1 00:00:00,000 --> 00:00:01,000 ELENA KRAMER: All right. 2 00:00:01,000 --> 00:00:03,735 Well, thank you so much for coming out here today. 3 00:00:03,735 --> 00:00:05,855 It's a thrill to be here with you. 4 00:00:05,855 --> 00:00:08,855 So, as you just heard, I'm a plant biologist. 5 00:00:08,855 --> 00:00:12,765 I study the evolution of genetic programs that control the development, 6 00:00:12,765 --> 00:00:14,595 particularly of flowers. 7 00:00:14,595 --> 00:00:17,855 Now, I have to admit, I've sat next to a lot of people on planes 8 00:00:17,855 --> 00:00:20,135 and told them that I was a plant biologist. 9 00:00:20,135 --> 00:00:26,015 And often, their reaction is, help me find out why my fern is dying, 10 00:00:26,015 --> 00:00:27,515 or something like that. 11 00:00:27,515 --> 00:00:29,765 I don't keep things alive. 12 00:00:29,765 --> 00:00:31,655 This is the problem. 13 00:00:31,655 --> 00:00:34,535 I like to grind them up in little test tubes and things like that, 14 00:00:34,535 --> 00:00:36,695 so don't ask me how to keep something alive. 15 00:00:36,695 --> 00:00:41,615 But we talk in plant biology about plant blindness. 16 00:00:41,615 --> 00:00:43,715 People often don't appreciate the importance 17 00:00:43,715 --> 00:00:45,995 of plants in our everyday life, but I think 18 00:00:45,995 --> 00:00:50,345 we can all agree that flowers are very compelling and charismatic. 19 00:00:50,345 --> 00:00:54,365 As a human civilization, we've had a fascination with flowers, really, 20 00:00:54,365 --> 00:00:55,565 from the beginning. 21 00:00:55,565 --> 00:00:58,535 And what we're really trying to understand in our lab 22 00:00:58,535 --> 00:01:01,985 is how we have, over the course of evolution, 23 00:01:01,985 --> 00:01:05,705 generated this enormous diversity of floral morphology, 24 00:01:05,705 --> 00:01:09,695 and then, particularly, how changes in the genetic programs that 25 00:01:09,695 --> 00:01:14,945 control these developmental processes has generated the change in morphology. 26 00:01:14,945 --> 00:01:18,749 And in order to do that, we work at the intersection 27 00:01:18,749 --> 00:01:20,165 of sort of three different fields. 28 00:01:20,165 --> 00:01:21,975 We ask different types of questions. 29 00:01:21,975 --> 00:01:25,835 So for instance, we're very interested in development. 30 00:01:25,835 --> 00:01:29,315 Very specifically, what are the patterns of cell division and cell 31 00:01:29,315 --> 00:01:32,075 expansion, the duration, their localization, 32 00:01:32,075 --> 00:01:35,765 and how do these different patterns generate the morphology 33 00:01:35,765 --> 00:01:36,965 that we're interested in? 34 00:01:36,965 --> 00:01:41,775 So we ask very specific questions about how morphology changes over time. 35 00:01:41,775 --> 00:01:45,545 And then we overlay this information with information about the genes. 36 00:01:45,545 --> 00:01:48,755 How do different genes interact with each other on a local scale, 37 00:01:48,755 --> 00:01:50,975 on a global scale, on the genomic scale? 38 00:01:50,975 --> 00:01:55,505 And ask how these genetic programs control the developmental processes. 39 00:01:55,505 --> 00:01:57,545 And then we take all of those questions and we 40 00:01:57,545 --> 00:02:00,065 put them in an evolutionary context, and say, 41 00:02:00,065 --> 00:02:03,575 well, when we ask all these questions in one species then 42 00:02:03,575 --> 00:02:07,055 we compare them to a different species, what aspects of these processes 43 00:02:07,055 --> 00:02:09,245 are the same, what aspects are different, 44 00:02:09,245 --> 00:02:13,415 and can we understand how those differences arise? 45 00:02:13,415 --> 00:02:15,305 So when we put all those things together, 46 00:02:15,305 --> 00:02:19,445 that's a field called evo-devo, or evolutionary developmental biology. 