Storer Lectureship feat. Sean Carroll, University of Wisconsin, Madison | March 24, 2009

Intro um anyway so and thanks everyone for coming out uh if you did a second time or if you're here for the first time well you got the better introduction um so for the theme of my talk today I'm going to pick one of my favorite quotes that comes from a physicist from a while ago jean-parent who said that the key to Scientific progress to explain the Theme complex visible by some simple Invisibles did you walk away with a pointer oh there it is sorry um by some simple Invisibles and then the more you think about this I think the more true it is and I really hope um oops this thing is now running on its own there we go um I hope to convince you today that we're making some progress um in exactly this vein by explaining the complex visible diversity of animals with some simple Invisibles and today I'm going to address the central research focus of my lab how does form evolve but of course since animal form is the product of development this question really boils down to how does development evolve now this is not at all a new question you know the uh Central importance of development to Evolution was certainly recognized by Darwin it was recognized by The Architects of the modern synthesis here's a quote Julian Huxley um right out of the preface of Julian huxley's evolution of modern synthesis where he said a study of the effects of genes during development is as essential for an understanding of evolution as are the study of mutation and that of selection but of course in 1942 and probably for more than four decades after that we couldn't say anything about the effect of genes during development we didn't know anything about the genes that govern development until the last 25 years or so now we have learned a great deal in those last 25 years about genes and development that bears directly on our understanding of how development evolves and I want to underscore so much that without the great advances in developmental biology as a foundation without Devo there is no evo Devo and that we can think about the diversity of animal forms as really sort of natural mutants those phenotypes that natural selection has allowed and then the challenge to us as developmental geneticists slash biologist is to figure out where the functionally meaningful genetic changes are that account for the diversity of forms so what I hope to explain today are how we can now integrate this knowledge of Developmental genetics and case studies of species Divergence with consideration of the roles of mutation and selection to arrive at some general genetic rules governing the evolution of animal form but before I present current data I want to take a little while to explain how how and why thinking has developed along the lines that it has so to get everyone all on the same page I want to talk about some of the general findings out of evodivo that set the table and directed research to the path that is now taking and I want to start by asserting that the key Discovery from developmental genetics is that most genes that govern development have many roles this the Mosaic Plyotropy term for this is called Mosaic plyotropy that was coined by ernstadorn in the early 1960s now what do I mean about Mosaic plyotropy plyotropy meaning many many ways these genes have many roles well let's just take a fairly famous Gene like Sonic Hedgehog a vertebrate Sonic Hedgehog homologue of a of a fruit fly Gene called Hedgehog but Sonic Hedgehog has many roles and I'm just showing you a few of these roles visually in C2 hybridization to a developing chicken um so here's Sonic Hedgehog being expressed in the limb Bud where we know it's essential for the formation of the proper polarity of the digits of that developing limb it's also expressed in the neural tube where it plays an important role in organizing various features of the architecture and differentiation of components of the neural tube but later in developments also expressed in the development of feather buds now what do feather buds neural tubes and limb buds have in common nothing nothing at all so the Mosaic plyotropy the fact that an individual Gene would be used in the development of such disparate structures that don't have any sort of embryological relationship to each other they have no evolutionary relationship to each other it didn't have to be this way but this is what developmental geneticists discovered is that there's a group of genes involved in building and patterning body parts that individual genes can have quite disparate roles in involved in the development of structures that have nothing to do with each other well why is this important well then this constrains the genetic path that Evolution can take because the gene that has this many jobs just speaking in terms of gross participation in the development of different tissues let alone the many jobs it actually has at the molecular level um there are going to be constraints on changing on how this Gene can be changed okay so the question that this Mosaic plyotropy then raises is so how are the effects of mutations limited to one body part or one sex how do we Tinker with just feathers or just digits um right or just the head or something like that and not all the other parts of the body that a gene might be uh influencing and so I'm going to assert that the key determinant of the genetic path of morphological evolution is circumventing the plyotropic effects of mutations mutations in Sonic Hedgehog that for example affect the protein activity have devastating effects they have limb defects they have neural tube defects profound effects on the development of animals so if that's the case then how can Evolution Tinker with Sonic Hedgehog how did Sonic Hedgehog for example even acquire these different different roles in the development of different structures so what I want you to be thinking about is the genetic path of morphological Evolution and which paths are allowable and which paths can functionally affect changes such that one body part can can change without others so the key consideration here then leads us Genetic Paths to ask well then what genetic paths will allow us to circumvent the plyotropic effect of mutations and if we think about sort of broad categories of the types of mutations the types of genetic events that happen in genomes we might be asking them what are the relative contribution of different molecular genetic mechanisms to the evolution of anatomy and so let's think about maybe the most obvious way to circumvent the plyotropic effect of mutations one that's been thought about for a long time which is Gene duplication and Divergence obviously by duplication by providing redundant copies of genes that can circumvent potentially circumvent the plotropic effect of mutations because an ancestral function can be retained while new functions can be explored of the duplicate Gene so almost 40 years ago now a very influential book by sisumo Ono emphasized the creative role of Gene duplication and evolution and right here in the preface Ono stated that natural selection merely modified while redundancy created and inside the book he explained how allelic mutations have already existing Gene loci cannot account for major changes in evolution it's quite fair to say that Ono did not conceive of how new forms even new capacities could come about without new genetic information in the form of new genes this is clearly not correct now I must say that this idea was uh pervasive in developmental genetics in fact some most influential articles written in the early days of of Developmental genetics anticipated that Gene duplication was going to be a big part of the story of the evolution of physical diversity of animals but it just turned out not to be the case so let me just review for you a little bit of the empirical data that has turned us away from thinking about Gene duplication as a major contributor to the diversification of form so foremost among these is it contrary to the original expectations that would there would be some association between the expansion and the number of genes and the expansion of animal complexity for example and the expansion of animal diversity duplications of loci and code transcription factors and components of signal transduction Pathways and these really are the key components of the bodybuilding genetic toolkit these duplications have been relatively infrequent over very long periods of animal evolution I'll show you the examples of that in a second but furthermore while I think the prevailing thinking was that Gene duplicates would be free genetic parts to play with there's some evidence that those duplicates may not be neutral for these sorts of genes because these genes participate in dosage sensitive processes and there's actually empirical evidence to suggest that new duplicates may not be neutral and they're actually selected against because of their dosage effects on one or more of the multiple traits that they affect so new genes of this type may not be free they may come with a negative penalty so what's the actual data look like well a lot of the thinking about the role of Gene duplication and Body Evolution has been focused around the Hox genes and we know there's a series of genes that govern the