 So my name is Stefano Vitalia, and I am at Duke University. So if you don't know where that is, it's North Carolina in the United States, in the south of the country. I was actually trained as a physicist, and I switched into biology about the end of my undergraduate which was in Italy in physics. Then I actually spent a year a year, and I've been finally back after about 14 years, and it is really a great place to work. Not only the science is great, you can just go to the beach for lunch breaks, and it's a wonderful park and a castle around here. So it's really a great place where you get the best of everything. You have great science, and you have great environment, and nature, and history around you. But then after a year, a year I moved to Rockefeller, where I actually became an experimental biologist, and that's what I've been doing ever since. At first I started with yeast and doing cell cycle control, but then as a postdoc, a Princeton has switched to embryology, and this is what my lab thought you saw now, and this is mainly what I wanna talk about. And most of what you learn in this school is really gonna be about morphogenesis, and now embryo develops, or even adults, and the whole goal is like, can we apply more physics and quantitative way of thinking to our forms, and morphogenesis begins in biology, and it was really through the vision of Antonio of realizing that this is really a topic that where physics can make a contribution, and that this school came about, and that was asked together with Calphely-Paisen, but that will give the next lecture to be one of the director of these schools, and the idea is that to try to explore actually more quantitative and physics way of thinking can contribute to our understanding of development, and so what should we care about embryo? Well, the first obvious reason, that's where we all, all animals, we all come from embryo, so we all start from release. This unicellular organism, that this is like actually a human development, you start from a little fertilized egg, and then it goes on to a series of transformation to actually make a baby, or my favorite model organism, which is the fruit fly or drosophila, you also start as a single cell, and eventually to a series of transformation will make an adult, so that's of course a really important biological question. There's also another reason that I think is very fascinating to study embryos, and is that embryo just great system to study biology in general, and I think this is reflected very well in this sentence that Victor Hamburger was one of the father of modern embryology, and he was actually the one who recruited Rita Levi Montalcini to work at Wash U for the Italian and the people who have connections with there who know about that, and so what he said was, our real teacher has been, and still is the embryo is incidentally the only teacher who's always right, and is especially this second part of the sentence that I found very interesting in the sense that embryo really provide a great system for us to study biologists in its physiological context. You can really get an embryo and look at like basic processes of biology, like how generous pressed out the cytoskeleton is regulated, and you can really study that is real, that is really what happens. It's not like you are taking a cell and putting into a dish, you sort of remove yourself from many of the artifacts that are often you're forced to do when studying biology, you can really study biology as it happens in the embryo. And the other great thing about the embryo is the embryo, we can get embryos from many different organism and they show both some phenomenon that are the same. So you could study basic cellular processes and you can really start seeing principle that emerge because there will be some aspect of how an embryo develop from a fly, a frog or a mouse that are actually the same. But there are also many other things that are specialized and their difference. And those are equally interesting because it could tell us either some insight on evolution or it could tell us some insight on like, maybe the fly embryo is developing a lot faster than it is developing a lot faster than a mouse embryo, are there specialized mechanism or control that are for example required to do things rapidly doing biology versus doing things at a slower pace? And actually I will touch slightly on that so now you control an embryo. Embryogenesis very fast. And a lot of what I thought you saw in this lecture will have to do with cell division and the importance of cell division and how you start from a single cell organism but you need to end up with an organism that's a lot of cells. So cell division contributes significantly to embryogenesis. And so before getting into the embryology let me just remind you some high school textbook the fact about cell biology which is how cells can divide. So most of the time when we think about cell division we think about one cells then make two daughter cells that are genetically identical to the parent cells. So these cells will have a set of chromosomes. Usually most organisms we are interested in are deployed which means you have two copy of each of the chromosome and then this chromosome get replicated and then they're split equally between the two daughter cell. So that if you think about genetic material or DNA content this cell has the same one that the starting mother cell and so does this. There's a different way to split the chromosome which is very important when the life of embryo begins a fertilization which is meiosis. So in this way what you do you equally replicate the chromosome but then instead of splitting them into two cells you actually split it into four. So now of course