The genetic basis of evolutionary change in morphology, phenotypic adaptations, and behavior
Transcript of Part 1: Introduction
00:00:00.18 Hi, my name is Hopi Hoekstra and I'm a professor 00:00:03.10 at Harvard University. Today what I'm excited to do is to 00:00:06.23 tell you about the field of evolutionary genetics, and in particular, 00:00:09.26 the genetic basis of evolutionary change. I'm going to tell you two 00:00:13.29 stories, one about morphology and one about behavior. 00:00:16.23 So, here's the outline of the 3 segments of my presentation. 00:00:21.12 So what I'm going to do now is give you an introduction 00:00:24.10 and introduce you to some of the longstanding questions 00:00:27.17 in the genetics of adaptation, and give you a sense of how 00:00:30.14 we're addressing these questions. And then the second segment, 00:00:33.21 in particular, I'll tell you a story about how we're tracking down 00:00:38.09 the genes and developmental mechanisms involved in camouflaging 00:00:42.11 and color differences between two species of wild mice. 00:00:45.08 And then in the third segment, I'll tell you about how we're using 00:00:47.25 very similar approaches to track down genes involved in burrowing behavior 00:00:51.25 differences in these same mice. 00:00:55.26 So I want to start today by talking about Darwin. Because in 00:01:00.06 2009, we had a number of celebrations celebrating everything that 00:01:05.07 Darwin knew on his 200th birthday and on the 150th anniversary 00:01:09.12 of his magnum opus, On The Origin of Species. Now Darwin 00:01:13.19 certainly knew a lot about evolutionary change, but there is one thing 00:01:19.00 that he didn't get quite right. And that is the mechanism 00:01:22.09 of evolutionary change, or the genetic nuts and bolts about how 00:01:26.02 organisms adapt to their environment. Now Darwin knew the traits were 00:01:29.04 inherited, he knew that offspring resembled their parents, but he didn't 00:01:32.14 know how. And this maybe isn't surprising because during this 00:01:36.02 time, of course we didn't know about DNA or genes, much less 00:01:39.08 the whole genome. And that's really what I want to focus on 00:01:43.21 today, is this mechanism of how changes in genes actually produce 00:01:47.26 variation in phenotypes on which natural selection can act. 00:01:52.08 So I'm going to start by telling you a brief anecdote that links 00:01:55.22 Darwin to a second great discovery, and that is the discovery of 00:01:59.06 DNA. So what I'm showing you in this next slide is Darwin's 00:02:04.24 last publication. Now I don't expect you to read it, but i just want 00:02:08.05 you to appreciate the fact that you're looking at his last publication. 00:02:12.08 It was published in 1882, just two weeks before he died 00:02:15.03 in a prestigious journal called Nature. The title of this article 00:02:18.23 is called, "On the Dispersal of Freshwater Bivalves." 00:02:21.28 And what this really is, is a report of the finding of a freshwater 00:02:26.10 beetle clamped to its leg was a freshwater clam, or a cockle. 00:02:30.15 So why you may be wondering was this published in such a 00:02:34.14 prestigious journal even 100 years ago? Well, this actually 00:02:40.06 resolved this great debate about why freshwater cockles were so similar 00:02:44.10 among disjunct lakes in the British midlands. One hypothesis 00:02:49.09 was that these cockles could migrate from lake to lake, 00:02:52.17 thereby homogenizing the populations and thereby, making them 00:02:56.23 very similar in size and shape. But the big question always was, 00:03:00.15 well how do they get from lake to lake if they can't cross 00:03:03.07 terrestrial habitats? Well here was a mechanism. They could 00:03:06.15 hitchhike by attaching to things that could fly or traverse 00:03:11.29 this terrestrial habitat -- in this case by clamping to the leg of a 00:03:15.22 beetle. But that actually isn't the point of telling you this story. 00:03:19.16 The point of telling you this is to mention that Darwin was sent this 00:03:24.08 beetle with a cockle clamped to its leg by a shoemaker who 00:03:28.22 was working in the British midlands, who was an amateur naturalist. 00:03:31.21 And his name was Walter Drawbridge Crick. Now this name should ring a 00:03:36.12 bell, because his grandson was the one with his colleague, Jim Watson, 00:03:41.04 that made the second great discovery. That is the discovery of the 3-dimensional 00:03:45.09 structure of DNA. And it's in this DNA text that we find even more 00:03:52.14 evidence for Darwin's great theory, that is our 3 billion year 00:03:56.20 existence, the shared evolutionary history of all living organisms, 00:04:00.18 and the subject of what I want to talk about today. And that is the 00:04:03.20 mechanistic basis for evolutionary change. 