identity of segments and body parts along the major axis of the fruit fly and if we look at some of its arthropod and near arthropod relatives the prediction was that while a fruit fly has many differentiated segments those animals with fewer differentiated segments would require few of these genes that differentiate individual segments but that's not at all the case so a centipede which has a long trunk of repetitive segments of the same identity has exactly the same complement of Hox genes as a fruit fly and anonocopherin these soft-bodied animals that are the sister phylum to the arthropods have perhaps just four differentiated segments again they have the full complement of oxygenes that we know from fruit flies and centipedes and so the simple inferences is that the last common ancestor of lobopodians or onycopherins and arthropods which lived at least 530 million years ago or so and probably predated the Cambrian explosion had this full complement of Hox genes and there's been no new Hox genes added to the arthropod repertoire in 500 million years in fact insects actually have fewer functional Hox genes than their ancestors I can tell you the exact same story for tetrapods relative to for example sarcopterygians that coelacanths have more functional Hox genes or more oxygenes than do tetrapods and the story of tetrapod hock Sheen evolution is a static number of genes or actually a loss of genes in certain lineages so the four clusters of roughly 39 Hox genes that we have have been around for 400 million years and we don't know of any lineage specific additions in tetrapods so all of tetrapod diversity has evolved around essentially the same set of Hox genes all of arthropod diversity has evolved around the same set of Hox genes so if not Gene duplication then what might we also be thinking about here well how about protein sequence Evolution sorry maybe the proteins are Protein Sequence Evolution evolving in some way that we can discern but again here there were surprises in in terms of experiments that developmental biologists did and the most stunning surprise was that certainly Protein Swapping contrary to expectations that homologous transcription factors and Signal transduction proteins from long diverged tax may be functionally equivalent so the observation you may be most familiar with is the experiment where genes have been swapped between for example mice and fruit flies and shown to be functionally equivalent so for example the mouse version of the pac-6 gene is capable to induce the formation of eye tissue in a fruit fly despite the fact that those animals have undergone more than a billion years of divergent evolution the reason why that and this is a finding that's been occurred at least 20 25 times of various bodybuilding genes it's not always the case but it's more the rule than the exception why is this sort of propound we know we know how these proteins work a transcription Factor interacts with various parts of the host proteome and the nucleus chromatin remodeling factors cofactors things like this for the mouse protein to work in a fly proteum it means it's interacting with these components efficiently enough to make the phenotype so it's telling you that those proteins and all the interacting motifs on them have been stable for a billion years and that's not what you would expect if these proteins were evolving adaptively in some lineage specific way such that the proteins were changing they we wouldn't expect them to be functionally interchangeable among such unrelated taxa now we also have some other suspicions why these proteins are constrained one of the reasons why these proteins were so readily identified between different taxes is that their sequence as the translated protein sequences are highly conserved various domains like DNA binding domains may be perfectly conserved over 500 million years or so and we know from developmental genetics that functional changes in most of these developmental Regulators they have plyotropic effects catastrophic effects most of the time if it's loss of function mutations so there's constraints on how these proteins can evolve but the one of the measures of those constraints is that they're functionally interchangeable often over very large taxonomic distances okay so if not Gene duplication and protein sequences Gene duplication and protein sequences well what else well the key idea was anticipated some time ago and there's various roots of this idea in the literature I'm going to show you one very influential paper that came from Mary Claire King and Alan Wilson in 1975. and what they did is they looked at a very small set of proteins of chimpanzees in humans some they had sequences for some they had immunological data some they had electrophoretic data so this is sort of the first paper in comparative genomics although the word didn't exist and they weren't looking directly at genome sequences and they looked at these proteins of chimps and humans they realized that they were nearly identical or identical in most cases and the conclusion they reached was with Gene we couldn't really explain the behavioral and anatomical differences between these two species at you know with this minimal or zero molecular distance that we saw in the the proteins of these species so they suggested that evolutionary changes in anatomy and way of life are more often based on changes in the mechanisms controlling the expression of genes that on sequence changes in proteins now unfortunately um Alan Wilson was not able to live to see the data to emerge to test this hypothesis but there's now a vast body of data from evodivo from this evolutionary developmental biology where we've been looking at the expression and the deployment of genes in taxa at various distances and see uh usually a very close correlation between differences in body form and how particular genes with particular functions are deployed so there is a fair amount of supportive evidence now not in 1975 but by the mid-1990s I would say that or in the late 1990s that there's um a good evidence that changes in the expression of genes are associated with Gene expression and anatomy changes in anatomy so what does that boil down to in terms of the genetic path of evolution how is the expression of genes controlled well of course it's controlled by proteins that govern the turning on and off of genes but I've already told you about some of those proteins and how they're not changing but those proteins act through regulatory sequences in the genome so-called CIS regulatory sequences and so we think that um a reasonable possibility or have been thinking for quite a while a reasonable possibility for how changes in gene expression could evolve would be through changes in these regulatory sequences and so attention has turned increasingly towards CIS regulatory sequences as units of evolution and why have they done it so before I tell you anything about data sort of from an evolutionary framework let me tell you what molecular developmental biologists know about these CIS regulatory sequences which played a large role in our thinking that these indeed might be critical to the evolution of morphology first is that and these are CIS regulatory sequences well-established body of data from developmental biology that the spatial expression of virtually every developmental regulatory Gene is controlled by sets of modular independent CIS regulatory elements now you have to start to develop a picture in your mind of what I mean by modular independent CIS regulatory elements and to do that I'm just going to show you CIS regulatory elements sort of a hypothetical Gene what I mean by that so imagine a gene that's used in the formation of a few different cell types or tissues say in a human the it's used to say build the brain and the spleen the kidney and maybe to differentiate red blood cells if you looked at the coding region of this Gene it might occupy a pretty modest base just like every other protein but associated with it would be non-coding regulatory sequences and the regulatory sequences that govern the expression of that Gene in each particular part of the developing body would be discrete from regulatory sequences that govern expression and other parts so the brain you might have an element the controlled expression in the brain one that control expression in the spleen one of the kidney one of the blood cells and so forth what do I mean by these elements these elements are generally let's just call them on average let's say 500 base pairs of DNA would govern say expression in the brain that's actually a fair amount of information because what these elements contain are binding sites docking sites for regulatory proteins and very often to specify expression in a tissue like the brain at a given developmental time point there are several inputs that need to come into that regulatory element to govern uh gene expression at the right time and place so for the bodybuilding genes I've been talking about very often the regulatory information is much larger part of a much larger fraction of the locus than is the coding information so these are what I mean by these modular CIS regulatory elements and they have the property sorry I'm going to Regulatory networks just stay another thing uh another uh water the brain got