(dramatic electronic music) (audience applauds) - Thank you. (laughs) Well, thank you, Allison. And thank you to the
Society for this award. So don't worry, I'm not gonna be
talking all about my research, only a little bit of it tonight. But what I'm gonna be talking about is the fact that life is complicated. And this isn't sort of a
statement of existential angst, it's a statement of what we
see when we look at biology. Biological systems are complicated and that level of complication exists at the level of molecules
that build the cells and maintain the cells, but all the way up to the
interaction of individuals within populations and the
societies that they form. And when we see complex things in biology, we almost immediately
reach for Darwin, okay? When we do that, we typically look at a group
of organisms like this. So if we look at the visual system, if we look at the evolution
of the visual system, there are four examples of organisms which will react to light, okay? So in the top left there is
a single-celled organism, Euglena. And if you can make out
that little red patch that's a light-sensitive patch, but this is a single cell
that can respond to light. Below that is a flatworm. I think the flatworms
have the derpiest eyes in the animal kingdom. They're permanently
cross-eyed. It's a simple cup. And so it's a simple
eye light-sensitive cup. And then on the right-hand
side, we have two predators which have highly
developed visual systems, including the cephalopod
having a camera eye. And we look at these things and we see, well, this is the product of selection. This is selection operating
on simpler precursors, generating more complex
forms that are better fitted for the lives that these organisms lead. And that's exactly correct. But the other reason where we can know that we're dealing with
complexity is we can say, "Well, why do the organisms "have the level of
complexity that they have?" So the flatworm for instance,
doesn't need a camera eye. Camera eyes are expensive, so evolution is not going to go to the trouble of building something
that is expensive to run. How do we know that? Well, we can look at cave fish. So fish that live in the absence of light, their ancestors originally lived
in light, so they had eyes. And they've lost the ability. For various reasons, they've
lost the ability to make eyes for the simple reason that
eyes are complex things. If you're living in the
dark, you don't need eyes and so you have a significant advantage, a significant reproductive advantage, if you can avoid that cost. And people have estimated that juvenile fish expend about 15%
of their resting metabolism on maintaining their visual system. That's about the equivalent to what it is for the human brain in terms of percentage energy costs. So we are familiar with this. I mean, a lot of you
will be going, "Oh God, "I hope the rest of the
talk's not gonna be this dull. "I know this stuff. I
learned this in high school." What I want to do is present
the flip side to this, okay? So we're all familiar, yes, with selection and natural selection, but we're often not familiar is the ability of natural
selection to do its work is dependent upon the
number of individuals in the population that we're looking at. And I want to explain, spend a bit of time
explaining why that is. And in order to do that, we have to drill down to first principles. So here's the first of three
quotes from a guy Matthew Hahn, who's a population geneticist. And I like this quote because as he said, "Evolution begins as one
mutation, on one chromosome, "in one individual." Now it's not explicit, it's implicit. In one individual. That individual is part of a population. And in order for the fate of
that mutation to be decided, we have to take into account the other individuals that
don't have that mutation and how many of them there are. So we have to delve into
population geneticists. Sorry, don't delve into
population geneticists, delve into population genetics. And this is another famous quote. And what I like about it, it's a reworking of a more
famous quote from Dobzhansky, the original quote being that, "Nothing in biology makes sense
in the light of evolution." Michael Lynch, who's effectively one of the leading
population geneticists, says, "Well, no, nothing in
evolution makes sense "except in the light of
population genetics." So I'm gonna give you all a crash course in population genetics. Just briefly to warn you, about 10 years ago, it may be a little bit older, further ago, I took over teaching population