47 00:02:19,445 --> 00:02:23,585 So we're essentially asking how morphological diversity is generated 48 00:02:23,585 --> 00:02:26,685 over time at a molecular level. 49 00:02:26,685 --> 00:02:30,245 So, why don't we use plants as a model system? 50 00:02:30,245 --> 00:02:33,185 So if you'll let me proselytize for a moment about why plants 51 00:02:33,185 --> 00:02:34,925 are such a wonderful model system. 52 00:02:34,925 --> 00:02:36,545 I'm a developmental biologist. 53 00:02:36,545 --> 00:02:39,125 If I was an animal developmental biologist, 54 00:02:39,125 --> 00:02:44,945 I would only be focused on a very narrow window of that organism's life-- what 55 00:02:44,945 --> 00:02:50,235 we typically call embryogenesis, from fertilization to some type of hatching. 56 00:02:50,235 --> 00:02:54,845 But in plants, the fantastic thing about plants is that if they are alive, 57 00:02:54,845 --> 00:02:57,035 they are undergoing development. 58 00:02:57,035 --> 00:03:00,305 If we look at this plant embryo, you can immediately 59 00:03:00,305 --> 00:03:04,895 recognize that while, yes, there was a developmental process that took this 60 00:03:04,895 --> 00:03:09,605 from a single cell to this embryo, this is not 61 00:03:09,605 --> 00:03:12,915 the entirety of the body of a plant. 62 00:03:12,915 --> 00:03:17,405 So this enormous oak tree started out as this tiny embryo, 63 00:03:17,405 --> 00:03:20,615 and it is in a constant state of organogenesis 64 00:03:20,615 --> 00:03:24,155 and producing new organs in order to generate its body. 65 00:03:24,155 --> 00:03:24,985 And all around us-- 66 00:03:24,985 --> 00:03:26,735 I mean, we finally got to go outside today 67 00:03:26,735 --> 00:03:28,925 and enjoy a little bit of nice weather-- 68 00:03:28,925 --> 00:03:32,675 you can see trees all over campus, things coming up from the ground, 69 00:03:32,675 --> 00:03:35,405 these are plants that are undergoing organogenesis even 70 00:03:35,405 --> 00:03:37,955 in what we would consider their adult body. 71 00:03:37,955 --> 00:03:41,805 And plants can do that because all throughout their body, 72 00:03:41,805 --> 00:03:45,315 they have reservoirs of undifferentiated cells-- 73 00:03:45,315 --> 00:03:46,145 stem cells. 74 00:03:46,145 --> 00:03:48,959 Now, my animal colleagues get really excited about stem cells. 75 00:03:48,959 --> 00:03:50,375 Stem cells are nothing for plants. 76 00:03:50,375 --> 00:03:53,095 They've got stem cells to spare, and their stem cells 77 00:03:53,095 --> 00:03:56,915 are arranged in these specific regions called meristems. 78 00:03:56,915 --> 00:03:59,885 They're domes of undifferentiated cells. 79 00:03:59,885 --> 00:04:02,887 So when I tell you that I want to understand floral development, 80 00:04:02,887 --> 00:04:03,845 this is where we start. 81 00:04:03,845 --> 00:04:06,845 We start with a dome of undifferentiated cells, 82 00:04:06,845 --> 00:04:10,805 and we look at how that set of cells changes over time 83 00:04:10,805 --> 00:04:13,335 to generate floral morphology. 84 00:04:13,335 --> 00:04:15,135 So, this is our big problem. 85 00:04:15,135 --> 00:04:18,635 We have about 350,000 species of flowering plants 86 00:04:18,635 --> 00:04:22,535 that have arisen over the past 150 million years. 87 00:04:22,535 --> 00:04:23,705 That's a big problem. 88 00:04:23,705 --> 00:04:26,225 I don't try-- I'm not looking at that scale, right? 89 00:04:26,225 --> 00:04:29,585 You have to start with something tractable, and what we work with 90 00:04:29,585 --> 00:04:33,665 is a model system called Aquilegia-- that you may know as Columbine. 91 00:04:33,665 --> 00:04:36,545 A lot of you may have this in your yards at home. 92 00:04:36,545 --> 00:04:40,152 It is what we call a recently radiated species flock. 93 00:04:40,152 --> 00:04:41,735 So let me explain to you what that is. 94 00:04:41,735 --> 00:04:43,835 Why is Aquilegia a good model system? 