dehydrated right in the middle of that sentence and forgot where he was okay so what else do we know about these regulatory proteins and the regulatory elements well really for the last few years we didn't quite appreciate just how big how vast the regulatory networks were that they were part of now what do I mean by being parts of regulatory networks imagine this is a transcription Factor so the way it influences the development of these tissues is it goes on and either turns on or turns off other sets of genes now how many genes might it turn on or off it turns out with development of new Transcription factors technologies we now realize that transcription factors developmental transcription factors are parts of vast regulatory networks that contain dozens to hundreds of direct Target genes so this is plyotropy on a vast scale that even you know most of us did not conceive so we know examples of a trend of transcription factors that might regulate say four or five hundred different Target genes directly in a developing animal so and they might be doing that in multiple tissues at multiple times so if we think of this as a transcription factor maybe acting these different tissues what we're trying to say is this protein encoded by this Gene might then go and in say developing erythrocytes regulate a series of genes developing kidneys and regulate a series of genes developing spleen cells regulate a series of genes developing brain regulate a series of genes so this is what I mean by a picture of plyotropy is these proteins are parts of fast networks so the change gene expression to evolve some change in this network for the network to evolve it means we have to change linkages between transcription factors and the target genes and these Target genes of course have these CIS regulatory elements through which these transcription factors add so so how do we do that the trick is you know how do we change one linkage without altering all of the others so if we change the protein prediction is going to change most of these outputs but there are ways to change one linkage without altering others it may be obvious to you by now it's it changes in one CIS regulatory element mutations in a brain element or mutations in a spleen element mutation is the kidney element they don't affect the function of other CIS regulatory elements they're minimally plyotropic so mutations in individual CIS regulatory elements have the property again this is from developmental biology alone and a vast amount of Developmental biology that I realize is largely outside the view of working evolutionary biologists but nonetheless this is what the the progress that's made in developmental biology that these mutations are minimally plotropic they affect the deployment of a gene in just one tissue at one point in time and not all the others and they don't affect the activity of the protein so this is why we've suspected that these vast regulatory regions with these modular regulatory elements are the site are are the regions where the evolutionary of most relevant evolutionary action in these bodybuilding genes and while that was a suspicion the challenge has been to directly test that idea and the the challenge in testing that idea has been that um a lot of evodivo studies comparative studies were done over pretty large taxonomic scales in the 1990s you know comparing things like you know pythons and chickens and fruit flies and butterflies and things like that as it turns out it's very hard to isolate these regulatory elements in non-model species and it's also true that these regular if you're talking about traits um you know some certain traits don't change very much over pretty vast periods of time so if you want to study the evolution of limb number and insects and it's been six six six six six you know you don't really have something to study there in terms of the evolution of diversity so you need to study traits that are actually changing but changing on a time scale where you can actually get at these functional bits of regulatory DNA so we had to change our strategy somewhat beginning In fact when our team joined the lab was to study traits that were rapidly evolving in a study closely related texts that were genetic and transgenic approaches Changing strategy were going to become feasible and so the traits we chose to focus on was pigmentation patterns they were pigmentation patterns in drosophilid flies now I know you all just need to take a little breath for a moment yes you're giving up the next half hour of your life to color patterns in fruit flies but after all that grandiose thinking of the Animal Kingdom here's my argument my argument is that these patterns they're complex enough to be interesting they're spatial patterns restricted to particular body parts um they're uh good surrogates for more complicated things that might involve for example shape shape is a really tough thing to get at developmentally but these two-dimensional color patterns are a little bit simpler and when I say they're simple enough to be tractable it means that the number of genetic inputs in them is probably modest enough that we can identify some or all of the components that govern their formation and change but as it turns out we weren't sure about this but thankfully this is true these patterns are regulated by some of the very same top Regulators the big guns of animal development that evodivo's been thinking about for 25 years that shape body plans so even the big guns worry about you know the color of a fly's rear end amazingly so do I um okay so um what I'm going to do for the rest of my talk is I'm going to show you a few condensed case studies regarding the evolution of pigmentation patterns and I'm going to work from simpler to progressively more complex examples and increasingly greater taxonomic scales and I'm going to argue from this evidence and from the background I just gave you about the big picture of Gene duplication and functional equivalence and changes in gene regulation across taxa that some general principles of molecular evolution can be derived before I do that I want to tell you about the people who've done this work over I'd say roughly a decade span so here are current or recent Collaborators members of the lab whose work I'm going to highlight saying Zhang Tom Williams marker base and Tomas Verner in particular with help from Vicki Kastner and Jane sealig here's a list of alumni and collaborators I think you know this fellow and the last chapter I'm going to tell you about was started by RTM but former members of the lab Nicola gampa Benjamin Prudhomme John true Tricia whitcop and tonus rokus our team that all have their own labs and key collaborators particularly from the population genetics and ecology expertise side including John Poole who's spending some time here at UC Davis but I don't think you can be here today for the talk but you're right away going to see some things that John has collaborated we've collaborated with John in doing so um let's get started and let's get started uh in Africa and let's get started with the most familiar fruit fly to anyone in In Africa biology but as it turns out it's a bit of a variable beast and what John Poole and chipaquadro documented was that in East Africa there is a quite a range of pigmentation phenotypes in drosophila melanogaster and if you just focus here on the abdominal pattern or any individual segment these numbers we don't get too hung up on but these are pigmentation indices that John measured and the idea is that there's quite a range of pigmentation intensity from lighter to darker flies all flies have these dark bands on the abdominal segments but you can see either the more just ahead of the dark bands how dark the individual segments are varies quite a great deal up here to the to the the darkest form that they isolated and they think that this phenotype of the the darkest flies is adaptive because they saw an association between the elevation at which they found these flies and the pigmentation index so this non-random distribution of Darker flies suggested that there might be some logical reason for the darker flies to exist at at higher altitudes now they are also able to show a genetic Association of this phenotype with a particular pigmentation Gene in fruit flies and before I tell you a little bit more about that I want to introduce you to the sort of the three three star genes for most of my talk I realize that for most of you you Fruit Fly Genes know fruit fly jeans are not your favorite nomenclature but I have to use these names I don't want you to worry about the pathway God no I just want you to just focus on these three things I've circled here the three genes I'm going to talk about are ebony tan and yellow but here's your brain has to work upside down because fruit fly geneticists often name their genes from mutant phenotypes but the wild type function of these genes is opposite then to the mutant phenotype so ebony actually inhibits dark pigmentation and promotes the formation of a lighter yellow pigment so mutations anemone become darker but anyway Ebony's wild type function is to promote lighter pigmentation and I think is my battery fading or just you guys can still see it okay and the yellow and the tan jeans both promote thanks both promote darker