hello today is lecture I'm going to be talking to you about experimental evolution in the outline I'm going to explain what experimental evolution is and why it's a powerful approach looking at the evolutionary genetics of adaptation through and understanding through approaches using experimental evolution the evolutionary genetics of sexual selection exposed by experimental evolution and also experimental evolutionary ecology and finally some applications such as the evolution of resistance and I won't be talking about shaping artificial life so let me think about evolution we often think about a very long time scale we might think about our evolutionary ancestors or we might think about the fossil record or even weird and wonderful animals really deep leave far back in time like the trilobite however evolution does occur on a on a contemporary level and we can see it all around us in our dogs and cats and other domestic animals that is something that the layperson if you like might recognize as domestication and also as a a artificial selection but there is a definite parallel between the changes that have happened through that human imposition on those animals as well as a slight difference from experimental evolution in the context of domestication we choose particular traits that we want to be passed on to the following generation of animals and that leads us to evolve from the wolf to rediculous pooches to highly productive cattle from ancestral wild cattle from the wild jungle fowl to the chickens of today that lay an egg every day rather than laying 15 eggs once a year that's because we've targeted particular traits experimental evolution slightly different we create an environment or a test and evolutionary experimental test where we create conditions and allow them to determine what happens and test evolution that way the first example of experimental evolution that I'm aware of was work by William Dellinger from 1839 to 1909 is when he lived now he had a an amazing for its time experiment he had this contraption which is basically an incubator in which he kept bacteria and he sourced his bacteria and they were living in conditions of about 16 degrees C and he kept the bacteria in this broth and he slowly increased the temperature over numerous generations and over a period of seven years he increased the temperature from 16 degrees C to 70 degrees C and the bacteria was still able to live a 70 degrees C that evolved and he knew that they'd evolved because when he took the ancestral cells and he tried to grow them in the in the derived environment they couldn't and the derived cells couldn't grow in the ancestral environment so they had evolved now Darwin really would have loved to know about experimental evolution because for Darwin it was a frustration that we couldn't watch evolution in accident action Darwin said in looking for the gradations by which an organ in any species has been perfected we ought to look exclusively to its lineal ancestors but this is scarcely possible we are forced in each case to look at a species at the same group that is to the collateral descendants from the same original parent form so if you imagine the the sort of tree of life and each species are the twigs on the Tree of Life then we can examine differences between those species but we can't follow each individual species back through time to its lineal ancestor which is what Darwin was saying but with experimental evolution maybe we can maybe we can make fine scale assessments of how evolution is working to get from one species to another or the ancestors at one species to the original origins of so why is experimental evolution such a powerful approach there's first of all that's control so the experimenter determines what experimental test is going to be done and you can rule out many of the environmental effects and the ancestral phylogenetic effects that affect other processes when we're looking at evolutionary biology secondly there's replication so if we look at the evolution of particular species then that particular species is only evolved once we may see convergent evolution where evolutionary problems are solved in a similar way but in this case we can say all right we're going to set up this environment we're going to test this hypothesis and we're going to ask that these three replicates if you like of that particular species and ask whether they will evolve in the same way as as each other or whether evolution will have different outcomes another thing which Gallinger did is the quantification of divergence so we can keep our ancestral population we can evolve our derived population and then we can test them against each other is one fitter than another in a particular environment or not so having this control and replication and the ability to step back in time and look at the ancestral versus the derived populations means that we really have an amazing ability to look at the basic underpinnings of evolutionary biology perhaps not going back in time through our own species history but in some circumstances as you'll see definitely going back in time in some experimental situations so some of the questions that we might ask of evolutionary experimental evolution about the evolutionary genetics about adaptation are when the most beneficial mutations occur how do they spread is genetic improvement happening all the time even in stable populations is adaptation a smooth process and is it reproducible these are things that we can ask of experimental evolution now for the first part of this lecture I'm going to concentrate quite heavily on the work of Richard Lenski and his long-term