because the DNA was fixed you have four copies. Everybody's getting one copy which what that means is that through this process you get an applied cell and generate which replicates DNA and you generate four applied cells. And this is of course very important because to make a deployed organism you have to apply the egg and the sperm that fuse together a fertilization to start life. So that's where all embryogenesis begins and what it begins with is the when a sperm and egg meet. So this is actually a cartoon taking from this really nice book. You'll see a lot of them on principle of cell cycle control by David Morgan. And so the actually there's the process of just making a egg as well as making a sperm which is not depicted yet are very complex and they've revealed a lot of interesting biology. But for us what we find is the fact that you generate what is called an oocyte. So before maturation the egg is actually called an oocyte it will just grow and then it will undergo the first phase of meiosis. Now as I told you before at meiosis you take one cell and you make four applied cells but of this four you only wanna use one, right? You know because you wanna use one of those cells infuse it with the other cell or the other applied cell that's coming from the sperm. So to do that what the egg needs to do is to push out essentially three of those cells. So at the first meiosis when you do the first meta the first cell division and meiosis you push out two of this chromosome site and you are only end up with two that are now stuck in what is called meiosis two and when the sperm actually fertilize the egg then what happens is that the second meiosis is completed another set of chromosome is extruded what are called the polar body and then this applied and now applied embryo is left to meet with the applied sperm nuclei and this nuclei refuse to form a single nucleus and that's where you start embryogenesis. So a great system to study this is the C elegance embryo I'm not gonna talk much about it there is a great poster from Alessandro who is the student in Los Angeles actually coming to the postdoc with me and what you'll see in this movie let me start it again is that you'll see exactly the process of nuclear fusion so you have the female pro-nucleus and the male pro-nucleus that come together they fuse and after they fuse they become a deployed nuclei and they start dividing and the early process of embryogenesis begins. There is a big, big challenge that embryo phase as soon as they start at this process and the question is a lot of embryos for most organisms like that we are interested in like flies or frogs or like marine species like sea urchin they've been classical, but the organism it is actually organism laid their eggs outside so a fruit fly will like get into a rod and fruit and will just drop a egg there and the egg will need to develop but because it's outside it's essentially unable to feed itself until it has developed enough to have a mouth and you don't want an embryo to be able to exchange things in and out so really all these very small molecules like oxygen can diffuse in and out what that means is that essentially embryo are unable to grow or to acquire nutrients until they hatch until they develop enough to be able to hatch and go out and feed themselves so essentially what that implies is that the embryo has to have enough nutrient inside to be able to go through all these early phases of embryo genesis so the way that this problem is solved is that the most metatone like lay embryos or eggs that are really big so let me just give you a sense of scale this is the usual somatic cell in your body and this is how big an embryo is so they are about probably a million times bigger so the radius is about 100 times bigger so this is great now there is enough material but this is supposed to be a real challenge and the challenge is this is a single cell when it starts so it's only a set of two chromosomes so now how can these two chromosomes keep up with this huge egg and so and you of course feel free to interrupt me if there's anything I say is not clear at any point then I can clarify so let me just now let's just now go through some simple number and see why this is a problem so from a lot of estimates of people that measure molecular biology processes we know that the typical rate of transcription so the dogma of molecular biology right obviously goes that you make a RNA from DNA and you make protein from RNA so the process of making RNA from DNA is called transcription the process of making protein from RNA is called translation so you could look at these basic cellular processes and get a sense how quickly can they go so the rate at which you make an mRNA from DNA the typical transcription rate is about one kilo basis a minute so you essentially transcribe about 15 bases of DNA in a second that's how fast usually they can go there is also as a polymerase is moving on the RNA you can maybe at the DNA you can load another one but it's not that you can really load an infinite one because everybody's taking some space so typically you can probably fit within one KB above six polymerase so what that will mean is that sorry if you had a single genetic locus you are interested in you can probably make about six KB of mRNA a minute out of that locus you can put in some some numbers that you can kind of get to the literature similarly for protein usually let's say it is a rough estimate you can probably make a best about a thousand amino acid a minute of a protein and maybe you can load in such a protein about 10 ribosome per mRNA the ribosome is the machinery that will take the RNA and translate it into proteins so again you end up with a rough estimate maybe