pigmentation so all you really have to know in these dark circles these are genes that promote darkening and this is a gene that promotes lighter pigmentation okay and what um John Poole and Chip Ricardo showed was that there was a genetic association between variation at the ebony Locus and that dark pigmentation phenotype in um drosophila melanogaster and so uh Mark joined the the battle now to figure out exactly what was going on at the ebony Locus that would explain the variation in these fruit flies and one of the first things he looked at so these are the the most extreme light and dark lines that John isolated he said okay well how is the ebony Gene deployed in these two lines and this is in C2 hybridization this is a very reproducible sort of thing and this is showing you that ebony expression is high shown here in purple is the signal of ebony transcript in light lines but considerably lower in dark lines now so remember when you have less ebony you have a darker fly okay so this is very reproducible so it's showing you that there's a both a quantitative and a spatial aspect of this of a change a difference in Ebony expression between the light and dark lines so then he's set to mapping how could this occur Transgene Experiment and what he did was he did a a transgene experiment sort of a complementation experiment to look at which parts of the ebony Gene from light and dark flies was responsible for the phenotype so what he set up was a situation where it flies a trans a host had no ebony activity there was an ebony mutant which these guys are really dark and you can show see this sort of a close-up on an individual segment these segments are really dark and then he took the ebony Gene from either the light strain or the dark strain and said well how does this rescue the ebony phenotype well the gene from the light strain makes these guys much lighter pretty much a full Rescue of the of the ebony phenotype but the strain sorry the gene from the dark strain does not these flies are still dark and then he made just chimeric genes where he took the Upstream part of the light Gene and the downstream coding part of the dark Gene and vice versa and said well what how the complimentation worked there well if you had the Upstream part of the light allele and the downstream part the coding part of the dark allele these flies are now light and if you took the Upstream part of the allele and the downstream coding part of the light allele these flies were dark so what this tells you is the the meaningful differences between these two alleles map outside the coding region in fact Upstream of the gene um somewhere between Ebony and the next Gene here in the drosophila genome so this is for us the typical sort of evidence we'd seen in the past of a difference in non-coding regulatory sequences affecting the activity of this Gene in the developing abdomens of these flies and so we wanted to map this in more detail and then we had to really roll up our sleeves and figure out where Regulatory Elements all the regulatory elements are that govern ebony expression and this is just something to appreciate in terms of understanding how regulatory sequences work and their contribution to evolution is that there's no code for these things there's no algorithm that gives you the location of these regulatory elements they have to all be found empirically marching through sometimes pretty vast tracts of non-coding DNA to identify them and Mark was able to identify a whole bunch of different elements that act independently of each other in the developing genitalia the hall tier which are the hind wings of the fruit fly the four Wing in various bristles in legs and part of the head and parts of the thorax and the abdomen so the ebony Locus contains many modular CIS regulatory elements and then he looked at the activity of these elements between light and dark strains and I'm just going to show you a little Montage as we walk through these various elements and the way you monitor the activities elements is to hook the regulatory DNA to a reporter Gene in this case a green fluorescent protein and look at where in the body the reporter is expressed and at what level so this is expression of a reporter in the head and it's equivalent between the light and the dark strain regulatory elements regulatory DNA same with bristle driven expression in the in the legs same with expression in various parts of the sternite and sorry about the spelling of the genitalia you know genitalia is where the genes come from um uh this is the hind Wing again expression is equivalent in those tissues this is the thorax expressions equivalent in those tissues this is the four Wing expression is equivalent to those tissues but when you compare expression in the abdomen of that regulatory element there's a clear difference the light strain that regulatory DNA drives a much higher level of the reporter than does the orthologous piece of DNA from the dark stream okay I wanted to walk through this to you and you're not going to get this kind of detail in any of the other studies I'm going to tell you to give you an idea what are you seeing here you're seeing evidence where there's been changes in the activity of one of these modular regulatory elements just this one what the element that governs expression in the abdomen the activity of the Gene and all these other tissues which is controlled independently is all the same right and there's no evidence for any participation here in coding changes in differences between these Changes in Differences strains so what's Mark been able to do well once we've identified that there were changes in this regulatory element he's been able to pinpoint the meaningful substitutions and what John Poole had done previously he had noticed in fact one reason we were drawn to this work was that John had noticed that throughout Uganda the dark lines he isolated shared a common haplotype that was quite extensive not just involving a chunk of the ebony Locus but even beyond that and this looked like very good evidence for a selective sweep at the ebony Locus that uh for this dark phenotype and all these dark lines shared a small number of substitutions and we've gone through and tested each of these substitutions for their contribution to the level of Gene regulation and so we know what the individual contributions of these mutations are some appear to be neutral but a couple definitely contribute quantitatively to gene expression and there's one of them here for example that contributes most of the variation between the the two strains and there's another little piece of the regulatory element that also contributes to differences all I want to give you the sense is is that we're able to now pinpoint they're about nine nucleotide differences between the light and dark strains over the meaningful part of the regulatory element and at least three of those but probably not all nine have contributed quantitatively to the differences in gene expression now that's not very interesting arithmetic it's just to give you a sense that how to two alleles differ how to two alleles and you know in Uganda differ from one another Well a few meaningful substitutions in one particular regulatory element this is this is the abdominal element um uh are contributing to this difference in in pigmentation that we think is adaptive both for the non-random distribution of the Flies and for the genetic signature of what looks like a selective sweep at the ebony Locus okay so that seems to be the story I mean that's one story for you from within species um let's talk about Divergence between species so let me give you I think a pretty simple case study we're going to swim swing over swing over to the west coast of Africa to the island of santome and we're going to look at these two flies found on these islands uh drosophila yakuba and drosophila Santa Mia and we know that Santa Mia is derived this is the ancestral species that gave rise to this fly and this fly is pretty obvious to you what's different about it well one of the obvious features about it is lost most of the pigmentation of the abdominal cuticle and what we did was given these are again closely related this is a fly that evolved from this parent parental species is we wanted to understand what were the contributions Tan Gene what was the genetic basis and I'm going to dial right in to the meaningful Gene it's a different Gene that I've told you that I've told you about in the introduction the gene tan so this is a gene whose wild type activity is to contribute to pigmentation we were suspicious that tan could be involved for a variety of reasons but I'm just going to show you the the sort of the uh critical experiment that is that tells us that it's tan it's sort of like the ultimate qtl experiment where you suggest that Gene is involved but what we did is we actually just transformed in the entire tan Locus from a Darkly pigmented fly into drosophila santamia and so that required you'll see why the this whole construct including all the tan exons and Upstream regulatory sequence even Beyond this Gene and if we put in this I'm showing you fly pigmentation in a little different format this is if you butterfly to fly much as you butterfly a shrimp this is an easier way to look at pigmentation and if you look