Ancient DNA — What It Is and What It Could Be: Beth Shapiro at TEDxDeExtinction

Ancient DNA — What It Is and What It Could Be: Beth Shapiro at TEDxDeExtinction

August 31, 2019 20 By Stanley Isaacs


Translator: Carolina Becerra Merino
Reviewer: Shlomo Adam (Applause) I am a molecular paleontologist Sounds sophisticated? actually, I just made it up. I spend most of my time in the lab or in front of the computer but during the summer, I prefer to spend my days and nights in the far north, where I divide my time between waging war against mosquitoes and searching for the remains of Ice Age animals, which I bring back to the lab to extract their ancient DNA. As an early career scientist I was interested in
ancient DNA for three reasons: First, it was an oportunity to combine my interests in paleontology, geology and molecular evolution. I could get DNA from fossils and use this to watch species evolve through time. Second, it was a brand new field absolutely bursting with enthusiasm and excitement, a career in ancient DNA
promised to be different, definitely cool and hopefully useful. And third, I really wanted to go to the Arctic. So, what about the Arctic
is so good for DNA? well, it is the best place for the long term preservation of DNA, and the reason why
is quite straightforward it’s cold. And it’s been cold consistently for at least the last million years. This period was characterised
by dramatic shifts between very cold glacial intervals and much warmer interglacial periods like the one we’re in today. During the glacial periods, or ice ages, the sea level was a lot lower
than it is today, because much of the planet’s water was taken up into the glaciers
forming on the continents. The lower sea level exposed a mass of land that connected the Asian and
North American continents, this land mass was called Beringia, and it’s where I do much of my work. Now crucially Beringia was not glaciated during the last Ice Age, instead it was a rich tundra grassland that was home to an equally rich diversity of Ice Age plants and animals. Today, most of these species
are extinct, but their bones and other hard bits – tusks, teeth, hair, sometimes even full mummies, survived frozen in the ground called permafrost. We look for frozen bones wherever the permafrost is melting along riverbeds, around lakes and at active gold mining sites such as this in Canada’s Yukon territory, where the gold miners
wash away the permafrost to get to the gold there,
in gravels beneath, thousands of bones are unearthed. Over the last decade I and colleagues from around the world have extracted DNA from tens of thousands of these bones and used them to better understand how Ice Age animals were affected by climate change, in particular, the last Ice Age. We learn, for example, that bison, horses and mammoths really liked it when it was cold, this is when their populations
were the biggest, their genetic diversity, the most large. As it got warmer many of these populations
started to decline and eventually went extinct. We watched brown bears
during the last Ice Age dispersed out of Alaska, across Asia, and into Europe, presumably chasing after this
growing populations of herbivores And we’ve asked questions such as why do some species, including the cave lion,
go extinct? while others like the caribou,
continue to thrive today. And today we find ourselves in the middle
of the genomics revolution, once again with growing enthusiasm
to push this a little bit further, we find ourselves more and more frequently being asked and asking each other: can we sequence a complete genome of an extinct animal and, dare we say it, actually bring one back to life. To answer this, let me first recap, very quickly what I’m going to call the seven, badly described, but deeply over-simplified steps, to bring an extinct species back to life. First, you have to sequence the genome, and somewhere in there are the genes some of these As, Cs, Gs and Ts, the genes that code for the proteins that make the cells, the bodies
and the behaviours of the things we’re trying to bring back. Then, we have to get this genomes under the chromosomes because chromosomes are the way
that genetic material is carried in the cells, then we have to get the chromosomes
into the nucleus and the nucleus into the cell so they can start to divide and then we have to get the cell
at the embryo into a surrogate mother so that it can actually
grow up and develop and become something, and presumably her genome isn’t going to have
too much to say about this developmental process. And then, she’s got to take this developing embryo
or fetus actually to term in this case this is also not simple, there’s a very large size difference between the much large mammoth and a smaller elephant and it’s not known whether
it would actually be physically possible for a female elephant to carry a baby mammoth to term
without disaster. But assuming disaster is averted then a mammoth can be born
and learn how to be a mammoth we then have to find a place