you can make 10,000 amino acid a minute and now you assume that you are doing this in a way you forget about the gradation all of this is stable so you make RNA it's growing up linearly as the RNA grows up linearly you make protein so then protein will go up quadratically this is just a a simple model and and so you can you can imagine a typical gene which is about three KB and let's say encodes a protein of 500 amino acids and you assume that this goes up quadratically because the marinade is going up linearly and the protein which is the integral of the translation of the marinade will go up quadratically and you end up with some number about 20 protein a minute times the time squared expressed in minute times the number of genetic loci so now if you only add one loci will be very hard to transcribe it but you could maybe have more because if cells had undergone cell division so now let's just plug some reasonable number let's just think that you are a fly embryo and you wanna express your an enzyme that is driving some cellular processes is about 15 nanomolar that's a typically good concentration for enzymes in cells is actually maybe even a little low so if you do a little bit of algebra you can convince yourself the 15 nanomolar is about 10 molecules per micron cube but a fly embryo is huge is about 10 millions micron cubes so this means essentially if you wanted to make this amount of molecule per micron cube in a fly embryo within 15 minutes you will need to make about 100 million molecules only in 15 minutes and if you plug it in the number of loci that you plugged in the expression before you realize that you need about 20,000 loci if you wanted to get this from a single loci you will need to take the square root of this so it wouldn't take 15 minutes it would take of order of a day or more than a day to make that protein however a fly embryo really develops within hours so it's clearly not gonna work from a single cell you need multiple cells yeah the question I think it was just some rough estimate I found it may be a little higher but you still will end up that you need a border of 1000 right so the fly embryo is essentially not growing and throughout this process you are really everything you can assume that all the basic machinery is constant there is no growth is probably not in this balanced state this will be how many copy of DNA that are spitting out this protein you need it's essentially if you only had one right if you wanted to make this enzyme and you only have it in one for most of genes you are interested in they are present in one copy right there will be in one chromosome in one copy so if you wanted to just make it out of that one copy so much protein it will just take too long essentially but a different way to get so many loci is that you undergo DNA a lot of cell division that way you generate a lot more cells or a lot more nuclei and that way you can keep up and do development and that's exactly what I'll tell you that's the way that the embryo solved this problem and but before I get there so so if you think about what that means for a fly embryo which has about four loci per nucleus because they are diploid so they have two but they are also usually in a phase of the cell cycle where the chromosome have already been replicated so you actually have four then you'll end up with about five thousand nuclei which is which I'll show you is exactly when actually they start making transcription because what embryo do is essentially the early phase of development they don't use transcription at all so the early phase of embryonic development there is a very little zygotic expression with zygotic so when you look at the marinade in a early embryo you can distinguish two sources zygotic the marinade are the marinade which are actually produced from the DNA inside the embryo but there is also maternal marinade once the mother lays the eggs he can put the marinade and protein in there and so those the marinade are called maternal source because we are provided in the mother so what happens during early development is essentially an up to a transition which is called the maternal to zygotic transition which is a genomic blastula transition there is few hours of development where there is no transcription it's all controlled the mother lays a lot of the marinade and protein in there these are everything that is needed for the early cellular processes of development and that's what is driving development and the major thing that this process drives is just cell division so you wanna make a lot of DNA so that you have enough that transcriptionally you can actually start development you can keep up with the demands of development so this is for example a movie from a Xenopus from a frog eggs and what you'll see is that in the early phase it's just dividing like boom boom it undergoes this very rapid cleavage division about one every about twenty five minutes and it just keeps on going if there is no zygotic transcription is just cell division all driven because the mother has put in everything that is needed for cell division in the egg and that's driving the process yeah you had a question why you put both mRNA and proteins because some of the product for example for and I'll show that to you for a lot of cell cycle process are actually controlled by product in the gradation so there are certain product and you need to make destroy them make them again destroy that make them again because what you really you need an oscillator to be able to go to cell division we'll get there in a second so so you want to put both product but also the mRNA so that and also most product in the weather do I have a short lifetime or do I have a fixed lifetime so if you if you stop