at this is the wild type santamia male and if you put in one copy of the tangene you start to get pigmentation back in the posterior segments a little bit of the stripes two copies you get more pigmentation back Etc so this is just Visual Evidence that in fact there is a meaningful difference in the activity of the tangene the contribution of the tangene to the drosophila yakuba versus Santa Mia pigmentation okay well this is the meaningful stretch of DNA where are the functional changes well they're not in the coding region because if you look on the two lower lines here this is the alignment of the tan coating sequences between drosophila yakuba and drosophila santamia and there's not a single amino acid replacement throughout the protein okay so if you did say androsophilus said to me if you did knock out the tan protein it would be a much lighter fly but tan just like ebony just like all the other genes I'm going to tell you about tan is plyotropic it has other functions for example Envision and in the nervous system as well as in the pigmentation of other body parts so you'd affect all of those traits not just pigmentation of the abdomen so we don't see changes in tan between these two species in the coding region so that must mean something is going on elsewhere and it's pretty obvious what's going on when we look at expression of the tangene this is expression of the tangene seen by in C2 hybridization and drosophila yakuba both the females and the males you see high levels of tan expression in the rear ends of these flies but if you look over here in Santa Mia they've lost expression so if the genetic difference is mapped to the tan Locus if there's a difference in expression no coding changes guess what suspicion is this is these are again changes in regulatory elements of the tangene we were able to localize one of those regulatory elements that governs the expression of tan in the abdomen and it was inconveniently located between two other genes that's part of the other game that these genes play as regulatory elements can be nested in introns they can be sitting by other genes they can be Upstream they can be Downstream in this case it's nested between these two genes a good number a good distance Upstream of the transcription start of the tangene anyway so here's the point if we take this little piece of DNA and hook it to gfp that piece of DNA from melanogaster draws a beautiful pattern of expression of the reporter in the abdomen of these flies if you take the same piece of DNA from drosophila yakubit does the same but if you take the same piece of DNA from drosophila santamia it's dead as a dorneo okay so there's a selective loss of tangene expression in the rear ends of drosophila Santa Mia in the abdomens which account for the pigmentation differences between these species okay now the reason I showed you these two examples one example intra-specific one example inter-specific is I thought they would sort of allow you to see the most of the elements that not only just go into the experiments go into our thinking of why we think changes in these non-coding regions these modular CIS regulatory elements play such a major role in the evolution of gene expression and of visible form it's because restricted change is restricted to one element only affect one particular trait one particular feature of that Gene's function now this is I didn't present this to you in any historical way these are actually two relatively recent examples this was just recently published in the ebony story I told you about is not published yet but we've been Gathering case studies of these maybe about 10 more from my lab and there are other examples from for example stickleback fish and from mice and from drosophila where either the association of non-coding sequence Divergence has either been directly demonstrated or reasonably well inferred from genetic analysis we have maybe 20 case studies and each of those case studies all the genes involved are plyotropic and the definitions that I've given you and we don't have a single case of a plot of a coding change in a plyotropic gene accounting for Divergence between closely related species or within species variation So based on all the background that I gave you in a bunch of case studies I'm going to spare you the details of but I just summarized for you I think we can arrive at a general rule a genetic rule for the evolution of form that when these following conditions exist and the Genetic Rule conditions matter the gene product functions in multiple tissues mutations in the coding sequence are known or likely to have plyotropic effects and the locus contains multiple CIS regulatory elements then the regulatory sequence evolution is the more I'd say far more likely mechanism of morphological change than is coding sequence evolution okay so I I'm comfortable enough with this rule that for us the question is really changing not from why CIS Case Studies regulatory Evolution predominates but how CIS regulatory function actually evolves and so what I want to do is to close up by showing you two more case studies that look in in more detail about what's actually evolving in in CIS regulatory elements and the first I'm going to highlight the formation of new regulatory linkages by co-option to generate an evolutionary novelty and then a sort of a surprising story I think about how one can remodel CIS regulatory elements and generate new functions so if you like maybe one way to characterize the the last two case studies in these is the last two case studies were sort of the modification of existing patterns and in the next two case studies I'm going to tell you about new patterns emerging through CIS regulatory sequence evolution okay so the story I'm going to begin with here are two drosophila species the males of each species see if you can pick out the obvious morphological difference between them so in our strategies of picking which species to study you know we try to find you know the simplest example of the phenomenon we want to understand in this case it's the presence of the wing spots in this species but not some of its close relatives but before I show you how these spots form let me show you why we think these spots exist to show you this little film clip from John true this is the male of the species stepping out in front of the female I'll show you this to you again but anyway so flies that bear these spots um do this little courtship dance in front of the female so whoops let's try this again Okay so um John true is continuing to work on the meaning of this courtship dance and the meaning of these spots but it's a reasonable inference that this may be a pattern that's under sexual selection or has been under sexual selection I really am trigger happy today um so that's why we think the spots are there uh now let me tell you a little bit about how those spots evolved and before I show you the data I want to show you the model if you get this model you kind of get one of the general pictures I want to give you today okay so I'm going to show it to you in schematic form and let's ask how would you get a new spot in a wing that was previously unspotted and we know one of the genes that promotes the expression of that spot or that pigment Gene called yellow we know the ancestral condition before the formation of these spots was a sort of light expression of this yellow Gene across the wing so this is the ancestral condition this is the derived condition of spot formation in a bunch of species that do this courtship dance how did this come about well imagine which is the truth imagine that there are there are regulatory proteins transcription factors deployed in various patterns in developing wings and in fact turns out there's tons of these patterns they're cryptic to RI they're not necessarily associated with any morphological landmarks but imagine one of those proteins would say an activator in purple and another protein is there is a repressor in green and that binding sites for those respective transcription factors occurred at some point just a bit random in a the regulatory element to govern this more uniform expression across the wing well with those new binding sites and those the inputs of these transcription factors into this already existing element if this is an activator and this is a repressor the combined effect of those two inputs would be to upregulate the expression of the gene in this sort of distal quadrant of the wing and that's exactly what we think has happened that already existing regulatory element and already existing transcription factors have evolved a new interaction simply by simple sequence changes in this regulatory element that generates a new pattern of gene expression and a spot for sexual selection or natural selection to play with okay now let me show you the evidence for that so if we look at the expression of the Yellow Gene yellow Gene between unspotted and spotted flies it's pretty obvious spotted flies Express this yellow Gene this is late pupae when just before pigmentation is going to fill in in late pupia it's expressed at very high levels okay in these developing wings and flies that are sort of outside the spotted clay they don't do this this is drosophila obscure and melanogaster doesn't do this either so it's clear that there's