genetics in Aberdeen. And I did a bit of research to see how I could maybe
teach population genetics, what the students' previous views were, and I found a whole load of
assessments from previous years. And population genetics scored the worst in student assessments. One of the comments was,
"Far too much maths." Ah, that's probably fair. One of the other comments
was really odd though. It was, "Boring. Too many beavers." (audience laughs) There won't be any beavers, but there will be population genetics. And there won't be any
math, there'll be be graphs. So what I want to tell you
about is why do population, why does the size of the population, determine the fate of
a particular mutation? We have a single mutation
arising in one individual. Why does the number of individuals in that population that the
individual finds itself in, why is that a factor? And it's because of genetic drift. So you're all familiar with the idea that we've all inherited two
copies of our genetic code. One from our mother, one from our father, except for the X and the Y
and the mitochondrial DNA, which we'll just ignore. And you'll pass on one
of those two copies. Yes, mixed up, but we're
not gonna go into that. So what we have here then is two individuals who are
gonna produce offspring. And they have two different versions. They both have the same
pattern of difference. One has a red version and
one has a blue version, and there's a 50% chance that
they're gonna pass that on. And they can't pass on both. So they have to choose. Well, they don't choose, but only one is going to make it through and that is where you're seeing the first instances of genetic drift. So they make a choice. And the individual, the offspring, inherits the two blue versions. And if this was our very,
very tiny population, what effectively has happened is the red version has been
lost in the next generation. So it's gone. The blue version is the only thing that's present in this generation, so it's been what we call fixed. Now obviously this is
a ridiculous situation 'cause this is not a population. So let's look at some more
simulations doing the same thing. So the same thing is happening that I showed you on the previous slide, just on a slightly different scale. So this is from a simulation
from Graham Coop's lab. And we're seeing on the side there, on the left-hand side there, we've got 10 individuals. They're each the same as
the one that I showed you in the previous slide. So they each have one
red copy, one blue copy. And we're going to perform a
simulation with random mating. Now at this stage, the red and the blue copies
are equivalent to one another. There's no selection going on. The only thing that determines what happens over the next generations is genetic drift, okay? The random sampling of one or
other of those chromosomes. And below that is a graph showing the frequency
of the red or the blue represented by the line in the population. So you see, we start out, all the individuals are
what we call heterozygous, so the frequency is 50%. But you'll see as we run the simulation, very quickly the red population crashes or the number of red variants drops until it disappears from the population. And now the blue variant obviously takes over the population. We say again, just to remind you, the blue variant has now been fixed. Okay? Now this is a population size of 10. So I'm gonna run this
exact same simulation, exact same conditions, but we're gonna increase
the number of individuals in the population. It's constant across the generation, so we're gonna keep it stuck at 100. And you can see, I think very clearly, neither of those variants
has gone extinct, okay? They're being maintained more
or less at the same level over those different generations
all the way up to today in this simulation. Okay? All that's changed
is the population size. So this is telling you straightaway that the population size can determine what the fates of
individual genetic variants. This itself is an important thing, but what's even more important is what happens if you introduce
selection into the story. So I'm gonna make it relatively easy. We've got the two blue variants, you have 1,000 offspring. If you have a blue and a red variant, you have 1,000 offspring. If you have two red variants, on average you will only
have 999 offspring, okay? So you have a reproductive disadvantage. So I'm gonna run a simulation slightly differently this time though. What I'm gonna do is run the simulation with 10,000 individuals
over many generations. This is over 100 generations.