completely translation eventually most product will disappear because probably not live forever so but if you resupply them then you reach some steady state right by the balance of production and degradation so you always need the source to make new proteins that's another question so actually that is true but the it turns out that the mRNA lifetime is very regulated so a lot of mRNA that would be very unstable later on in development they're very stable and that is the during the early embryogenesis up to this maternal to zygotic transition and actually if you see these uh... what I was showing you here there is a large class of maternal mRNA you as really really really stable and then you get to this transition and boom they're completely degraded so it is true that if you look in east and you look at any generic mRNA as a lifetime of two minutes probably but that's all regulated so you could change your knobs into the biology so that those mRNA are very stable and that's what the embryo does so a lot of mRNA are really really stable during embryogenesis and then you eat this maternal zygotic transition and you can express enzymes essentially that target them from degradation so it's a good point but it is all regulated and the embryos develop way around it okay so the early phase of development really characterized by this very rapid dispension and instead of talking about fraud eggs most of what my lab works on and most of what I tell you today is actually the development of the fruit fly or drosophila embryo this is a really specialized embryo in the sense that actually it's a single cell and goes through most of his early development as a single cell so it's a syncytium so it's one big cell and that's what happens is that you don't get cell division you really get nuclear division so a nuclear split they just spread around this huge cell but there's really no cell membrane dividing them they divide inside and then there goes series of division that are inside and then they all migrate to the surface to form what's called a blastoderm few will remain out the inside but most of the nuclei will come outside and they will keep dividing and then after thirteen division and so it's thirteen division and we'll talk a little bit about how they count to thirteen they like actually stop and they undergo this maternal to zygotic transition the star gene expression and also what the as the star gene expression what they do they synthesize all the machinery that is needed for this special special way of doing cytokinesis in which you grow membranes and you essentially close all the nuclear cells and that's when you actually get cells and then the embryo goes on with morphogenesis so so you can look at this process so this is a single cell with about at this stage is about you know you can guess about six thousand nuclei if you get left out it starts with one and then the nuclei keep splitting so all these black one are it's it's uh... depicting the nucleus so you start with one obviously a fertilization but then they divide but there is no cell division so essentially you end up with one big cell which has multiple nuclei inside eventually that was they get grown into membrane each one of them become a different cell now apoptosis all it begins way later few hours so at this point it's all cell division there is no cell there right so this is all driven as you get zygotic gene expression here they gen they they they they express essentially some enzyme that start this process of membrane ingression and they regulate you know myosin contraction which pulls this membrane out so it's all regulated by so it's part of this maternal to zygotic transition as you activate zygotic gene expression you make some enzymes that are required for this process essentially right so there is few so you essentially what happens and this is true throughout a lot of early development you you load a lot of the basic machinery but you only miss a couple of key enzymes and those are then providing very special and special cues so it is also true for differentiation or a lot of morphogenesis you do have all the basic machinery but then the temporary spatial instruction all requires one factor or few factor that are missing and that are expressed very specifically both in space and time and that's how essentially the whole process of morphogenesis is controlled so accurately is that the machinery is there but you only miss one or fewer limiting component and you express specifically at the time and the place you want to drive those processes say that again so so this is true too it's 13 cell division which means you have about two to the 13 nuclei so it's at 8000 nuclei yes I mean like every transcriptional process if you really look precisely there is noise and there are way to filter out that noise I think probably Thomas Gregor who will talk tomorrow or the day after or once they will talk about it so so we can leave that I'm sure you'll learn more about it later from why do they migrate to the surface oh these are so-called pole cells so those are eventually the cells that will make eggs or sperm so these are very specialized cells are germ cells that are specified to become either sperm or eggs and they actually get sorted out about these eight or nine division they're pushed I mean they're not really out they just you know have a little they actually already have membranes these are actually cells already at this point is a specialized mechanism to generate a special pool of cells that will have a special role in the life of an animal which is generating germ cells so there is special cues right I mean these may look like there is no information but there's already some factor that are segregated only