Natural Selection, Genetic Variation and Magic Populations - Dr Paula Kover

The Director of the Milner Centre for Evolution, Professor Turi King, talks to Dr Paula Kover whose research focuses on understanding how an organism's environment affects the evolution of traits under natural selection.

Episode Transcript

(00:02):
Hello and welcome. You are listening to a podcast by the Milner Centre for Evolution at
the University of Bath. I'm Professor Turi King, your host, and today I'm talking to Paula Kover,
whose research focuses on understanding how an organism's environment affects the evolution
of traits under natural selection.So, Paula, you're looking at one of

(00:27):
the pillars of our understanding of evolution, natural selection. So, talk me through what is
natural selection, for somebody who doesn't know.Well, first thing you have to have variation on
a characteristic or a trait. If you don't have any variation, then selection can do nothing.
As I always tell my students, maybe there's an incredible advantage to having three eyes,

(00:48):
but we just don't have the phenotypic or genotypic variation to respond to that advantage.
So, selection first thing is the fact that the environment you live, the conditions that you live
might favour a certain trait, whether you respond to that or not, it depends whether you, as I said,
genetic variation and phenotypic variation. And if you have plenty, then natural selection will

(01:11):
favour the ones that have the trait that increase your fitness, which often we think about as having
more babies, but we have to be careful. It's not just more babies, but good babies,
healthy babies that are also going to give you grandchildren and so forth and so on.
And if that's the case, then the genes that are responsible for causing that trait will be
increasing in frequency over time. Because every single time that you have a whole population,

(01:35):
whoever had the genes that gives that traits will increase the chances of making more babies. So
therefore, over time, you're going to see an increase on individuals with that trait
and therefore an increase in alleles, which is the little bit of DNA that encodes that trait,
also increasing on the population.So, what were you studying when
you were doing your PhD?Ah, yes. I started working on

(02:00):
plants and pathogens. I did my PhD at Indiana Bloomington, and I met Curt
Lively. Curt Lively has been working on the Red Queen hypothesis forever, and I thought this is
the most fascinating thing ever. And the idea here is that sometimes populations have to be
constantly evolving just to stay in the same place, when they're trying to fight pathogens.

(02:25):
So, pathogens are constantly trying to adapt to the host. And of course, the host is always trying
to change. But if they're both evolving relatively fast and the pathogen tends to evolve a bit
faster, because their generation time is smaller and they have a bit more genetic variation, then
they're always on top of the host. So, the host is constantly evolving, but it's going nowhere

(02:46):
because the pathogen has managed to catch up with it. And I thought that was a fascinating idea.
And also, the Red Queen hypothesis is one of the explanations of why organisms actually
have sexual reproduction, which is another very, very big enigma in evolutionary biology.
So, if you think about it, about what I just said, natural selection should always be favouring the

(03:11):
best genotype. So, you have the best genotype, you've managed to attract 20 mates, and you can
have lots of little babies. But every time you make a baby, if you have sexual reproduction,
you're kind of destroying that perfect genotype and combining with someone else.
So why would you do that if you just had won the lottery and have the best genotype ever? And it's

(03:34):
difficult to explain, and we're still trying to explain why organisms do that. There's plenty of
organisms that are successful without sexual reproduction. So why is it that we'd do it?
And the Red Queen hypothesis is one potential explanation for that, because it says that,
you know, you have to create genetic variation to fight the pathogens, and the pathogens very
quickly evolve to you. So, you can never stay comfortably at your evolved place because you

(03:58):
have to constantly be responding because a pathogen is constantly coming after you.
So that's the idea of the Red Queen hypothesis, which has been around forever,
and it has been like everything else demonstrated in bits and pieces. But my
dream from ever was to have a population where I knew everything about the genes, and I could
throw pathogens at it, and I could see the Red Queen hypothesis happening in front of my eyes.

(04:22):
And that's how I create the magic population, which is one of the things that has the most
citations of all the work I have done, which is a population of Arabidopsis, which has a lot of
genetic variation, and it's completely sequenced. And you can use that to do a lot of natural
selection experiments and see how populations are responding to selection at the genetic level.
So, tell me about your research.Right. So, what is really exciting

(04:46):
me at the moment is trying to understand leaf shape. So, plants stay in place, and they get
their food by photosynthesizing. Leaves are the major way in which they can fixate carbon. So,
if we think about what natural selection is, there should be very strong selection on having

(05:08):
the most perfect leaf that you can possibly have, because it's key for your survival.
So, it's okay to understand that different plant species, because they live in different places,
they might have different leaf shapes as being ideal. But why would you have in one population
plants with different leaf shapes? That is very puzzling. And you don't see that very much,

(05:34):
most of the time the variation in leaf shape is between species. But there's this one species
that I'm really interested in studying, which is a close relative to Arabidopsis,
that have an incredible variation in leaf shape.And a lot of people think that if we understand
which leaf shapes are more adapted to certain climates, we could change,

(05:54):
for example, the shape of leaves in crops. It's not difficult to change leaf shape genetically,
but we don't really understand what is the consequences of these different leaf shapes. So,
we're trying to use this plant called Cakile maritima that has spread all the way from the
Mediterranean to Iceland and has this leaf shape variation across all the populations.