00:04:07.17 So like Darwin, one of the big questions in evolutionary biology 00:04:11.12 today is what gives rise to this amazing diversity? How is variation 00:04:16.02 generated and maintained in natural populations? 00:04:19.10 But thanks to Watson and Crick, we can look for that answer 00:04:22.11 in the genetic code. So the big question that we're focusing 00:04:26.02 on is what is the genetic basis of fitness-related traits? 00:04:30.14 By fitness-related traits, I mean traits that improve the probability 00:04:35.10 of survival or reproduction of organisms in natural populations. 00:04:39.07 So finding the genetic changes or the precise DNA changes 00:04:44.04 that contribute to variations either between populations or between 00:04:47.15 species, is a fun endeavor. And we certainly can learn things about the mechanistic 00:04:53.14 aspects of evolutionary change. Like how do changes in genes 00:04:57.15 actually produce changes in phenotype? But I'd like to argue that 00:05:00.21 we can actually learn even more about the evolutionary process. 00:05:04.11 So what can finding genes tell us about how evolution works? 00:05:09.03 Well there are a number of longstanding questions that I think we're 00:05:12.13 just now starting to be able to answer, because we're armed 00:05:16.06 with molecular biology and the ability to link genotype and phenotype. 00:05:20.17 So I'm just going to list a few of these big questions. 00:05:24.03 So for example, how does evolution proceed? Is it through 00:05:28.27 many small changes? Many small mutations, each that have a small 00:05:33.13 effect on the trait, or can evolution take big leaps? 00:05:37.08 That is, can mutations have large effects that are beneficial? 00:05:40.08 We also want to know about the dominance of these mutations. 00:05:45.07 So, do adaptive alleles or mutations that appear, do they tend to be 00:05:49.21 dominant or recessive? So J.B.S. Haldane, one of the founders 00:05:53.28 of population genetics, argued that adaptive mutations tend to be 00:05:57.19 dominant. Because when they first appear, they're visible to selection 00:06:01.03 and then can quickly spread through the population. 00:06:03.01 Compared to a recessive mutation, which would have to build up enough 00:06:07.14 number in a population to be contained in the same individual, 00:06:12.12 and that recessive trait then expresses. We also want to know, how 00:06:16.23 many -- how do these mutations interact? So if multiple mutations 00:06:19.26 are responsible for changing the phenotype, do they interact 00:06:24.03 in a complex way? Or does each mutation additively affect 00:06:28.07 that trait? We also want to know where these mutations, these beneficial 00:06:35.08 mutations are. Do they occur in the protein itself? For example, 00:06:39.17 amino acid changes that affect that structure and function of that 00:06:42.15 protein. Or do they occur in what we call non-coding DNA, 00:06:46.22 which affects the regulation, let's say the timing or place of expression 00:06:51.21 of that protein. And then we want to know where these mutations come 00:06:55.25 from. So for example, if there's a change in the environment, do we have to 00:07:00.13 wait around for new mutations to appear in that population? 00:07:03.08 Or are there these mutations maybe at a low frequency in the 00:07:07.18 population already that are pre-existing that can be selected 00:07:10.27 on almost immediately? And then finally, if we find mutations in one 00:07:16.03 population that are responsible for an adaptive trait, and we 00:07:19.01 have a similar trait involved in another population, is it the same 00:07:22.08 mutations and same genes that are responsible for those 00:07:25.16 convergent traits? Now importantly, all of these questions that I've listed 00:07:29.17 don't have simple yes or no answers. And in fact, we're more 00:07:33.08 interested in the frequency, whether for example, more often 00:07:37.13 beneficial mutations occur in regulatory regions versus structural 00:07:41.17 regions. But even more importantly than that, we want to know 00:07:44.26 why. Why in some cases do we see protein changes and in other 00:07:49.09 cases we see regulatory changes. Now these I would argue are 00:07:55.01 still largely open questions, but questions we can start to answer 00:07:58.25 by making the connection between genotype and phenotype. 00:08:01.20 So the context in which we're studying the genetic basis of 00:08:05.15 adaptation looks like this. That is, we're trying to make the connection between 00:08:09.16 environment and phenotype. In other words, trying to implicate 00:08:12.21 a role for natural selection in driving that phenotypic variation. 00:08:16.00 That is, suggesting that the phenotypic differences affect fitness. 