in the anterior or the posterior so this is really not a uniform embryo there's already a lot of special cues at this stage of embryogenesis and it's these special cues and some of these factor that control the generation of the pole cells there is a lot of them right and that you know it's a big it's a really big embryo right so so this right essentially yeah right it's essentially because until the latch which is twenty four hours later when there is a lot more cell than these that they essentially are unable to get any nutrients in so I mean they could make new ribosome these ribosome binogenesis to replant but there isn't any net growth so essentially from an energetic point of view is an isolated system so he has to just make what they have also they have a big storage of what is called yolk is very proteinous and essentially burn a lot of that energy to translate it into more protein so that's it they contain some really really dense stuff that they can burn out to generate more proteins but energetically it all has to come from the egg Antonio essentially I think that's probably correct I don't think so I don't think it's uh... I think it happens well so in the mouse and it happens day eleven right or day twelve that you start specifying the gonads I think so well I think I I'm not sure I mean one of the thing you you have to to realize for example that there is a process of cell specification they start very very early in embryogenesis so it may be in part be linked to that I will assume so for example X chromosome compensation can be visual as soon as they start transcribing their genes a cycle fourteen you can already see that the male nuclei is transcribing as much as a female nuclei so there's already gone on all this process of dosage compensation so if you have two X chromosomes or one X chromosome you express most of the gene on the X chromosome at the same level and this is already true like castrolation so so it is probably maybe linked to that I'm not sure evolutionary why you want to specify them so early but it's possible that has something to do with the fact that actually a lot of process of sex specification start already in embryo must start in embryo but then there are other organism that can keep by doing that a little later so I don't know I wouldn't make from these a general principle that is the first thing that needs to be specified so this is a movie from out of the dorsophila egg develop and what you're seeing at Eastern type with GFP Eastern is the material that keeps the DNA together wrapped around the nuclei so at the beginning all the nuclear were inside now they come to the surface and then they'll go this really beautiful synchronized mitosis and they will do one and they will do one more at which point now they decide that they made enough nuclear and that they can actually keep up with development and that's when this process of cellularization now is happening and separated all the nuclei in different cells and then as the cellularization is completed they start actually morphogenesis and now you really start getting like self specification different group of cells that divide a different time and then just start shaping itself so for the topic of this course we're mainly going to talk about how the cell proliferation is controlled in space and time so let me just remind you about cell cycle and cell cycle control so the goal of the cell cycle and regulation of cell division is to start in one cell and make two cells that are the same at least genetic intermogenetic content from the first cell for that you need to do two things you need to replicate your DNA and you need to split it equally between two daughter cells so the see the phase in which you synthesize DNA it's called usually S phase for synthesis phase and then mitosis is the process through which you split them equally between two mother cells and somatic cells are a lot of what you have heard about the cell cycle is that they are between this S phase and M phase there are gap phases so there are some phases during which the cell checks am I big enough have I grown enough that I should be able to divide actually the embryonic cell cycle they get rid of that they don't need it they know that they want to divide as quickly as possible so they just replicated in a split it replicated in a split it so that they can go much faster so they're very stripped down cell cycle and therefore they've been a great system to actually study cell cycle control and from the work of many many very good people what has emerged is that essentially the way you control cell cycle is that you have an oscillator it is really are somewhat of if you want a repetitive process you deplicate your DNA you split it now the cell every single cell has to redo the same thing to make two more cells yes to replicate yes they may split it so what you have you have some enzymatic activity they go up as they go up they trigger S phase then they when get higher they trigger mitosis then you destroy the activity and so you have reset and the cells now a split so now you have two of these cells and if you look really look inside one of the single cell you have the activity go up and down and that's how you repeat the process and the the way you get this oscillator is because you have some proteins that go up and down and most of what the if you read that cycle textbook what you learn it was these are actually from the original paper from T-month and is the work from which you won the Nobel Prize in 2001 was to realize that it was this protein called cycling there's actually a class of cyclins and what this product indeed i think it would be this product here he was at and he was looking during the early stage of