95 00:04:43,835 --> 00:04:46,925 So obviously, it's very pretty, but it's also 96 00:04:46,925 --> 00:04:49,595 very tractable from a molecular standpoint. 97 00:04:49,595 --> 00:04:53,045 It has a relatively small genome, about 300 million base pairs, 98 00:04:53,045 --> 00:04:55,445 and you'll take my word for it that that's manageable. 99 00:04:55,445 --> 00:04:57,695 We have a fully sequenced genome. 100 00:04:57,695 --> 00:05:00,065 We also have a lot of functional tools that 101 00:05:00,065 --> 00:05:04,475 allow us to trick the plant into doing things that we want it to do. 102 00:05:04,475 --> 00:05:08,045 It also has a wonderful array of morphological novelty. 103 00:05:08,045 --> 00:05:10,655 So, I mentioned that this is a radiated species flock. 104 00:05:10,655 --> 00:05:11,655 What am I talking about? 105 00:05:11,655 --> 00:05:15,455 I'm talking about that when we study the evolutionary history of this genus, 106 00:05:15,455 --> 00:05:20,135 when we reconstruct its phylogeny, and understand how the species are related 107 00:05:20,135 --> 00:05:25,115 to each other, we see that these species have radiated over a period of one 108 00:05:25,115 --> 00:05:27,825 to six million years, which is relatively recent. 109 00:05:27,825 --> 00:05:31,535 And in fact, the North American radiation that's pictured here 110 00:05:31,535 --> 00:05:33,305 is less than a million years. 111 00:05:33,305 --> 00:05:36,135 So these are very closely related species. 112 00:05:36,135 --> 00:05:38,459 That means that they're also interfertile, 113 00:05:38,459 --> 00:05:41,375 and that's something that makes the geneticists sit up and take notice 114 00:05:41,375 --> 00:05:43,895 because we can cross these species. 115 00:05:43,895 --> 00:05:48,575 Even species that look quite different from one another, we can hybridize them 116 00:05:48,575 --> 00:05:54,065 and then look at how their traits segregate in later generations, 117 00:05:54,065 --> 00:05:58,505 and map the molecular changes that are responsible for the differences 118 00:05:58,505 --> 00:05:59,715 in morphology. 119 00:05:59,715 --> 00:06:02,825 So this means that we have traditional developmental genetic tools, 120 00:06:02,825 --> 00:06:05,585 and we also have a lot of evolutionary genetic tools 121 00:06:05,585 --> 00:06:08,425 that make it a good model system. 122 00:06:08,425 --> 00:06:10,335 OK, so let's step back a second. 123 00:06:10,335 --> 00:06:12,805 I'm going to tell you a little bit about floral morphology 124 00:06:12,805 --> 00:06:13,875 and floral development. 125 00:06:13,875 --> 00:06:17,015 So, this is what you usually think of when you think about a flower. 126 00:06:17,015 --> 00:06:20,165 You may have a little bit of plant biology, maybe in elementary school, 127 00:06:20,165 --> 00:06:23,225 and they told you that there are four different kinds of floral organs, 128 00:06:23,225 --> 00:06:23,765 right? 129 00:06:23,765 --> 00:06:26,285 So you have sepals that are the productive organs, 130 00:06:26,285 --> 00:06:30,125 the attractive petals that are involved in manipulating pollinators, 131 00:06:30,125 --> 00:06:33,005 the male stamens, and the female carpels. 132 00:06:33,005 --> 00:06:36,155 And these organs are arranged in what we call whorls-- 133 00:06:36,155 --> 00:06:37,815 they're concentric circles. 134 00:06:37,815 --> 00:06:41,045 So moving, you get sepals, petals, stamens and carpels. 135 00:06:41,045 --> 00:06:43,835 And about 30 years ago, molecular biologists 136 00:06:43,835 --> 00:06:46,265 working in a number of different model systems 137 00:06:46,265 --> 00:06:50,865 defined a genetic program called the ABC Model, 138 00:06:50,865 --> 00:06:55,685 and this is a model that tells us how each of these floral organs 139 00:06:55,685 --> 00:07:01,852 figures out during development what flavor of origin it wants to be. 140 00:07:01,852 --> 00:07:03,935 So for instance, there are three different classes 141 00:07:03,935 --> 00:07:07,505 of genes, the a, b, and c-class genes. 