a difference in yellow expression between these species with this novel spotted pattern in the spotted species well we're able to map in a regulatory element of the yellow genus activity so we're able to show Again by hooking regulatory DNA up to Green fluorescent protein that while the athogous piece of DNA from drosophila melanogaster gave you a uniform pattern across the wing blade and the orthogous DNA from pseudo obscure gave you a uniform pattern across the wing Blade the unspotted species the spotted species had elevated expression driving elevated expression of the reporter in this quadrant of the wing and we're actually able to tease out which part of this regulatory DNA was responsible for the spot pattern and let me just tell you a little bit more of that spot pattern so we can inform infer a little bit more about the process and the genetic path of evolution Spot Pattern one of the questions you might want to ask and don't look at this for a second um one of the questions you might want to ask is well how many steps would it take to to put a spot on a wing now if there were transcription factors already expressed in spots it'd be pretty easy all you need to do is evolve an interaction between a spotted transcription factor and a regulatory element bang you get spotted patterns of gene expression the other way to draw a spot might be if you have both activators and repressors already existing and that there you get binding sites for both evolving in some number of steps and that the combined input there would be to draw a spot and that's what's happened now the way we got tuned in the way we got a hint that that's what was going on it's really a simple set of experiments where we just manipulate this regulatory element we start trying to tease apart where the inputs are into this regulatory element so this is just bashing away it's this regulatory DNA and seeing what happens when we remove various bits of sequence and the two informative experiments were that when we removed a little piece of this regulatory element this element I'll just refer to as the spot element that drives expression up here in the front half of the wing that as we remove some DNA we saw expression bleed out so it was now also expressed starting to be expressed at higher levels in the back half of the wing so this is the front half of the wing up here what we call the anterior compartment this is the posterior part of the wing so that tells us that there's some negative input here removal of DNA causes the pattern to expand that's Telltale sign that there's repressive input and on the other thing it's also positive input because there were other sequences that were mutated we blew away expression activity of the of the element altogether okay so that's telling us there's positive and negative inputs so what well tells you that this is at least a multi-step multi-hit process modifying the activity this element but it also gave us a hint as to what one at least the identity of one of these regulators and there's a more important broader lesson from the identity of one of these regulators those of you who know a little bit about fruit fly development I told you about the front Fruit Fly Development and the back half of the Wings deliberately is because we know a lot about the subdivision of fruit fly wings and in fact insect wings and insect segments in fact arthropod segments into front and back compartments anterior and posterior compartments the gene who does that is a gene called engrailed a transcription Factor Engrailed expressed throughout the posterior compartment not just the fruit fly wings not just of insect wings but of all arthropod segments and the close correspondence between where engrailed was expressed and how the expression of the how the activity of this regulatory element was restricted to the anterior compartment suggested to us that maybe with a negative input was this protein engrailed so how do you test an idea like that well we look for in grail biochemically we look for engrailed binding sites in this regulatory element I'm showing you those here in yellow if we mutate those in grail binding sites to a different sequence so they now no longer can buy and grilled this is what happens the expression of the reporter Gene is now de-repressed in the posterior compartment so it's repressed here de-repressed in the posterior compartment so that's telling us that one of the direct inputs into this pigmentation Gene is this ingrail regulatory protein so what this is getting worse by the minute okay no this is the so what then Grail protein is not a dedicated pigmentation Gene it didn't have anything to do with pigmentation of the Wings until this little group of flies evolved about 15 or 18 million years ago and it then got recruited into the process of deciding where the pigment pattern was allowed and not allowed in a developing Wing this Gene has been around in insect wings for over 300 million years and it just picked up this job in a little clay to fruit flies as I said in the last 15 or 18 million years and there were no changes to the ingrail gene nothing's necessary to change in Grill it's a good old transcription Factor expressed in the rear end of segments that posterior half a segment for hundreds of millions of years it's already available no changes have to happen at the engralled locus all that has to happen is a binding site for engrailed or multiple binding sites for Ingrid have to evolve in a regulatory element and The Binding sites for these regulatory proteins are simple enough that there's lots of sequences in any stretch of DNA that are just one or two base changes away from being a decent binding site so the ability of new binding sites to evolve the probability of new binding sites to evolve is very high in any reasonable size population okay so that's why we're excited about this story this is an ancient Gene picking up a new job tinkering with wing pigmentation patterns so let me shift to the almost The Last Story Sexual dimorphism again a story of novelty but there I told you a case where already existing transcription factors in our existing regulatory element if you just get binding sites to evolve in the regulatory element you now have a new regulatory interaction can draw a new pattern now I'm going to give you just a slightly different theme one that surprised Us in terms of the generation of a new pattern this has to do with the origin of sexual dimorphism um many of you here of course would be quite familiar with sexual dimorphism in fact some of these models um the good news is you know sexual dimorphism is widespread throughout the animal kingdom the bad news is the story I'm going to tell you about not any of those Majestic animals I'm going to tell you about fruit fly sexual dimorphism I'm going to tell you about pigmentation again drosophila melanogaster is the show horse for this because again folate drosophila abdomens in species like drosophila melanogaster the female has this stripe pattern on the actually you look familiar this is RTM slide he must be thinking wow this is an old one this is classic 2000 2001 there buddy so this is in fact a study that our team pioneered um that these flies uh the males have this dark pigmentation in the posterior whereas fairly related flies are monomorphic with respect to pigmentation where the males and females are pigmented identically so what do we see here we see segment specific and sex-specific pigmentation and for drosophila geneticists this Rings all sorts of bells because we know about the genes and genetic Pathways that govern sexual identity and segment identity so the suspicion is is this pattern is somehow being influenced by the sexual identity pathway and by segment identity genes so how is this done how did pigmentation evolve to be male and segment specific well let me just show you in sense form and then show you a little bit of the data it's via the evolution of sex and segment specific regulation of a particular regulatory protein actually a pair of proteins encode at a certain Locus called bricabrack I'm just going to call it bab now I have to tell you all this because again I got to turn your mind upside down bab is a repressor of pigmentation so what has to happen is in order for those males to be pigmented at their rear ends you have to repress a repressor all right a lot of double negative logic in biology this is another example let me just that's sentence form let me show you visually again rtm's data so in females this uh these bric-a-brac repressor proteins are expressed in all the segments and this suppresses just the sort of male specific form of pigmentation these other Stripes can form but uh the the heavy pigmentation is turned off throughout these segments but in males bricabrack is regulated differently and you can see it's down regulated in the last couple abdominal segments and that allows for pigmentation to take place in those males so the difference in dimor in sexually dimorphic species is that this Gene is on throughout the female posterior and in this case uh an off in the male posterior okay so what do we know about this well Regulation the regulation of this Gene let me just step back for a second the only way we can you know dissect this is we have to figure out all the functional bits the regulatory