(06:17):
So, we're doing crosses with them and trying to find what the genetic basis of it is. And we also
trying to use photographic recordings, so we don't contribute to the carbon cost of fieldwork. And
also trying to find out if we can do more work with everybody's pictures of plants everywhere.
And trying to see if 1) If different leaf shapes are more common in different parts of the world,

(06:38):
and 2) What is it about the leaf shape that makes them better or worse for the plants?
This is an ongoing work, but we do know that contrary to our first hypothesis,
it's not environmentally determined. So, if you are a plant mother that makes a leaf that is
completely divided, then you have babies that also make leaves like that, no matter what environment

(07:00):
you grow in. And if you have completely undivided leaf, your babies will do exactly the
same thing no matter what environment you grow.So, it's clearly genetically determined. And for
some reason there is no very strong selection or there some inability to respond to that selection
for some reason, and we see this variation.So, is there any pattern whatsoever? Is there

(07:22):
a hypothesis as to what's going on here?There's some very strong hypothesis. There
is even like theories repeated on and on and on that, more divided leaves are supposed to keep
the leaf cooler, better than whole leaves. So, the expectation is that warmer parts
of the world will have more divided leaves, and cooler parts of the world has the complete leaf.

(07:48):
Now, nobody has been able to show empirically that that's the case, which is why I'm really
interested on it, because most of the variation is between species. And when you look at between
species, there's other things playing a role on what leaf shape they have.
So, this particular species, because it is the same species going all the way from Iceland to

(08:08):
Spain, offers a particularly interesting situation in which you can actually test that. It's very
cool, people have built leaves off like very thin metal material, and put into a chamber
of a certain temperature, with a certain wind and measure all the physical capabilities of leaves.
So, under those conditions, that's what you would expect. But yet that's not what you see in nature.

(08:33):
So, we are trying to do now reciprocal transplant experiments and put these plants
in different parts of the world and seeing if we can measure directly whether having a
certain leaf shape can definitely improve your fitness on hot versus cold climates.
And I suppose the idea is that eventually, with this knowledge, you'll be able to apply it to

(08:54):
our own crops to make better decisions about what crops we grow or potentially change leaf
shape to be better, or this kind of thing.That's right. So, it's really interesting
because a lot of the times when you're trying to solve a crop problem, you tend to think about
the very specific problems. So, for example, people do research to try to find what genes

(09:15):
are turned on when a plant is under drought, and then try to modify that pathway. But it
could be that you could solve this problem at a much basal level. So, the plant not even get to
the point of experiencing drought by changing the leaf shape. And it would be something that
occurs somewhat naturally in most species or genera. So, I mean, not that I think it is a

(09:36):
problem, but I think it would be much more acceptable for the public to have
a change like that than some GM specific trait.Yeah. So, let's chat about that, because sometimes
people get worried about this sort of thing. What's involved in terms of changing a leaf shape?
So, in Brassica which is the family that I work. So, brassicas are your cabbages, your broccolis,

(09:58):
your cauliflower. And for those of you that don't know this is all the same species. So,
Brassica oleracea is the same species, whether you eat the cauliflower,
the broccoli, the cabbage, the brussel sprouts, whatever it is. So, you can see that that
species has a lot of phenotypic variation.And if you want to change the leaf shape,
you can cross with some very close ancestral form and change the leaf shape or the flower

(10:22):
shape as you have already seen with all these. So, it doesn't require an introduction by
bacteria of a gene that only exists on a completely different species. It's just
about a process of selection for increasing the dividedness of a leaf or reducing it as you need.
Now you've got a new paper coming out soon. So, what is it that you've been

(10:45):
researching and what did you find?I mentioned that I've been working
with Arabidopsis. And Arabidopsis is a great organism because we have sequenced many times,
that means we know all the genetic information that Arabidopsis have. More importantly,
we have collected Arabidopsis from different parts of the world and sequenced them also,

(11:08):
which means we know where on these instructions there is some variation.
Now, the best way to actually identify the genetic basis of any trait is to be doing crosses. But
it’s really time consuming to do a lot of crosses. The other thing you can do, which we
do with humans, which we cannot convince to cross as we like, is to do something, maybe some of you

(11:35):
have done, which is the ancestry DNA, right?So, you can send your DNA out and you can
identify whether you have variation. And then you can see, okay, everybody that has diabetes
have this genetic variant. Everybody that doesn't have diabetes doesn't have that genetic variant.
This is called an association analysis.So, in plants you can cross, and you can