142 00:07:07,505 --> 00:07:09,635 If you are primordium in the first whorl, 143 00:07:09,635 --> 00:07:13,715 you're only exposed to A function, and that tells you to be a sepal. 144 00:07:13,715 --> 00:07:18,005 In the second whorl, you have A and B function that tells you to be a petal. 145 00:07:18,005 --> 00:07:23,255 Stamens are exposed to B and C together and then carpels to C alone. 146 00:07:23,255 --> 00:07:26,135 So this is what we call a combinatorial code. 147 00:07:26,135 --> 00:07:30,845 Each little position in the floral meristem has a different set of genes, 148 00:07:30,845 --> 00:07:34,065 and that tells the organ what kind of organ to be. 149 00:07:34,065 --> 00:07:36,255 And the way that they came up with this program 150 00:07:36,255 --> 00:07:39,515 was by examining mutants, and developmental geneticists, 151 00:07:39,515 --> 00:07:40,505 we love mutants. 152 00:07:40,505 --> 00:07:41,885 We make mutants all the time. 153 00:07:41,885 --> 00:07:46,175 We like to knock out gene function and then infer from the way 154 00:07:46,175 --> 00:07:47,225 the plant looks-- 155 00:07:47,225 --> 00:07:48,845 or any other organism-- 156 00:07:48,845 --> 00:07:51,995 the way it looks what the function of that gene must be. 157 00:07:51,995 --> 00:07:56,435 And the genes that comprise the ABC program have something in common 158 00:07:56,435 --> 00:07:59,375 that when you knock them out with mutation, 159 00:07:59,375 --> 00:08:02,255 you dramatically change floral morphology. 160 00:08:02,255 --> 00:08:08,105 You transform the identity of one organ into the identity of another. 161 00:08:08,105 --> 00:08:12,335 And so here, for instance, if you eliminate this B-class gene, 162 00:08:12,335 --> 00:08:16,145 then all you have are A and C functions in the floral meristem. 163 00:08:16,145 --> 00:08:20,135 And the petals turn into sepals, and the stamens turn into carpels. 164 00:08:20,135 --> 00:08:23,465 And this is a complete transformation, so this 165 00:08:23,465 --> 00:08:27,305 highlights how minor genetic changes can have really 166 00:08:27,305 --> 00:08:30,795 profound effects on morphology. 167 00:08:30,795 --> 00:08:32,915 So these are the kinds of genetic programs 168 00:08:32,915 --> 00:08:36,465 that we think about when we think about floral evolution. 169 00:08:36,465 --> 00:08:39,305 OK, so how are we applying this kind of model 170 00:08:39,305 --> 00:08:42,935 in the context of our model system, Aquilegia? 171 00:08:42,935 --> 00:08:46,475 So there's a lot of interesting things about Aquilegia flowers. 172 00:08:46,475 --> 00:08:50,315 These guys, up here, those are the sepals. 173 00:08:50,315 --> 00:08:53,075 And usually, you think about sepals as being green and protective, 174 00:08:53,075 --> 00:08:56,135 but these are not just green and protective, they're bright red, 175 00:08:56,135 --> 00:08:58,505 or sometimes blue, or white, or yellow. 176 00:08:58,505 --> 00:09:02,165 They're what we call petaloid, which just means that they're showing. 177 00:09:02,165 --> 00:09:03,825 So this is an interesting question. 178 00:09:03,825 --> 00:09:07,715 Here, we have the ecological function of pollinator attraction 179 00:09:07,715 --> 00:09:11,225 that has been transferred from the inner whorl of the flower 180 00:09:11,225 --> 00:09:13,085 to the outer whorl of the flower. 181 00:09:13,085 --> 00:09:16,955 So how does that happen, and what's going on at the molecular level 182 00:09:16,955 --> 00:09:20,765 in order to make these sepals bright and showy? 