I think it's 400 generations. And we've got five different populations that start out the same. Okay? They don't differ from one another. So the only differences
are stochastic differences caused by the random
sampling at each generation. And you can see very clearly
that in all the populations, the number of red variants goes down. And you're all going "Well, yeah, "'cause you've already told me "they have a reproductive disadvantage." And importantly, I hope. Well, maybe the colors
don't come out well on this, but we'll manage. There's a blue line running through that which shows what happens if you
have an infinite population. So now let's look at 1,000
individuals, same principles. You can see there's a lot more noise because with smaller
numbers of individuals, the random sampling creates
greater fluctuations. The proportional effect is
stronger in a smaller population. But generally speaking, there's still a downward
trajectory for the red variant. We're not monitoring the blue variant. I should have pointed that out. Now we go to 100 individuals. Now you see something quite startling. You can see the blue line I hope, which represents the infinite population just going down smoothly. But you'll see in only two cases has the red variant
disappeared, been lost. In two cases, it sort of
fluctuates up and down. But by 400 generations, it's still there even though it's conferring a reproductive
disadvantage, okay? And then the green
population at the top there, the red variant has become fixed. Every individual in that
population has the red variant. They started out having the blue variant, which gave you a
reproductive disadvantage. They've lost it and they've lost it because of the effects
of genetic drift, okay? This is one of the most profound outcomes of population genetics. The fact that the fate of an
individual genetic variant is dependent in part on the impact that that variant will have on the biology. But the fate is already also determined by the number of individuals
in the population that that variant finds itself, okay? And this is, I think, a relatively underappreciated phenomenon. Certainly often as biologists, and I know I speak as one of the guilty, before I started teaching
population genetics, I assumed that everything
that we see out there at all levels of hierarchy and biology was there because selection put it there. In actual fact, a better way
to think about genetic drift is, as Holly Dunsworth has said, "It's the survival of that
that doesn't suck too badly." In other words, as long
as your genetic variant doesn't cause too much of a phenotype and your population is sufficiently small, it can survive. So this is quite a striking thing. And one of the things
that this is so striking is when we look at the level of genomes, which is what the rest of
the talk is gonna be about, we see the really strong
impact of genetic drift or the survival of that
doesn't suck too badly. So again, a lot of you were going, "Oh God, he's put that up." Just to remind you the
flow of information. So we have this lovely elegant model from decades of research which mapped the flow
of linear information encoded in the letters of DNA, the four chemicals that make
up DNA in our genetic code. That has to be converted via an enzymatic series
of enzymatic reactions into a messenger RNA. And that messenger RNA is
effectively the cousin of DNA. And that messenger RNA is
then used as the substrate to burn that linear array of nucleotides, those chemicals, into a
linear array of amino acids. This is a beautiful picture. But by the mid-70s, people studying things
that are not like bacteria, which is what were used to work out a lot of the genetic code, people studying things more like us found this rather disappointing fact. So what's shown here is one
particular human gene to scale. The things that are labeled exons, those are the bits that code for protein just in the slide that
I've just shown you. And they're interrupted
by these bits of DNA which I've labeled as introns, which don't code for anything. Worse than that, they have to be removed before you can do the important
bit of converting the RNA into the protein. Why would you do this? This seems nonsensical. And if you just look at the size for this particular human gene, down at the bottom there,
that little orange rectangle, that's the size of the messenger RNA, the region that codes for protein. You can see most of the gene consists of regions that have
to be removed, the introns, before you can do
anything sensible with it. So where have they come from? Well again, we could be thinking,
"Ah, there's a lot of them "and they're big. "They're probably doing
something really important." And you'd actually be right because now they will be
doing something important. But that's not the question
I'm asking is that, where did they come from? Did they come fully fledged into life doing something important? And in fact, if we look at the eukaryotes, so I like this is an evolutionary
tree of the eukaryotes. It's actually hard to define eukaryotes. Basically things that aren't prokaryotes. Oh, that's not very helpful. If you're looking for yourself,