the valve of embryonic development what this product does is made and so it's level go up and then as the cell is about to split and you can see here this is my talk some kind of my product index the product boom disappears and then you make it again and disappear so the eternal that is products called cycling are really an essential component of the cell cycle so what they do is that they bind they bind something called city k1 and once these two things are together they form an activity and these activity can drive the cell cycle in some enzyme it will go out is a kindness will go out will phosphorylate substrate as the substrate of phosphorylated mitosis a mitotic event begin but then you'll be stuck in mitosis if you keep this activity high so the activity needs to be reset for the cells to exit mitosis and restart the process for that usually what happens is that this process is driven by product in the gradation so there are machine so not only product in synthesis is regulated product in the gradation is also regulated I said the cell sometime was just to get rid of certain products to do that they use is a process called ubiquitination so you can put these post-translational modification on a protein and it gets decolonized by the proteasome and it gets rid of the product you just choose it up into pieces the product is cut no more function and there is a special complex ubiquitination complex called the anaphyse promoting complex and what this complex will do is recognize cycling being just degraded so so now how do you make an oscillator out of this thing and the first simple view of how you can make an oscillator is that you could have a system of negative feedback which will actually not work so you need an extra ingredient and this is this idea here is that you need also a delay so what is happening is if CDK1 will activate the APC the machine that destroys it and then this machine will destroy it and especially if it does so with that delay what will happen is the following cycling activity will go up and our cycling level will go up and eventually they will go up enough that they can start activating the APC but there will be a delay so in a way you have overshoot and as the APC activity goes up it will destroy the cyclins but since you need CDK1 activity to activate the APC as CDK1 activity disappears soon with a slight delay also the APC activity will go down and now because there is no APC activity the cycling can accumulate again and you can repeat so actually you can show mathematically it's very very difficult if you just have a simple feedback to mechanism to get actually an oscillator will happen they will just go to homeostasis but if you have the delay actually this is a system to build an oscillator right it's all product right so this is it's because the mother lays a lot of mRNA so there is a lot more right so the calculation I did before assume that you started from zero RNA and the RNA was was accumulated yeah that's sufficient if you have a lot of mRNA you can make as much product as you want I mean it doesn't matter how many nucleus you have because the mRNA is all loaded in the cytoplasm then the nucleus is not contributed to translation at all until you get to them to the MBT right there is no transcription so there's a lot of RNA you transcribe product from this RNA actually you can even get rid of nuclei a lot of what we know about this is made from cycling from Xenopus egg extract and sometimes you can just get this extract and just get the cytoplasmic phase with no nuclei in it and is able to oscillate so you don't even need the nuclei to actually build the oscillator the cytoplasm is sufficient yeah yeah cells will become smaller and progressively smaller and smaller but you start with really huge cells now I'll show you why I mean you have a good point why is this going up linearly and this not the activity is not yeah I'll show you the reason is that this is not on top of these really really what the embryo wants to do or most cells want to do is they don't really like that they don't really like to control processes through a linear accumulation what you really usually want you want an activity that spikes up very abruptly so you because the view here is that essentially what you want to do is when you enter mitosis you just want to enter and so you want to enter very very quickly so the so essentially what the machinery is built in you also have built in all this positive feedback so as you make cycling city k1 actually the city k1 sub sub straight is modified by some other product to be inactive but eventually you just get enough and this positive feedback can trigger activation and that's why you can translate the sun is essentially linear to something that is very stepwise these old efforts because through positive feedback you can essentially amplify signal and what it was shown by and I'm just gonna put you some if we just zoom in on this particular feedback and we look at what what this nature is so what happens actually I told you about cycling and in city k but what happens that on top of being controlled by cycling synthesis there is a kindness we want that can put an inhibitory phosphate on city k1 and a phosphatase that will remove it and they also operate in a feedback so that cities twenty-five activities positively controlled by city k1 activity and we want activity it's inhibited so this could generate positive feedback as soon as you get a bit of city k1 activity you will have amplified so on activator and will repress is on inhibitor and that way you can transform a little bit of activity into a lot very quickly and that the other thing that you get from something like this is by stability so this is a plot in which