for it to live so it can go about doing mammothy stuff and eventually make more mammoths. Pretty straightforward. Ok so, easy might be an exaggeration, or an outright lie, but let’s get not caught on semantics. So, where are we, in this, with the mammoth? The first attempt to sequence
a complete genome of a mammoth was published in 2008
in the journal Nature. The authors managed to sequence 3.3 billion base pairs of mammoth DNA. When all was put together
and mapped to the elephant it looked like about 50%
of the mammoth genome had been sequenced. 50%, that’s about half so instead of this, we’re working with this, but that’s ok, that’s great that was state of the art in 2008, and it’s pretty much
state of the art today, in fact, does anybody know
how many vertebrate genomes have been completely sequenced, including modern vertebrates,
living vertebrate species? Give you a hint, is not a very big number. None. Not even our own genome, for which an honest to goodness
ridiculous amount of sequence data has been published. We do not know the complete sequence
of our own genome. I’m being a little unfair. We do know
more than 99% of the genome, the part of our genome
that actually codes for things, where the genes live, the euchromatin. It’s the other part, the heterochromatin, that is made of tightly condensed,
repeat sequences, that are just super hard
to sequence through. The heterochromatin doesn’t make up
a very big part of our genome, and it might not actually
be that important, but we don’t know, because we haven’t managed to sequence it for a living thing. For the mammoth, we don’t know the heterochromatin, we also don’t know where the genes are, much less, where the specific genes, that make a mammoth
look like a mammoth, are, so we can find them, and use that to hybridise– So what do we do,
how do we fix this problem? Well, seems pretty clear,
let’s finish it. We go into the Arctic, we find the really well-preserved
mammoth bone, we take a chunk out of it, and bring it back to the lab,
and we sequence its genome, that’s where the fun begins. If I’m gonna sequence my own genome, I could start with, say,
a piece of my hair. And I could dissolve it off into extraction stuff, and sequence everything that was in there, and the results would look like this. Pretty much everything
inside that DNA extract would be me, my own DNA, that’s because I’m alive, my DNA is in good condition, my hair hasn’t been anywhere
particularly weird. Not so for the mammoth. A similar experiment
performed on the mammoth bone from Alaska, resulted in about 50%,
a little more than half of the sequences in that bone
being mammoth. The other half was a combination
of plants, bacteria, other stuff that had gotten into the bone while it was in the soil, and stuff that might have been
introduced into the bone during the process of
excavation and sequencing, we touch the bone
and breath on the sample. But this is still pretty good, permafrost preserved more that 50%. Here’s a similar experiment, for a Neanderthal,
from a cave in Croatia. Here, only 3% of the sequences
in that extract were actually primate, and that includes both
Neanderthal sequences and any potentially
contaminating human sequences, but this is still good, there’s still Neanderthal DNA
in this sample, just not much of it, so presumably, what that means is all we have to do is sequence a lot more of it. And to some extent,
that’s absolutely true, but there is another problem. You imagine that our own DNA is like a rope, or a ribbon, or party streamers,
say like those we’re gonna put up to celebrate the birth
of our first cloned mammoth. Then a mammoth DNA itself
is actually going to be more like… confetti… that’s been run over
by a herd of mammoths… in the rain. Water, oxygen, solar radiaton, bacterial decay, all of these things will begin to act
on the long strands of DNA immediately after an organism dies. And, eventually, that DNA will be chopped down into smaller and smaller fragments, until there’s nothing left. So, this
is what we have to make this, to make these, and eventually that. It’s a hard problem, it’s a problem that probably
won’t be solved without new and different biotechnology than what’s available today. But as we progress toward this goal, we’re going to learn a tremendous amount about these extinct animals, knowledge that I have no doubt will help us to win battles
against the extinction that are happening in our world today. We’ll learn where the genes are, we’ll learn what genes
make them look and act the way they did, we’ll learn a lot more about how genes
interact with the environment. And, we’ll finally, hopefully, discover why some populations and some species are so much more
susceptible to extinction than others are. At the same time,
we would get better at extracting DNA from our ancient — the ancient bones. We will learn how to fix the broken bits so we can start piecing together
these little tiny fragments into longer and longer fragments. If it’s what we want to do, we will eventually be able to sequence the complete genome of an extinct animal. And then we will have completed step one. Thank you. (Applause)