by the way, there you are. That's the animals. The eukaryotes are incredibly diverse. Most of them are single-celled organisms. They will not appear on
David Attenborough shows. But all of these organisms
have one thing in common. They all have some of these introns, these regions interrupting
their coding regions. And the answer to why they're there goes back right to the very beginning. That stem going down that's labeled eukaryotes into prokaryotes. Because the other rather startling thing, which is maybe not so startling now, I think many books, many
popular science books, have been written about it, is that eukaryotes are a union between two types of prokaryotes. The alphaproteobacteria, shown there in red. And in blue, they're beautifully named Asgard archaea bacteria. And effectively they
came together in a union. The red alphaproteobacteria
became the mitochondria, the thing that powers the cells, and the rest of the cell was
contributed by the archaea. And this union was not necessarily equal. This small alphaproteobacteria
getting swallowed by what possibly was a larger cell. We don't really know this, of course. This is rather fanciful, this drawing. But the important thing is
that the alphaproteobacteria and these archaeal-like cells
had these small entities. Which I've labeled there
as Group II introns, okay? So those red entities there
that sit inside the DNA and you can see them today
in modern alphaproteobacteria and archaea. The key thing was, and for
reasons that we don't understand, but bear with me 'cause we
think we've got a good reason is when this happened, when
this union happened, the Group II introns just went crazy. So they leapt from their original location because these things are able
of moving around in the DNA in which they find themselves, and they just landed everywhere. Calculations suggested that nearly every gene would've been hit. And when I say hit, I mean interrupted. So what's shown there is the
coding region being interrupted by this enormous Group II
intron just landing inside it and obviously disrupting it. So that creates a problem. We think that the last common ancestor, that beautifully named LECA there, the last eukaryote common ancestor, would've had an
infestation of these things that it didn't really have
the mechanism to deal with. Now we think one of the explanations, we don't know the exact explanation, but we think one of the
explanations might be something to do with population genetics. So Group II introns do have
this beautiful property is that although they do land in the DNA, and they interrupt the coding region, when they get converted
into RNA like that, they very usefully have the
ability to remove themselves. Now this is a very simple
version of what could be a very complex cartoon. The only thing you need to
take home from this is that that looping structure between the what I've labeled the exons, is there to represent this innate ability that Group II introns have to excise themselves from the transcript so they're effectively
doing no harm, right? And if you've only got a
few of these in your genome, that's probably true. But you've got the early eukaryote, our ancestor had
thousands of these things. And worse than that,
each one of these things, I've only shown a few of
the signals that you need in order to be removed, those things that I've
highlighted there in blue. There's more than that. There's at least 200 different sites that needed to be preserved if you're going to remove
that intron by self-splicing. That's what Michael Lynch
calls, "Mutational fragility." You've created a situation where there are 200 times the number
of sites that can be mutated to prevent the flow of
information from DNA to protein. This is not a great place to be, okay? Now if you had high population numbers, like the bacteria that were effectively came together as a union, that's not a problem, okay? So what I've shown here
is three different cells that are being infested
with these introns, shown there in red, and below that I've shown
you the different dosages, if you like, of this infestation. Well, population tells us that you've now got differential
reproductive ability. Because the more infestation you've got, the more likely is one
of those will go wrong and doom that cell and that cell will be
removed from the population. But its neighbors that
have fewer infestations, such as the one there over on the right, has a better survival chance, okay? So if as long as your
population is high enough, selection can operate. But if your population is too small and you have genetic drift, selection will not be able to remove these and many of them will
start to go to fixation. Just as I've shown you
in those simulations at the beginning. That unfortunately is where
our ancestor found itself. But it dealt with it. And we know that because I'm standing here telling you this story, okay? How did it deal with it? Well, one thing we know is that because it had a low
effective population size, most of these infestation introns