experiments with E coli so this experiment was done or is continuing to be done at Michigan State University and it started on the 24th of February in 1988 so there are 12 vials of bacteria of E coli that are generated from a single clone so they all started off being genetically identical and they have a media as they live in which is glucose and citrate the glucose that bacteria can eat and the citrate was something that they couldn't eat but is a possible carbon source and in February 2010 they've done 50,000 generations now at each generation sorry each day the culture is refreshed so bacteria are taken and resuspended into a new vial and every week a sample is frozen at minus 75 and at any one point during the experiment you can go back in time to that frozen sample reactivate it and you have moved back in time to that ancestral point and at every general at every week the same process is repeated for measuring adaptation and that is how cloudy the liquid is after a certain period of time and that's measured on a color imita and you can measure Fitness against the background population the assert ancestral population and look at relative Fitness so at the beginning of this experiment you can see we've got the x axis here which is the number of generations and the relative fitness the fitness relative to the base population and you can see that adaptation to this normal environment has increased over time now there are two different curves here marked in blue and red now the the red curve is one in which it has an asymptote so an asymptote is where it never where it ceases to increase after a certain period of time the blue curve is one which doesn't have an asymptote but carries on increasing and interestingly the blue curve fits the data better and the red curve so most adaptation took place very early on between generation zero and generation 10,000 and nevertheless it seems like adaptation is still taking place even later in the sequence and that's confirmed by looking at the Fitness at generation 40,000 and comparing it to the generation at 50,000 so at the generation the fitness of generation 50,000 those data points all lie above the one-to-one line so in general the populations these twelve clones these twelve lines that came from that single clone are all more fit in generation 50,000 and they were generation 40,000 over here you can see the relative fitness of the average of the 12 generation 40,000 and 50,000 and the ratio between the two and you can see that indeed Fitness did keep increasing even in that stable environment one of the intriguing things about these populations was that three of the twelve lines evolved a mutation rate which was much higher than the ancestral level one hundredfold higher than the ancestral level and in effect these populations cycle through mutations and that much faster and that means that mutations in these hyper hyper mutated lines selection can pick out the favorable mutations at a much higher rate and you can see that the hyper mutation populations adapted to the media after about seven thousand generations at a much faster rate so it's interesting to think about how mutation rate might be an adaptation in itself so one of the big questions of the 1970s which seems a long time ago as many of you I'm sure but has sort of kind of became unresolvable was Eldridge and Google's proposed idea of punctuated equilibrium so this is where evolution continues at a always punctuated by periods of rapid change so there's a fairly static period of no change and then a jump now in this species of E coli cell size is a very good measure of fitness and you can see that positive relationship between cell size and fitness over there now when they were looking at the evolution of cell size over the first three hundred generations they fitted a stepwise model compared to a continuous model and it seems like the stepwise model fits better and what happens is that beneficial mutations arise but then it takes them a long time to get to a frequency where they affect the mean of the population so they gradually build up and then suddenly their effect on the population mean is obvious and then there's a big jump in the population mean as that gene suddenly sweeps to fixation and so you can see that stepwise outline of the population cell size distribution suggesting that there are these sort of punctuation points in the fine-scale evolution of cell science so the reproducibility of evolution is something that is really unique to studies of island populations or populations like these under experimental conditions and obviously the experimenter can really manipulate things to see how reproducible evolution is now the figure on the far side shows to adaptive Peaks so the path of the population one path goes up the red peak and one path goes up the green peak now what happens is that random processes might determine the early trajectory of the population either towards the red or to the green peak but once on the slope of that peak when there is a valley in fitness terms between the green and the red then that population is locked in to going up the green peak even if the ultimate fitness of that population is lower than the fitness of the population which when the red peak okay so subtle changes in early evolutionary processes can lead populations in to different adaptive peaks and that suggests that evolution social stochastic events might affect evolution quite strongly here we can see the evolution of cell volume in these populations and we've got the number of generations on the x-axis and cell volume on the y-axis and all of the populations increased in cell volume but