183 00:09:20,765 --> 00:09:25,835 Another really interesting aspect of the Aquilegia flower is this nectar spur. 184 00:09:25,835 --> 00:09:30,485 So the true petals in the second whorl have this long, tubular structure 185 00:09:30,485 --> 00:09:34,715 that terminates in a nectary, and you can just imagine all the things 186 00:09:34,715 --> 00:09:39,215 that Aquilegia is doing to manipulate pollinators to make them go all 187 00:09:39,215 --> 00:09:41,075 the way down here and get the nectar. 188 00:09:41,075 --> 00:09:43,775 So there's all kinds of things that we can study here related 189 00:09:43,775 --> 00:09:45,995 to how the spurs first evolved. 190 00:09:45,995 --> 00:09:48,305 They're what we call a key innovation that 191 00:09:48,305 --> 00:09:50,615 were important for the radiation of the genus. 192 00:09:50,615 --> 00:09:54,215 And what is the genetic basis of their diversification in morphology? 193 00:09:54,215 --> 00:09:57,525 They get longer and shorter, and they're curved into different shapes. 194 00:09:57,525 --> 00:10:01,550 So we're studying all these different aspects across species of Aquilegia. 195 00:10:01,550 --> 00:10:03,925 And then one of my favorite things about Aquilegia flower 196 00:10:03,925 --> 00:10:05,965 that I didn't even realize until I started 197 00:10:05,965 --> 00:10:10,405 dissecting them is that they actually have five types of floral organs. 198 00:10:10,405 --> 00:10:14,425 They have this fifth type of organ called the staminodium, which 199 00:10:14,425 --> 00:10:18,025 is a sterile organ positioned between the male stamens 200 00:10:18,025 --> 00:10:19,555 and the female carpels. 201 00:10:19,555 --> 00:10:21,745 So, how the heck does that-- how does that work? 202 00:10:21,745 --> 00:10:24,295 The ABC Model gives us a very elegant mechanism 203 00:10:24,295 --> 00:10:26,845 for giving us four types of floral organs, what 204 00:10:26,845 --> 00:10:30,385 happens when we intercalate a fifth type of floral organ 205 00:10:30,385 --> 00:10:32,365 into that genetic program? 206 00:10:32,365 --> 00:10:35,065 So we've been studying all of these questions. 207 00:10:35,065 --> 00:10:37,374 Another one that I'm going to show you today 208 00:10:37,374 --> 00:10:39,415 is, we've been interested in this question about, 209 00:10:39,415 --> 00:10:42,685 how does the flower know when to stop? 210 00:10:42,685 --> 00:10:44,245 You think about a flower. 211 00:10:44,245 --> 00:10:46,420 Once it makes those carpels, then it stops. 212 00:10:46,420 --> 00:10:49,675 It doesn't make any other floral organs, and that's actually 213 00:10:49,675 --> 00:10:52,435 a very carefully controlled genetic process. 214 00:10:52,435 --> 00:10:54,835 In most of the flowers you might think of, 215 00:10:54,835 --> 00:10:59,425 there may only be one whorl of stamens like in this Snapdragon. 216 00:10:59,425 --> 00:11:01,675 But here, we're looking down on an Aquilegia flower, 217 00:11:01,675 --> 00:11:04,825 you can see that there are many whorls of stamens, 218 00:11:04,825 --> 00:11:08,755 so how does the meristem decide how many whorls it's going to make? 219 00:11:08,755 --> 00:11:11,335 And to investigate this, we started by actually 220 00:11:11,335 --> 00:11:14,785 working with these C-class genes because this 221 00:11:14,785 --> 00:11:16,285 makes sense when you think about it. 222 00:11:16,285 --> 00:11:19,555 It turns out, the C-class genes confer the identity 223 00:11:19,555 --> 00:11:24,445 of the stamens and the carpels, but in conferring identity of the carpels, 224 00:11:24,445 --> 00:11:27,145 they tell the floral meristem to stop. 225 00:11:27,145 --> 00:11:30,385 So, what we want to ask is, what happens when we knock out 226 00:11:30,385 --> 00:11:32,515 the function of these genes? 227 00:11:32,515 --> 00:11:35,365 And in Aquilegia, we use a molecular tool 228 00:11:35,365 --> 00:11:38,875 to trick the plant into silencing our gene of interest. 