would've gone to fixation. That in itself, as it recovered
because it surely did, would've created selection pressure for a mechanism that would deal with that. And one of the things that we know about all eukaryotic cells is
that they have a machine, an incredibly complex machine. The human's machine has 200 components, more than 200 components. And this machine exists to help
those Group II introns that, if you like, have done the equivalent of fallen over and can't get up anymore, okay? They've become damaged
by mutational processes and they can no longer remove themselves. So the eukaryotes had to
invent a mechanism to do that. It was either that or no more eukaryotes. So you can see this tension
that would've existed. And it's no surprise then that, I mean, this could have been a
cataclysmic end to the eukaryotes. But they gradually picked
up, they built this machine, and this machine now
exists in all eukaryotes. And its job is to remove these bits that really it would be
better if they weren't there, at least at the beginning. So there's two memes for the
price of one in this one. This one could also be, this slide could be called: How it started versus how it's going. So we started out with
our common ancestor LECA struggling to deal with
its burden of introns. And then now today, we've got a whole array of eukaryotes that have ended up in different places. So at the bottom there we have humans. They're not exemplars
necessarily of this state, but you're all human so I
thought you'd appreciate it. Lots of organisms have
lots and lots of introns and those introns are often very big. Now it would be remiss of me to say that natural selection
did not operate on this. So yes, spliceosomes were a
bad thing that accumulated because our ancestral cell didn't have a high enough population to get rid of these mildly
deleterious elements. But evolution is nothing if not making the best of a bad job. And we now know that introns have an essential role
to play in ourselves. They provide the diversity
for coding regions, they provide the ability
to mix and match genes and to make them more diverse than they would ordinarily
be without introns. They're also sites for control. So introns often have elements within them that are essential for the
control of the individual genes. Okay? So I'm not saying
introns have no point now. But they had no point at the beginning. And we do see great differences. So at the top there is a trypanosome, the causative agent of sleeping sickness. And that has only a handful of introns. It's retained only a handful of introns. We think it's retained
them 'cause it needs them. Studies from yeast have shown that introns have important
roles in growth control and yeast, Saccharomyces cerevisiae, is another organism that has
lost most of its introns. And again, what we think
what's happened is that population genetics has brought about the loss of those introns. Organisms which have very
large population sizes, that thick gray bar, means that their populations
will rise sufficiently, that selection can start seeing
mildly deleterious elements such as these introns. Because they still confer
molecular fragility. They can still break. And if they break, they
can break you. Okay? I think something like 2/3
of mutations found in humans that cause Mendelian genetic disease turn out to be splice mutations, so good example of how important getting this intron splicing is. Okay, so we can see population genetics can explain introns' origin, and it could also explain why our genomes might look like they do. It's not the full answer,
but it gives us indications. What I want to do in the
final part of the talk is, as promised, talk about what I work on, which is this beastie here. This is C. elegans. It's not unfortunately
that nice lime green. Actually, it doesn't come
out well on that screen, but. But it's a useful organism
because of its simplicity, okay? So it's about a millimeter long, it feeds on rotting vegetation, it's pretty unappealing and it's essentially just a
tube made of about 1,000 cells. But what's interesting, I
want to mention it in passing, is if we look at the number
of genes in C. elegans, and this is kind of, this is an old story, but I want to mention it in passing. If we compare, and these
are the latest numbers, literally within the last month. So that's how many protein
coating genes C. elegans has and you can see it's pretty damn close to the number of protein
coating genes that humans have. But this is here for me to tell you that C. elegans does splicing
just like humans do, but it does something more complicated than humans do as well. So it's taken splicing
and done something weird. So at the top there is what
I've already told you about. The mechanism that needs
to remove the introns from pre-messenger RNA and all of that so that RNA can be translated. And I've said that we have
the machinery to do that, the spliceosome, and we
call this now cis-splicing because you're removing an
intron from the same molecule. Cis-splicing was invented because trans-splicing was discovered and trans-splicing takes
