by Dr. Reich’s team in a paper recently published in
Nature: Just 4,500 years ago, long after the arrival of farming
in Europe from the near East—a transition that had largely
displaced the genes of the indigenous hunter-gatherers—another
“massive migration into the heartland of Europe from its eastern
periphery” occurred. People from the steppes northeast of the Black
Sea swamped the European genome with their DNA, and that relatively
new pool of DNA is still ubiquitous among Europeans today.

This tips the balance in another long-running argument among
anthropologists about the origin of the “Indo-European” languages.
From Irish to Sanskrit, there are close similarities of vocabulary
among most of the languages of Europe and those spoken in parts of
Central Asia, Iran and India—connections not shared by languages
like Basque, Turkish, Arabic, Hungarian and Finnish.

Two main rival theories have been offered to explain this
pattern. The first holds that proto-Indo-European was spoken by the
first farmers who left the fertile crescent of Syria, Turkey and
Iraq for adjacent regions. The second view is that the foundational
language was spoken not by these early farmers but, as certain
shared words seem to suggest, by horse-riding sheep and cattle
herders who spilled out of the Ukrainian steppe a few thousand
years later.

The recent research of Dr. Reich and his colleagues supports
this latter hypothesis: Indo-European languages probably originated
in the steppes just two millennia before the Christian era.

The discovery of the massive migration from the steppes 4,500
years ago was made possible by the analysis of DNA from 69
different individual bodies from between 8,000 and 3,000 years ago
and the comparison of nearly 400,000 different sections on their
genomes. This sort of massive analysis would have been impossible
just a decade ago, but since the advent of low-cost,
high-throughput DNA sequencing, as well as advances in statistical
analysis, it is now almost routine.

Before these technical innovations, reading DNA required the
laborious amplification of short segments, one at a time. By 2008,
companies such as 454 Life Sciences in Branford, Conn., and the San
Diego-based Illumina began marketing machines that could read
millions of DNA samples in parallel. In the past, researchers
wanting to study ancient or modern DNA had to sip from raindrops;
now they can drink from fire hoses.

For now, such work can only be done in a few laboratories—not
just because the sequencing requires big machines but also because
the procedures needed to avoid contamination of ancient samples by
modern DNA are elaborate and expensive, to say nothing of the
skills required to analyze the massive amounts of data produced. As
a result, says Greger Larson, head of a new ancient-DNA research
group at Oxford University, scientists are conducting this work not
at many different laboratories but in huge teams gathered around
the leading experts in the field, such as David Reich at Harvard
Medical School, Eske Willerslev of the University of Copenhagen
or Svante Pääbo of the Max
Planck Institute for Evolutionary Anthropology in Leipzig.

Dr. Pääbo is best known for his achievement in sequencing the
Neanderthal genome in 2009 and for his discovery that a small
amount (up to 4%) of Neanderthal DNA is found in modern Europeans
and other non-Africans. This suggests that when African emigrants
overwhelmed the Neanderthal populations of Europe and western Asia
some 40,000 to 30,000 years ago, they interbred with them to some
small extent—thus anticipating the scenarios of admixture described
by studies of later waves of migration.

In 2010, Dr. Pääbo and his colleagues startled the world again
by discovering (from the DNA in a 50,000-year-old finger bone found
in a cave at Denisova in the mountains of western Siberia) that a
hitherto unsuspected third type of early human lived in Asia at
this time. These “Denisovans” are as distantly related to the
Neanderthals as they are to us “Africans.” A small amount (up to
6%) of their DNA survives in the genomes of Melanesians and
Australian aborigines, which suggests that somewhere on their way
east from Africa, probably in southeast Asia, modern humans mated
occasionally with Denisovans.

Now comes evidence that Tibetans also have a Denisovan
connection. In the thin air of the Tibetan plateau, the local
people can survive only because of specially evolved versions of a
gene called EPAS1. In a study published last summer
in Nature, Emilia Huerta-Sánchez and Rasmus Nielsen of the
University of California, Berkeley, and their colleagues found this
version of the DNA sequence around EPAS1 in the ancient genetic
material of the Denisovans. Mating with Denisovans seems to have
enabled people to survive at high elevations in Tibet.

Ancient DNA is telling us, in short, not only who mated with
whom and when but which genes were then promoted by natural
selection in the resulting offspring to improve their chances of
survival. As Dr. Thomas of University College London points out,
changes in the frequency of particular DNA sequences are the stuff
of evolution itself. Directly measuring how DNA changed over time,
by comparing samples from different periods of human history,
allows us to see evolution not in the survival rates of organisms
(that is, through a middleman of sorts) but in genetic material
itself.

Consider, for example, the invention of farming in Europe about
8,500 years ago, a shift that caused rapid evolutionary change in
the genes of Europeans as they adapted to new diets, new pathogens
and new social structures. Some of this can be inferred from the
study of modern DNA, but ancient DNA can catch it in the act.

A forthcoming paper by Dr.
Reich’s group looks at 83 individuals from the period before,
during and after the arrival of agriculture. The study analyzes
300,000 different sections of their genomes and pinpoints just five
genes that changed rapidly.

The strongest signal came from the mutation for lactase
persistence—that is, the ability to continue digesting the milk
sugar lactose after infancy. Normally, mammals don’t need to digest
lactose as adults, and the necessary lactase gene switches off when
a baby is weaned from its mother.

This changed for human beings, however, when dairy farming
introduced milk into the adult diet. A mutation that prevented the
weaning switch-off spread in Europeans fairly late, around 4,300
years ago, probably long after dairy farming was invented, but it
gave its possessors a significant advantage: They derived more
nutrition from drinking milk (and suffered less indigestion) than
their rivals.

Two genes that affect skin color were also subject to rapid
evolutionary selection as early farmers tried to subsist on
grain-rich, vitamin-D-poor diets in northern areas with low levels
of sunlight. (Sunlight helps the body to convert a form of
cholesterol into a form of vitamin D.) The shift to pale skin—which
produces vitamin D more efficiently than darker skin—among northern
Europeans after the advent of farming appears to have proceeded
rapidly, pointing to some of the strongest selection pressures
ever recorded in human genetics.

Since the discovery of DNA’s structure more than a half-century
ago, genetic science has promised—and begun to deliver—a medical
revolution, but it keeps producing other kinds of revolutions too.
In the 1990s, it transformed the field of forensics, for example,
and now it is having a similar effect on history and archaeology.
Today, the prehistory of humanity is an open book as never
before.

The lessons of this DNA revolution are not just scientific,
however; they are social and political as well. The discoveries
made possible by our new access to ancient DNA show that very few
people today live anywhere near where their distant ancestors
lived. Virtually no one on the planet is a true native—an
instructive fact to consider at a time when ethnic and national
differences still abound and the world continues to throw human
beings together in new and unexpected ways.