229 00:11:38,875 --> 00:11:42,685 And when we do that-- here's a wild type Aquilegia flower, and here's 230 00:11:42,685 --> 00:11:46,945 a flower where we have used this trick to silence the C-class gene. 231 00:11:46,945 --> 00:11:50,395 And you can already see it's looking a little weird, but as it develops, 232 00:11:50,395 --> 00:11:53,995 you see that the stamens are completely replaced by petals. 233 00:11:53,995 --> 00:11:57,685 And the staminodium, that novel organ, and the carpels 234 00:11:57,685 --> 00:12:01,375 are both replaced by sepals. 235 00:12:01,375 --> 00:12:04,285 And when we look inside the flower, you may not be able to see this, 236 00:12:04,285 --> 00:12:07,045 but there are actually dozens and dozens of whorls 237 00:12:07,045 --> 00:12:10,675 of organs here, so it's clear that this gene also 238 00:12:10,675 --> 00:12:13,345 controls floral meristem termination. 239 00:12:13,345 --> 00:12:16,135 Now, this was actually just the first step in this project. 240 00:12:16,135 --> 00:12:20,545 We were trying to confirm that this gene functions the same way in Aquilegia 241 00:12:20,545 --> 00:12:22,435 that it functions in other species. 242 00:12:22,435 --> 00:12:26,205 We expected it to be conserved because, in fact, all around you, 243 00:12:26,205 --> 00:12:28,045 you see mutants. 244 00:12:28,045 --> 00:12:30,705 These are all mutant varieties. 245 00:12:30,705 --> 00:12:33,715 So here's a rose, a daffodil, and a peony. 246 00:12:33,715 --> 00:12:36,505 Where on the left, you see the wild type flower 247 00:12:36,505 --> 00:12:40,735 that has a single whorl of petals and many whorls of stamens. 248 00:12:40,735 --> 00:12:43,825 And on the right, you see an Agonis mutant 249 00:12:43,825 --> 00:12:46,855 that has the transformation of stamen identity 250 00:12:46,855 --> 00:12:51,175 into petal identity and the indeterminancy of the floral meristem. 251 00:12:51,175 --> 00:12:53,454 So actually, you're looking at mutants all the time, 252 00:12:53,454 --> 00:12:54,745 you may just not have known it. 253 00:12:54,745 --> 00:12:57,775 So homeotic or transformative mutants have always 254 00:12:57,775 --> 00:13:00,565 been very unpopular in horticulture. 255 00:13:00,565 --> 00:13:04,015 So, I hope I've convinced you that evo-devo is an interesting field. 256 00:13:04,015 --> 00:13:07,062 If you're not totally taken by plant biology as I am, 257 00:13:07,062 --> 00:13:09,395 you might be interested in some of my other colleagues-- 258 00:13:09,395 --> 00:13:13,735 Mansi Srivastava, who studies these fascinating little flatworms that 259 00:13:13,735 --> 00:13:17,545 have the capacity to regenerate their entire bodies from just 260 00:13:17,545 --> 00:13:19,285 small fragments of their body. 261 00:13:19,285 --> 00:13:21,325 They're actually thought to be immortal. 262 00:13:21,325 --> 00:13:24,055 My colleague, Cassandra Extavour, who studies 263 00:13:24,055 --> 00:13:26,225 the evolution of the genetic pathways that 264 00:13:26,225 --> 00:13:30,775 control germ line specification, which is incredibly important for evolution 265 00:13:30,775 --> 00:13:31,885 and development. 266 00:13:31,885 --> 00:13:34,585 Cliff Tabin over at the medical school studies 267 00:13:34,585 --> 00:13:38,515 the diversification of vertebrates and particularly the vertebral column 268 00:13:38,515 --> 00:13:39,115 itself. 269 00:13:39,115 --> 00:13:41,275 Why don't snakes have any limbs? 270 00:13:41,275 --> 00:13:44,005 Why do ostriches have such long necks? 271 00:13:44,005 --> 00:13:47,965 My colleague Terry Capellini over in Human Evolutionary Biology 272 00:13:47,965 --> 00:13:51,625 is actually using evo-devo approaches to understand the evolution 273 00:13:51,625 --> 00:13:53,545 of the skeleton in humans. 274 00:13:53,545 --> 00:13:57,715 So he uses mouse models a lot, but what he's ultimately interested in 275 00:13:57,715 --> 00:14:00,565 is actually human skeletal evolution. 276 00:14:00,565 --> 00:14:04,017 So that's everything I want to tell you about evo-devo, so thank you.