two separate molecules and joins them together. In this particular case, I'm talking about splice
leader trans-splicing, which takes, if you like,
a professional exon, that thing labeled in blue, and sticks it at the front
of most of the messenger RNAs in C. elegans. And again, the question might be, why? Why are you doing that? And by now you'll be going, "I bet he's gonna tell me "that it's not because
of natural selection." And you'd be right, partly. In order to explain what's going on, I need to explain a little bit
more about translation, okay? So that's the messenger RNA there being translated by this
dedicated machine, the ribosome. And the ribosome uses an
adapter molecule, an RNA, to interpret that code. The RNA brings the amino acid
that corresponds to that code. And all eukaryotic... I always feel that somebody's
gonna put their hand up. Actually, I know one that doesn't. But as far as I'm aware, all eukaryotes start
translation with methionine. Even as I say, actually I'm
sure I've read a paper that- Well, anyway, it doesn't matter. Let's assume that this is true. Most. Sorry, most. Let's go with most. Most eukaryotes start their
translation with methionine, which is AUG. So that's the signal that
the ribosome is looking for. And then it can carry on translating. But there's another signal
the eukaryotic ribosome needs, and that's this nice thing called the cap. That's the thing that I've colored there. It's actually a modified
chemical guanidine that is used in the body
of the messenger RNA, but it's chemically
modified in such a way. And that is used for the ribosome to say, "Oh, that's a messenger RNA." So what happens is one of
the sub-units of the ribosome binds to that cap, recognize it, and then travels along with other factors, I've not drawn them,
till it gets to the AUG, the thing that says, "Out of methionine. "Ah, right now I start translating." Now the important thing is this stretch between the cap and AUG, must be pristine because the ribosome
can be easily defeated. It gets bothered by all sorts of- It can get easily
distracted, shall we say. When it gets to the AUG, it
recruits the largest subunit and then translation begins. So the content between the
cap and the AUG is important and it's the content
between the cap and the AUG that determines how
well this messenger RNA will be translated, how many individual protein molecules this RNA molecule will make. Effectively, how many
ribosomes can you carry? There's the ribosomes there moving along and that little squiggle
coming out of the top, that's the amino acid chain assembling, curling up and folding and eventually forming that
sort of nondescript blob, which is the rule that we have
to use in molecular biology. Whenever we have proteins, we describe them as nondescript blobs. This is an example of what happens if the sequence between the cap
and the AUG is not pristine. If you have damage, a mutation, in there, then you can fail to recruit enough or you can fail to transit the ribosomes from the cap to the AUG
with the result that you end up with not enough ribosomes, so you have less protein
than you want to have. So this is obviously a
bad situation to be in. Natural selection will make
sure that doesn't happen in the brutal way that it does. So if you inherit that mutation, you probably won't reproduce
as well as if you don't. Now trans-splicing allows the nematode to
circumvent this problem because in trans-splicing, that front part gets removed and replaced with this dedicated
professional splice leader. It's short, so there's not really room
for extraneous mutations. And it means that the movement
of the ribosome from the cap happens really nicely, okay? So what I'm saying is
that if you trans-splice, you don't need to worry
about that particular region because you're gonna replace it. So you could be saying, "Aha, so that's what it's there for." Well, no, because you
don't actually need that. We think this is what happens, okay? So at the top there is the ground state, the beginning of evolution
of trans-splicing. You have a small RNA which at
that point is doing nothing. It's being made, but it has
no consequences for the cell. We call this, this is a neutral event. It's happening but the
cell doesn't care about it. Then in yellow, we have the
messenger RNA making protein. And then a change can happen, I've simplified this as simply the acquisition of a little blue box, that allows this small RNA to now recruit itself to the spliceosome and take part in splicing reactions. But that's not really of any use because the target messenger RNA doesn't have a signal which
will allow it to be recruited to the spliceosome. So what? This can neutrally go to fixation. If your population size
is relatively small, you might get a situation where everybody in the population
now has a splice leader. They don't know what to do with it, and so you can lose it again by mutation. Which is what that wee ratchet there is. Well, not ratchet, that wee cogwheel. It can go backwards and forwards, indicating that this is
something that is no longer, this is a neutral event