the system was tweaked so that you can ignore about the product degradation is by using a cycling that cannot be degraded so you are really only looking at this network the way it works and what they saw the the other deal where it was the supply cyclins and then what you see is that the city k1 activity is low is low is low and then as soon as you pass some threshold boom spikes up very very quickly then what they did in the next experiment they took the cycling off so you kind of titrated out and you kind of go back to low activity but you go back on a different if you see you go up in one part and you come down in a different part and this is the signature of having a best able system which is the resist is the there is right you get the mainly the main idea is that in this system you get a coefficient because you have multi-site phosphorylation so city k1 phosphorylates city c25 but don't like probably ten eleven sites if phosphorylates we won in three or four sites so so so I just believe that this multi-site phosphorylation is another mechanism to give some days effectively identical to what people would predict for quadratic by name when people and measured is all ill-coefficient for this reaction they all come up to be higher than one so there is enough nonlinearity to generate this sort of thing so what happens when you have a best able system is that there is a regional parameter in which the system could be in two states you could either be low high or low and if it is high or low in this region really depends on the history right so if you come down from very low activity and you plot a response versus input you're just gonna stay on this line and then eventually you you run and you jump here so this is like gives us a that this feature that we want and we talked about before about very abrupt response so you're going up and then all of a sudden you just jump so you are you get this very abrupt activation which is what drives mitosis but also it gives you this nice feature that once you are up here if you get a little fluctuation that brings you to the left is not gonna be enough to bring you down you have to go all the way here to come back down which means essentially this might give you some buffering of noise once you jump around here really you need a lot of noise to be brought back so that people really believe that by stability is a great way to regulate cellular transition that you really want to be irreversible and very rapid and mitosis is one of that once you enter mitosis you don't want to go back because it's so much happening you're like destroying the nuclear membrane you organize all the site of skeletal it's really a lot so someone to be sure they go in at the right time but once they go in they want to be sure to go all the way so by stability is believed to be the way you do this so this is just a primer on the way that most people think about cell cycle so that is positive feedback to sharpen the transition and is a medical feedback we delay to control that you really get an oscillator now let's see how this applies in vivo but before this a couple of questions here's one oh if you don't have a by-stable system what other mechanism could work to buffer noise I mean I guess it's a so if you just had a really really sharp transition that wasn't by stable right you will go here and you will go really really up as you get up here you will be very very sensitive to noise I mean probably you could make sure that you drive it by having an input that changes very rapidly so that it's very unlikely that you will go back so there may be other mechanism like this but this is just a an important feature I mean often you can remove and I will show you an example maybe I mean often you can remove some of the sources of this by stability and the cell still can cope with it so I'm not sure if it's really truly required but it's a feature that is often associated to sharpen those transition well in this particular case you can just think it's cycling synthesis right so you start with very little cycling you make some you make some you make some but as you make it is still inhibited and then eventually you have made enough that you can get it activated and then of course it's destroyed as by the degradation so essentially what you what you do in the cell cycle is that you do you kind of do this you go up here then you move maybe here and then so this is essentially you're almost having an oscillator that is doing a limit cycle around that I'll talk about in my next lecture all this is synchronized and there's actually a collective decision so the decision of entering metosis in the fly embryo is a collective decision that is influenced through these actually waves of activity that spread through but you'll have to wait until friday and then there's one in the back and then I'll come to you so these these mechanisms alone without the feedback school give you an oscillator and there are actually stages in early embryo genesis of the fly embryo of the Xenopus embryo where probably this is what is going on because these other feedback kind of become irrelevant actually so and I believe that the simplest model for the circadian oscillator is just also a negative feedback with delay so it's an interesting point of why biology always seems to add these extra layers and the easiest explanation is like he adds robustness I mean it's like I don't know it's a bit biology is often a bit more complicated that it needs to be and probably the function is that you want those transition to be sharper and it's easier to do it but my evolution has been easier to do it by adding an extra layer of regulation rather than really fine tweaking this parameter in a way they work robustly