the you can see that for example the blue population increased to a very much larger size than for example the black population or the yellow population so although all the populations changed in cell size some of them evidently ended up at a higher adaptive peak than others one way that this lens keaney's colleagues were able to look at the reproducibility of evolution was to go back in time to see how a particular trait had evolved so there's this idea that early that mutations may occur which predispose populations to a adaptive outcome and that's historical contingency and what they did was they found a population that evolved to be able to utilize the citrate substance that they provided and they wanted to know how reproducible this event was so they took because they've frozen samples all the way back through the experiment they went back in time and at each time point re-released that population into the media to see whether and when it evolved the citrate you the ability to use citrate and what they found was that somewhere around generation 33 thousand was the point where if you went before that in time they didn't evolve to use citrate and after that they did evolve citrate it seems like there are two mutations one which was needed and occurred at around generation three thousand and one that subsequently happened which alaya allowed them to use the citrate so when we come to think about other examples compared to that example of bacteria we might because we might think that okay this this is something that just can be done with generations are very with species a very short generation time but we are able to look at other species in this in this experimental evolutionary framework and those can answer questions about sexual selection amongst other things so here's a nice example by a friend of mine Bradburn yes a grad van and he works on these mites which live in the soil and I have some in my lab upstairs for anyone's excited about the prospects of working on lines and here we have a an experiment which he did looking at the role of sexual selection in reducing or removing mutations now these mutations weren't naturally arising they were induced by radiation so this was in the might rise of glyphosate Canopus and they have a very short generation time so he irradiated some males and mated them with a light dose of radiation mated them with some females and he had a control group and then he looked at a situation where there was polyandry and where there was mono monogamy and in the monogamous situation offspring and produced by with in the absence of sexual competition in the polyandry situation males have to compete for access to females and also their sperm have to compete for fertilizations internally within the female what he found was that the embryo viability which is a very good measure of the load of mutations that an individual carries the embryo viability was much higher when reproductive condition competition could take place a similar study in dung beetles done by lee simmons in this group and his group here on these dung beetles on the fakest torus showed that in a similar way these induced mutations were removed when sexual selection was allowed to take place so in generation 3 you can see that the fitness of the radiated males under enforced monogamy is much lower than the irradiated males when there was sexual selection and indeed that the fitness of the radiated males under sexual selection is almost restored back up to the control level I've been doing some work on this with a student of mine Rob Dugan and this is looking at the evolutionary genetics of sexual selection in Drosophila melanogaster by selecting four studs and duds so in this scenario the mail that mates with a female when there are two males and a female in a vial is designated as a stud male and the loser is designated as the dud male now in this system it's very interesting what we did was we took the studs and we took the duds and we bred them for stud leanness and Dudley nurse over numerous generations now normally the duds wouldn't get to leave any offspring so we're kind of selecting on the lower ability of males that would normally not leave offspring and what we found was that male mating success which was this measure of that we were using responding to this change in selection on males enhancing or enhancing the characteristics of studs and duds to produce population that differed in their mating attractiveness and in fact what we showed was that here when you're looking at the inbreeding in egg to imbrie depression in egg to adult viability was much greater in the deadlines suggesting that there's a genetic correlation between attractiveness and the the load of mutations which affect egg to our viability also in this department is work that's been done and the nice thing about this is that it's on a mammal but it is experimental evolution in mice by Renee Fuhrman now she did 27 generations of selection on mice either in a situation of polyandry or ma nandri where sexual selection is elevated or removed completely and she found several interesting things firstly the ejaculate quality of the polygamous the lines with the selection history of polygamy was increased that is the the sperm were more motile more progressive and more rapid and faster and more long-lived and these consequences of the sperm trait seem to translate into paternity and here we can see that the polyandrous lines sort of Li the polygamy lines have a higher paternity when the second male mates second you take their female mouth she mates to two males and the second male to mate when he is from the polygamous history mating