(11:58):
use a method called QTL analysis, which is when you cross you try to see which
piece of DNA goes with which trait.Now there has been a debate for a very
long time about what is the best way to find the genetic basis of a trait. And what I've done about
ten years ago, is I create what we call a magic line. A magic line is a cross between different

(12:24):
varieties of Arabidopsis. We put all that together into one population, and we know exactly what
the genetic variation for each individual is.So, then you can measure whatever trait you want,
and you don't have to de novo genotype or sequence every plant. So that has become
something that was very difficult to do. It took me like three years to do, and then it took like

(12:49):
ten years to do on wheat and then another five years to do in barley, but is amazing. Nowadays,
most crops have come to the conclusion that building a magic line is the way to go.
But when you do a magic line, you are using these crosses. And because you are crossing
these plants, the pieces of DNA are divided every time you do a cross. But it takes a long

(13:16):
time to become a very, very, very small piece of DNA. You would have to break that lots of lots,
a lot of times, right? So even though you can say, oh, I think there is some
genetic information on this bit of DNA that is important to determine when I plant the flower,
it is still kind of a vast amounts of information.Now that other method I mentioned to you,
the association method, you know, it's a much more pinpointed. You'll find like there is a mutation

(13:41):
here and everybody that has that mutation has that trait. So, what I've done on this paper is
that we show that you can use the magic line that I created to do both, to do QTL and association.
And it’s great because you can find much more variation that you do if you use association
alone. Because strangely, by keeping track of these genotypes, there's more information than

(14:06):
not knowing and treating everybody as if they are random. And doing this by combining both of these
approaches, QTL and association, we were able to find a new genetic variant that explains variation
in accumulation of metals in plants.So, it's more a methodological paper

(14:26):
to show that we can actually find the genetic basis of some really complex
traits relatively fast, using the magic lines.So, what really strikes me about your research
is not only are you developing methodologies to find these particularly interesting genes, but
then there's implications for use to help us in terms of, sort of, crops and this kind of thing.

(14:54):
That's the hope, yes. And I think people are becoming more and more aware of the importance
of not growing a single variety. And I think the magic lines are particularly powerful for this.
It's showing that you don't have to, again, genetically modified that one variety that
you like to create the trait. You could create a population, select for it, and has a little bit of

(15:14):
genetic variation that would allow you to do more than one thing at once and create a little bit
more variation in the field. And you know, that should keep pests at bay for a little bit, but
also it provides more an interesting environment for other beasts that should be cohabiting with
you. And when you have like a single genotype out there, sometimes it's very difficult to do so.

(15:35):
So, what's next for you?So, I have been really interested
recently on truly multidisciplinary, and I have happened to stumble upon a project with
my chemical engineer colleague, which is this idea that maybe we can tell when the soil is happy and

(15:57):
when plants are happy in the soil.It is clearly a sci fi project,
but there's so many things that have been sci fi in the world that got done,
and I think it's just the time for me to get my hands dirty with some project like that.
So, we are trying to see whether we can get electric signals from an environment that has

(16:21):
a happy soil with the right nutrients, or plants that are happy, instead of having a pathogen,
that we could monitor that and be able to use that in farms to do long distance
monitoring of areas for growing crops. So, it is totally sci fi, but we're very,
very slowly just starting to see little things.So, something that we have got in recently is that

(16:47):
actually plants growing with a very low current running in the soil grow better than plants that
doesn't have that. And we have just done a very, very small experiment about that. And it feels
like such 1970s research, right, but it's true.So, we're just trying to see if we can
use something to make plants a little bit healthier under very controlled conditions.

(17:11):
But if that works, maybe we can move up to something a bit larger scale, we'll see.
So that's really interesting. So, do we know why a little bit of electric current is making the
plants… it's like tickling the roots? What is it?You know we should know right? I mean it's a 70s
question isn't it. And again, that's maybe why science is so fascinating. So,

(17:34):
the old idea was that every cell has a little bit of polarity, isn’t it, and that maybe adding a
little bit of electricity just to make the whole metabolism work a little bit faster.
But nowadays there's some ideas that actually think that the current running in the soil might
be changing, actually, the nutrients in the soil, or more importantly the big word, the

(17:56):
microbiome in the soil. So maybe it's culturing different kinds of organisms in the soil and those
bacteria they’re growing in the soil, because it has a little tiny bit of current going on,
are plants that are also beneficial for plants.So, you're going to have to wait for the next
chapter for us to tell you why exactly it is.Paula, thank you so much for talking with me.

(18:21):
This was a podcast by the Milner Centre for Evolution at the University of Bath.
I'm Turi King and thank you for listening.If you have any thoughts or comments on
this or any other episodes, please contact us through social media. For
more information about the Milner Centre for Evolution, you can visit our website.