parts of the brick-of-brac locus and The Regulators of Bricco bracks are going to need it sort of piece together this story this was a bit brutal our team can tell you some stories um he started the work on the molecular work on the bric-a-brac Locus but this thing was a bear and what we didn't realize was that the total expression pattern of bricabrack you know we just say well this is bricabrack in the abdomen you might think it had an abdominal enhancer but it turns out that pattern is a composite of two of the outputs of two different regulatory elements there's a regulatory element that governs expression of this Gene in the anterior of most segments we cleverly call that the anterior CIS regulatory element and there's a regulatory element that governs expression of the gene in the posterior segments and this regulation is sexually dimorphic we call this the dimorphic element okay this is this is probably tenfold more than you wanted to know but just kind of focus on patterns here and then I'll I'll get to the take-home point what's really different here is that you have an element that's being regulated differently between the Sexes and in a segment specific way so this CIS regulatory element is somehow integrating both sex specific and segment specific information so we had to figure out what those inputs are this is the meaningful bit that gives you sexual dimorphic pattern of this Gene well what could those inputs be well segment specificity rings the bell that that could be Hox input we know the Hox genes that govern the identity of particular segments in the fly you can see them from this cartoon color matched and the rear end of The Fly is governed by a particular Gene named abdominal B and we in fact know that abdominal B is one of the direct inputs into that Gene well we also know the inputs that govern sexual identity and the ultimate effector of sexual identity and somatic tissues like these is a gene called double sex and there are sex-specific isoforms of that protein that have different activities in males and in females and the way you get that sex-specific pattern of the activity of that enhancer is actually by integrating those two outputs so I'm saying you all the biochemistry and all the rest of the molecular biology just to draw you a picture of a genetic switch that controls the dimorphic expression of this Gene and the activity of the switch is different between two Sexes for these reasons so just focus over here on the left hand side of the figure this is just to explain to you how the activity of the gene is different what's going on is that in females abdominal B and the female specific of double sex collaborate to activate the gene in the last couple of segments and that represses the male specific pattern of pigmentation so okay so bricabrack is on in the female in the last couple of segments that does not allow the heavy pigmentation in those last couple segments in males however the male specific isoform of double sex is a repressor and so it the combined input here of these two proteins gives you repression of this Gene in the posterior part of the male and that allows for pigmentation all right so this genetic switch is responsible for the genetic switch being both the Cisco regulatory sequence and the proteins that are binding to it is responsible for the pigmentation difference okay how did this evolve how did the sex and segment specific regulation of brick of Brack evolve this is unique to these sexually dimorphic species where did it come from well you might think since there's two regulatory elements one of which is sexually dimorphic maybe these flies evolved a whole new CIS regulatory element for controlling dimorphic expression the other possibility is maybe a regulatory element gained inputs from Hox genes and from double sex and we were wrong in both cases it's not what happened what happened was that an element had been around that had been around for a very long time was remodeled now let me explain what happened was was that dimorphic bricabrack expression evolved and expanded by a number of segments by the molecular Remodeling and ancestral dimorphically regulated CIS regulatory element what I mean by remodeling is changing the number and polarity and topology of The Binding sites for these two key Regulators abdominal B and double sex so I just want to show you the change visually not get into many more details this is the difference between monomorphic and dimorphic species in monomorphic species it turns out that regulatory element I've been telling you about does exist and it drives expression in the posterior most part of the female this little smidge of tissue what's happened in dimorphic species is that expression has now expanded two full segments up the anterior posterior axis of these flies that's a pretty important shift a couple of segments this is the kind of stuff that goes on in body plan evolution here an anterior shift to two segments how did this happen it took a nasty nasty nasty set of experiments to sort this out but Tom Williams had the backbone for it he analyzed the inputs of double sex and abdominal B in these regulatory elements of these two species and what made it nasty was for example there were 14 abdominal B binding sites in this regulatory element two double sex binding sites here's the summary of what he's seen the activity of this element differs between monomorphic and dimorphic species and I'm showing you abdominal B binding sites in yellow and double sex binding sites in white what's happened well you have changes in the number of binding sites a loss over here and one lineage gain in another lineage you have a reversal of a double sex binding site that polarity reversal accounts for a 30-fold difference in gene expression it's the same binding site just in two different orientations pretty remarkable and some other gains of binding sites over here so in the ancestor this of this dimorphic species this element existed and had double sex and abdominal B input but merely by changing the polarity of sites a couple tweaks of the number of sites and in fact the spacing between these sites the output of this regulatory element has changed dramatically and expanded two full segments of the body axis okay so just merely and and just for those of you into informatics and you know genome informatics if you are scanning these species you would say this was a highly conserved regulatory element these binding sites are collinear now over tens of millions of years throughout the regulatory element so any signal you look at you'd say oh that's the same regulatory element but it's dramatically different in its activity in Vivo so a little cautionary tale there that extensive sequence conservation does not mean functional functional identity of the elements okay well why do I get all excited about this well I get excited about this because this shifting of gene expression up the main body axis this is the same sort of thing we see going on at the body plan level the deployment of for example Hox genes invertebrates and arthropods um new domains of gene expression evolving so we're very happy that here by just looking at this you know little patch of pigmentation on the rear end of a fruit fly I think we have a very good model for a general phenomenon in the evolution of development which is moving around a hawks regulated trait in this case expanding pigmentation up a couple of segments so you know we think that the regulatory mechanisms that are involved in the small scale Divergence of species that I just showed you are also involved in large-scale body plan divergences because that's exactly what we've inferred from a whole lot of comparative data in evodivo furthermore climbing another step out on the limb I think the evolution of CIS regulatory DNA sequences are sufficient to explain much diversity and to account for the great constraint we see in the evolution of these proteins so I'll go so far as to say you know this uh great diversity of wing patterns Evolution that we see for example in drosophila flies I see that all these patterns could evolve without any changes in coding sequences whatsoever now this is well it might seem like a naked assertion but to me it's not a naked assertion because we're dead serious about working hard to dissect how more complex patterns have evolved and I want to give you one parting glimpse of what I'm talking about I think from the moment my favorite fruit fly drosophila good Affair the polka dotted Wonder that's that's not what it says in the drosophila books but to me the polka dotted Wonder this is kind of as close as drosophila gets to a butterfly I inhaled when I was in college okay um fortunately our team doesn't have those photos but anyway so here's the question I just for just for fun minute or two how would you make a pattern like this with this many spots how would you make a more complex pattern okay so I told you a story where making of a of a you know of a of a spot on a wing a new spot required the evolution of one CIS regulatory element so there's something like 16 spots on this Wing how would you make this pattern how is this pattern encoded how would it evolve so we'll just play audience participation pattern how do you make audience participation how do you make a more complex pattern