history as a higher for fertilization success of those offspring similar work on the dung beetle has been done where males and females were either in a high-density mixture or males and were isolated with a single female over numerous generations and this produced an evolutionary response in the relative testes weight of males so males that were not experienced aiming sperm competition evolved to have smaller testes I've done some work on with this might as I mentioned earlier looking at the the fighting morphology of males which is that the third pair of legs that you can see at this figure is much thicker in fighter males and they're larger than their counterparts who are scramblers so this is a male dome orphism which some of you will hear more about in behavioral ecology now I gave these mites a change in their habitat so normally we keep them in a simple plaster of Paris plane if you like in a petri dish but we stuck straws into the petri dish to make their habit more complex to make them have to climb around and all the food which is used that they was down the straws so they had to negotiate this complex habitat in order to reproduce and we left them there for 10 generations and we looked at the change in the relationship between body size and producing that fighter morphology and what we found was that in the complex habitat the ability of males to get around was much more constrained for the fighter males because of that third that very thick leg they couldn't negotiate the ups and downs of the straws so well as the Scrambler males and that shifted the threshold at which males become fighters to a much larger body size so a smaller proportion of the population became fighter males so that's an example of how the abiotic environment can be looked at in the context of experimental evolution so experimental evolution is taking place in the fields and hedgerows around us every time someone sprays a crop and leaves a a feed or a weed species to grow then that is an evil it that species that has been exposed to that chemical but and yet survives will leave more offspring and obviously this takes place as we know in hospitals all the time with multi drug-resistant anti bacterial being a massive problem you could study this evolution of resistance in things like ryegrass which is a terrible problem for farmers in Western Australia but the trouble is that ryegrass which grows along field margins gets a low dose of herbicide and then propagates resistant offspring and you could do a selection experiment on rye grass but knowing the particular characteristics of the spraying regime would either increase resistance or decrease resistance is very difficult in a species like rye grass which has quite a long generation time so paul nave and colleagues had looked at using instead of using rye grass use a single-celled alga an alga which i also have in the lab in culture for looking at the spray regimes which may lead to better results when it comes to resistance and try and wipe out existing resistant species the great thing about Clemmy - - is that it doubles every 4 hours which is much faster than rye grass in a field trial so here they used three herbicides that is atrazine glyphosate and car better mind and they used them in either you can see that a 0 means atrazine on its own and AG is pat Rosine and glyphosate and the one is how frequently so there's the weekly cycle the bi-weekly cycle or the try weekly cycle so and then we've got on the y-axis we've got the number of weeks to resistance on the y-axis and you can see that some combinations take longer for resistance to evolve than others and similarly in the figure be the you can see that some combinations don't evolve resistance at all what's interesting about this work is that it reveals how there is a genetic correlation between being resistance to one herbicide and being resistant to another so here we've got the crop the spraying regimes if you like on the rows and in the columns are whether they became resistant to that particular herbicide or not so we've got actress apply for Satan car better mind which were the trials and then we've got s metal or I so prove true or on and timber o try own and what it turns out is that if they had evolved resistance to atrazine and car bet amide then they were also resistant to those other herbicides as well so there must be a genetic correlation the same genes are evolved that are involved in resistance to one chemical are also provide resistant to another so that's a use for experimental evolution experiments and looking at herbicide resistance so there's a lot of experimental evolutionary ecology going on at UWA and you should think about doing maybe doing on us or visiting some of those labs to see if you can help out if you're interested so there might be some criticisms and caveats about experimental evolution and its utility so test tubes are an unnatural environment how realistic is it transfer the knowledge from test tubes to the natural world our test tubes too simple and how different our microorganisms and the micro evolutionary processes like that punctuated equilibrium that we saw from the punctuated equilibrium that was envisaged originally by Aldrich and Gould and so the question remains as to where the micro evolution can be scaled up so to conclude experimental evolution is a powerful way to address evolutionary questions and the interplay between ecology and genetics at the moment we've got some possibilities for Honor's projects which I'll just flip through here with me and also Bruno Rosato and Jason and these are some references they'll help you understand experimental evolution