that can be lost or gained for no consequences. The next step would be the
ability to do trans-splicing. So that's stage two. You'll see the messenger RNA has acquired a little blue blob. It can now be recruited to the spliceosome where it can be joined
to the splice leader and create that little
spliced messenger RNA. It'll be translated. But again, so what? The unmodified RNA has
translated just as well. There's no selection for or against, it's just sort of drifting. It might go to fixation
or it could be lost, okay? This is a neutral characteristic. Again, that cogwheel can go either way. Where things get interesting
is the next step. Effectively what we've done is we've preconditioned a
system to, if you like, become addicted to trans-splicing. The cell has acquired the ability to bring these two RNAs together. It's not important until one of those RNAs or rather the gene for the RNA acquires a mutation that damages
the ability of the ribosome to move between the cap and the AUG, that stop sign there. Remember trans-splicing removes that. Now all of a sudden,
you've created a ratchet. Now all of a sudden,
at least for this gene, if you lose the ability
to do trans-splicing, you have a problem because now you only have a transcript which has that damaging mutation and doesn't express protein as well. Doesn't switch on the
protein as well, okay? So I hope you can see
what's going on here. We have this freewheeling
system, which is for no cost, no real cost, is building the system that just at the right time,
not just at the right time, it's just sort of sitting
there in the background and then a mutation happens and now all a sudden,
you've got trans-splicing and you can't get rid of
it, at least for this gene. Now you're in business if you're
a trans-splicing machinery because now you can start assimilating the rest of the transcripts, okay? Now you can start saying, "Okay, well I'm important for this gene. "If you lose me." It depends on the
population size, remember? But let's assume that the
population size is big enough that now if you lose this, you're gonna have a
reproductive disadvantage. Now trans-splicing can start
operating on the other. The other genes will
start acquiring mutations and trans-splice and go,
"I can solve that for you. "I can solve that for you." And gradually, this ratchet
mechanism is operating at the level of the entire genome such that you end up potentially, C. elegans isn't there yet. About 80% of it, 85% of its genes, are dependent on trans-splicing. But trypanosomes are there. That organism I showed you about, the one that causes sleeping sickness, it has few introns. It has few introns. Every single one of its
messenger RNAs is trans-spliced and it has to be trans-spliced. Okay, so this is the
bit where you will go, "Okay, that's kind of a cool
story, but why do I care?" Well, there's three reasons
why you should care. The first one is here. So those magenta arrows,
remember this tree? Those magenta arrows are
pointing to groups of organisms where significant numbers
of those individuals have been shown to use
trans-splicing, okay? Some of them include
things like the alveolates at the top there, okay? They include some of the
major components of plankton. We'll get to the animals in a minute. Discoba there, that's
where trypanosomes live. That's also where leishmania lives, okay? So these are major pathogens
that use trans-splicing. Animals over there. Lots of animals have
invented trans-splicing. And that's when I say that.
So what's going on here? The evidence, and really, you really don't want me
to go into the evidence. We can do it in questions maybe
or over beers or whatever. But the evidence suggests that trans-splicing is not an ancestral
event that has been lost. Eukaryotic cells have a propensity
to invent trans-splicing. We estimate that they've
done it at least 14 times, 14 independent inventions. It looks like there's something about the way the eukaryotic cells work and the way they conduct themselves in terms of RNA splicing, that they have a propensity
to generate this system. Which probably lots of times appears, "Hey! We can trans-splice." And then is mutated away and disappears. And we're only seeing these examples where it has taken over the cell for
population genetic reasons. So that's one reason. Trans-splicing is telling us
something I think profound about eukaryotic biology. The other reason is
trans-splicing is important, okay? It is just like splicing, just like the evolution of
introns gave us the ability to do all sorts of funky
things with our genomes, the evolution of trans-splicing allowed organisms to do odd things that eukaryotes don't normally do. And one of them is to build these long arrays
of individual genes. So each color, each one of those boxes, is a separate gene coding
for a different protein. But it's all strung
together in one long RNA. Now eukaryotes are really bad
at translating these things, but splice leader RNA, trans-splicing, allows you to break that up
into individual messenger RNAs, which eukaryotes are
very good at translating. And there's lots of reasons