00:08:20.02 But we also want to understand the genes underlying that phenotypic 00:08:23.29 variation, and not just what those genes are, but how those genes 00:08:27.12 through let's say development, actually produce the differences 00:08:30.28 in variation. And then once we make those links, we'll have a much 00:08:34.18 more complete picture of the adaptive process. I think this is where things 00:08:38.09 can get really fun, because we can go back out in the wild and ask how traits 00:08:41.21 evolved in nature. So, to make these links between environment 00:08:47.07 and phenotype and genotype, my lab group is studying one particular 00:08:51.26 group of wild mice, commonly referred to as deer mice. 00:08:55.05 Or mice in the genus peromyscus. These are the most abundant 00:08:59.10 mammal in North America. And the reason we study them is because 00:09:04.04 first, they're found in almost every habitat type. So from the top of the Rocky 00:09:09.03 Mountains out to the shores of Maine, to the plains of Kansas, to the deserts 00:09:15.21 of Arizona. So they're very widespread in their distribution 00:09:19.12 and because they live in all sorts of different habitat types, 00:09:22.06 there's a lot of opportunity for local adaptation. 00:09:24.23 So in addition to all the variation that we find in the wild, 00:09:28.14 they also can be treated much like laboratory mice. That is 00:09:32.14 we can bring them into a controlled laboratory environments. They 00:09:35.24 breed in the lab just like laboratory mice, and we can do controlled 00:09:39.22 experiments. And finally, while we're still behind traditional 00:09:45.05 model organisms, my group, as well as others, is building a series of 00:09:49.27 genetic and genomic tools that are going to be useful in trying to 00:09:54.17 make these connections between genotype and phenotype. But I would argue 00:09:58.00 one of the main reasons for studying these mice is because 00:10:01.11 we have this amazing literature of natural history studies on their 00:10:08.02 ecology. That is, these mice have been studied for nearly a century 00:10:11.13 by natural historians who have described morphological, physiological, 00:10:15.05 and reproductive behavioral variation in natural populations. 00:10:19.21 Just to give you a sense of how these mice vary, here are just 00:10:25.01 a number of traits that I picked out of the literature that describe 00:10:29.01 traits that have been studied and traits vary either between 00:10:32.14 populations or between species of peromyscus species. 00:10:36.12 So they vary in body size, tail length, foot size, color patterning, 00:10:39.27 testis size, sperm morphology, et cetera. They vary in morphological 00:10:43.05 traits, physiological traits, and behavioral traits. So, using these 00:10:48.20 mice, we're trying to make those connections between genotype and 00:10:52.01 phenotype. And the next two segments of my presentation, I'm 00:10:54.23 going to focus on two of these traits. One morphological trait, 00:10:58.04 color patterning, and a second trait, burrowing behavior. 00:11:01.18 So, the second part of my presentation, what I'd like to do 00:11:07.05 is focus on the morphological trait. And in particular, camouflaging 00:11:10.28 and color differences between subspecies of peromyscus polionotus. 00:11:16.02 Both to understand the ultimate reasons why these color differences 00:11:20.10 evolved, as well as the mechanisms or the underlying genetics 00:11:24.21 contributing to these differences in camouflage and color. 00:11:28.08 And for the third part, we'll switch gears and focus now using 00:11:32.24 very similar approaches. But instead of studying a morphological 00:11:34.26 trait, we've substituted in a behavior where we're taking advantage of these dramatic 00:11:39.18 differences in burrowing behavior; there are species that build these 00:11:42.24 large burrows versus those that build small burrows. To try and 00:11:47.11 understand how genes can affect behavioral variations in natural 00:11:50.21 populations. So I've hoped to have gotten you excited about 00:11:54.28 biology, in the sense that we're at this amazing time where 00:11:59.11 we can use approaches like Darwin first did, that is studies of 00:12:03.29 natural history, observation and experiment in the wild, 00:12:07.14 but combine that with studies of modern day molecular 00:12:10.15 genetics. To try to understand the genetic basis of what Darwin 00:12:14.26 referred to as that perfection of structure and coadaptation which 00:12:19.01 most justly excites our admiration. So thank you very much 00:12:23.00 for your attention, and I hope you'll join me for the next two 00:12:26.24 segments, where I'll tell you about more detailed studies from 00:12:29.19 laboratory group, trying to connect genes and phenotypes for both 00:12:33.11 morphological and behavioral traits. Thank you.