you want to take any guess how many CIS regulatory elements you need to draw this pattern anybody come on shout it out two two four okay pretty conservative those of you who said two I hate you you're right okay we had to look through 37 kilobases of DNA you only came up with it in like two seconds four is an okay guess two 16 would have been high but we had no idea because we didn't know how this pattern was pieced together sort of cobbled together over the course of evolution and I could have imagined that it might have been more regulatory elements but uh and just so you have some sympathy for us to do this experiment it meant developing transgenic tools for drosophila good affera and dealing with all of its little peculiarities of mating and husbandry which took us over two years um so our team said something about the nice atmosphere in the lab but I am a slave master um because I wanted to see this pattern so uh one regulatory element actually gives you all but one spot and then this smudge that's between these veins comes from a different regulatory element actually so does all these very light striping that I didn't point out to you before uh you can start thinking about how you make this complex pattern we actually have some good hints but we're one experiment short of telling you the definitive proof so I'm not going to really tell you what I want you to do uh since you have figured out this is of course this is at least a fairly visual you know visibly complex pattern coming from this is a 277 base pair element so I'd say take our little polka dot enhancer as further evidence that CIS regulatory elements they're the complex they're the simple Invisibles sorry I'm going to get this right they're the simple Invisibles that explain a lot of the complex visible diversity of animal form thank you [Applause] yeah fire away um one or two bases away from changing to humans for example it seems like we're only one or two bases away from disaster like certain types of diseases but it seems like most of the diseases we've characterize so far in human genics are in coding regions so it seems that we're going to have to get a lot better bioinformatically at identifying it's just regulatory elements and finding out which ones are significant right we're still not afraid of that you could you could argue that a lot of the coding stories um are you know the low-hanging fruit if you're studying you know proteins and you can see you know stop codons or other things that that alter protein function that's obviously been the easiest stuff for us to see in genomes or our coding changes um I think of course I'd argue a lot of the physical variation of humans you see in this room are going to be non-coding um you know slight differences in cheekbones and forehead size and yeah you know forehead okay never mind we'll get to that trait but um yeah and and I think that you know for people interested in you know human variation and human disease there's this whole big component you know non-coding uh compartment of the genome that's been uh much more difficult to study um has to be done really empirically but we know there's a lot of stuff out there just from look if you look compare Mouse and human genomes you know twice as much non-coding DNA as under purifying selection as coding DNA so we know there's lots of stuff out there it's just been you know harder to study and it's of course harder to study in an in Vivo context because you know you can't take human regulatory elements and put them back into a human embryo and you know find these regulatory elements that are active in particular tissues so a lot of this is going to be done in cell culture but cell culture doesn't necessarily reproduce that spatial and temporal information that you get in Vivo so it's it's uh you know it's it's going to be a serious challenge I mean there's a fair number of cases of non-coding things that you know affect human phenotypes but um you know they're they're definitely smaller in Number the ones that have been identified than than those that are coding stories yeah in the back yeah yeah right um so it seems like the message you're giving us it's all about five but at the level of this radio is very elegant Elementary and so we over and over again it was that cre and we find the right City agreement if you climb the hints that is seen at times and and so it seems like the this regulatory element level it sort of takes us away from applied for the argument almost and sort of pushes and go back a level but is that is that what we won't take from this um I mean I've there's there's some you know these case studies are selected because you know the differences are so uh it's certainly between species stories are pretty discreet between the tax of being compared I think I understand that that what we say is a phenotype you know is a sort of a range of phenotypes in a fly so I don't know to a fly you know what those intermediate levels of pigmentation mean you know to uh to a to a fly in Uganda so you have this whole range of expression phenotypes that are that that you map to the CIS regulatory element um there's also I didn't emphasize a little bit as you as we're mapping those little mutations that element they have different effects on lateral patterning for medial patterning so while I will describe these things as you know abdominal patterns and that to me is the trait um the trait is changing in uh subtle ways and in different directions along the you know along the segment so while it at one level uh that is the story I'm giving you uh the little color I'm trying to add here is to say you know it's a little more nuanced than that that um I you know we don't know what aspect of these traits the animals care about right you know what what degree of pigmentation what spatial area that's covered things like that so it may well be that one cre is sort of multiple traits it's just not obvious to you and I what those multiple traits are so I think cautiously should you can you can think in that vein but with a little bit of of uh caution what's your impression of the unitarity of the modularity of the controlled by the area in other words how many if one looks quite broadly at gene expression how many cell types is there a single really a single module that regulates the expression as opposed to multiple modules cooperating or oh the module is actually working independently additively or whatever in other words you it sees out and these all work but I won't imagine that there are plenty of cell types in which two mod two cres are actually yes cell cell types are body regions so um yeah let me let me expand on that so um certainly plenty of cases where if you just talk about a tissue like a wing blade that we'll find multiple cres that contribute to the total pattern okay so from our point of view you might say well that's a wing blade but we'll find separate distinct pieces of DNA that will contribute to that um well you mentioned cell types and the reason why I'm just saying about uh oh that there would be more than one regulatory element separate regulatory contributing to level of expression a single cell yeah probably about a single cell in an organism you just need to think about one crazy well no I would say that there are certainly cases where more than one creative Locus contributes to the level of gene expression in a particular cell it's there's not any rule I can give you about that that's you know certainly things that we've observed so we we've observed in in a given tissue where the the uh you know a total pattern is clearly just the perfect complementary activity of multiple cres where there's no overlap in activity and there's other cases where we have multiple elements contributing to the expression in a given cell so in those cases where gene expression is evolving and you have more than one cre contributing then there's no doubt you would suspect that you could get contribution from multiple cres to The evolutionary shift either in the quantity or pattern of gene expression um so I still think you should be thinking about CIS regulatory sequences but it's not necessarily a one now I understand a little bit with your driving tour in for a given cell for a given region of the body sometimes you have overlapping inputs of multiple CIS regulatory elements for the same Locus and that probably makes sense how these regulatory elements have expanded they probably themselves have expanded by duplication and Divergence and so you have redundant regular if you want partially redundant regulatory elements within particular loci is that clear okay yeah the way the name cre why not are there class elements involved in this too it's a it's a cinema it depends how you grew up I refer to him as CIS regulatory elements these are enhancers I I sometimes you know people think of enhancer sometimes as promoter proximal elements as well and for these developmental regulatory genes these are usually not anywhere near promoters they're dispersed sometimes you know 10 20 30 kilobases away and um I just prefer myself describing them as this regulatory elements that's why I just Define them functionally I probably should have said uh others would just refer them as enhancers yeah but there's yeah anyway I'm really sorry if that misled you thank you okay thanks