why this might be useful. At least in C. elegans, it looks like building
these things means that you can be very economical with
the machinery of your cell. Whereas if say you had shown
there you've got four genes, each of those genes would need
separate control elements. Well, if you've got
four all glued together, they only need one control element. So you can be more economical and that clearly is an advantage. Why do we know that it's an advantage? 'Cause again, every
organism where we've found splice leader trans-splicing, we have found operon. Sorry. Polycistronic
RNA, that's what that is. Should have defined that. Okay, that's only reason number one. Reason number two, which
I kind of hinted at. A lot of the organisms do
really bad things to us are dependent on trans-splicing. There's trypanosomes, we've
already talked about those. I've talked about C. elegans. C. elegans is a nematode. There are some really
bad nematodes out there that want to do very bad things to you. Flatworms, schistosome, schistosomiasis, these organisms all use trans-splicing. Because they use trans-splicing and it's an essential
process that you don't use, that plants don't use,
and animals don't use. These parasitic organisms,
potentially we can target them using drugs that targets
this splicing machinery, because it doesn't exist in us. And that's one of the
things that we do in my lab. We're trying to develop drugs
that will treat the mechanism, the proteins and the RNAs involved in trans-splicing in nematodes. Okay. And then, we're
not too bad time-wise. The third reason is this. So this is my summary slide for what I've already told you about how trans-splicing evolved. But what I've told you is
actually a more general principle than we'd appreciated. So I don't like naming things too much unless it's a necessary, 'cause I think naming things
is not always necessary. But this process of developing complexity essentially for free
using neutral evolution has been called constructive
neutral evolution, okay? The story I've told you about with splice leader trans-splicing is only one example of this. And there are a number of researchers working in various labs that have found similar instances, okay? So Joseph Thornton's lab
has shown very clearly that a lot of the proteins, so mammalian proteins, which
interact with one another to form larger groupings of proteins, those groupings are important now but they didn't used to be. What happened was they went through the same sort of ratchet
process that I'm talking about. So effectively what you have
is you have a beginning state where things are free-flowing. So for instance, for
Joseph Thornton's work, they can show that you can
bring proteins together that interact with one another. That interaction's not important. It's just happening, okay? So it can go backwards or forwards. So the proteins can interact, a mutation can prevent them interacting, it has no consequences for the organism. They get to State B where all the proteins are interacting with one
another constitutively, but it's still that
interaction isn't important. And now a mutation can happen
in one of the components such that if you now prevent
these things from interacting, that mutation in one of the components kills the whole system. So again, you've established a ratchet. The important thing is the principle of the
two proteins interacting happened by that freewheeling state, okay? It wasn't being selected for or against. It was what we call this
phase of neutral evolution. But you're building molecular complexity. Trans-splicing is building
molecular complexity. It's making molecules exist that don't really need to be there. And in fact, early on
could have been got rid of. But they then serve as a
substrate for the next stage, which is usually the ratchet stage where some mutation
happens, where it said, "Ya-ha. Now if you get rid
of this, you're doomed." And this I think is increasingly
a general principle. So I just want to end then,
if you like, as a plea. I hope I have convinced you that a simple view of the way things look in terms of biology,
in terms of complexity, just by looking and just
by taking into account that, say, natural
selection has designed this, that is only going to be
telling you part of the story. I'm not denying that a lot of these things are going to be subject
to natural selection. In this picture, natural
selection has to work. The ratchet is the natural selection part. But just because you see complexity doesn't mean to say that
complexity has evolved for doing something, okay? It can happen by freewheeling,
by simple genetic drift. Genetic drift can build
complex structures, which then themselves become the substrates
for natural selection. Okay. So I'll end with the requisite end slide. I want to thank... This is members of my lab last Christmas. Two of them are here. (laughs) And we are funded by the the BBSRC. Again, I want to thank
the Genetics Society for allowing me to come here today and tell you all about their stuff. And I'll stop there and take questions. Thank you. (audience applauds) - [Host] Thanks very much.