but I think it's a good point why aren't the genetic circuits simpler but it could work you could get oscillator just out of these and I think if there would be specialized cell cycle during a mirroring development that I believe will work just through this essentially through all of these negative feedback with delay and two components and there was one here are you also sorry I'll come back to it while also the product can change because their degradation production but their degradation is controlled over time you only the great product as you are in mitosis so that then you go to zero then you make them again and then you destroy them again so you even if you don't have transcription you can have changing product concentration by having regulated translation or regulated degradation in this case the gradation seems to be the major player but there are another opportunity and I'll show you and I'll argue that this is how mainly the early cell cycle are controlled is that you can just control at the level of activity all post-translationally or by having different enzyme that act at different levels these are all different ways but that you can control there are intermediate steps so CDK one often controls another enzyme called Polo that then control so there is belief to be intermediate step and also just phosphorylation might take some time so there are built-in delays which are associated to either having a cascade or to having other events which are not depicted in this cartoon so I'm trying to just make it the simplest conceptual model yeah I mean that there is a paper from Jim Ferrell when I looked into this and I think they don't remember it's actually if they had any experiment in changing the delay it is also true that the mechanism by which the APC is actually controlled by CDK one is just starting being elucidated now there was just some paper out this year that start really understanding this so I think we could probably now as we get a better understanding of this interaction to start to generate that make genetic manipulation that can alter the delay okay so I just give you the textbook view of the cell cycle let's just see how this kind of concept apply to the early fly embryo and which one actually all proven which one actually turned out not to be relevant and you have all seen these already so this is the way that people looked at embryo back in the days and I'll show you how we want to do it these days and why I think that quantitative biology is helpful so this is really an heroic effort from one of my scientific hero Bruce Edgar when he was a postdoc what Bruce did here is that he measured the concentration of a lot of proteins and these are two cyclins these are the CDC-25, the activator and this is the CDK one component and what he did is he did this by Western blood. The Western blood is something which you crushers, a technique which you crush your cell or your embryo you get the lysate, you run it through a gel this is the gel so that you separate protein by their size and then you have an antibody that recognize specifically your protein and then you can recognize the antibody with another antibody that carries an enzyme and then you give it a substrate which generates luminescence and you put it under a machine or on a film that will read this luminescence so there's a lot of manipulation and what Bruce did was really the heroic effort so every point you see here is a different embryo so he got an embryo which was a cell cycle 2 in interface and then embryo cell cycle 2 in mitosis, this means anaphase so mitosis he means metaphase when the chromosomes are compacted and then anaphase is as soon as they start splitting and telophase and when they move apart a little more so he collects, I don't know, probably about 50 embryos or more at all these different stages and then he run this gel and blotted for all these proteins and the reason why I want to show you this is that what is really remarkable is that if you look at these early cycles you really don't see the protein disappearing so the protein is really not appearing, it's disappearing and yet I've just told you that this disappearance of the protein is what you need to reset and start mitosis so what is going on here, they synthesize their model into something like this and it's actually if you plot the level of cyclins as a function of time or as a function of cell cycle they came up with a model like this in which the level are really really high at the beginning, maybe declining a little bit but really you don't see oscillation and only later on the oscillation came up now we can just repeat this very easily and we can generate these data are much more resolved from these, they are quantitative and they all come from a single embryo so that's one of the great things that has happened since the 90s and it's been the discovery of green fluorescent proteins and a lot of fluorescent proteins what these are are proteins that if you illuminate them with the right wavelength they'll shine light back at you so what you can do, you can fuse your protein of interest with GFP and then under your microscope you illuminate this protein with blue light and it will shine light back in the green and so you can essentially visualize your protein and by doing that then you can develop computational algorithm and see how much of this protein is present as a function of time and this is a live embryo so you can just look at it develop so you can get essentially the same plot from a single embryo and it's truly remarkable and really speaks on how good these people were that essentially got the right answer from a much cruder technique but it remains true if you look at the very early cell cycle are really driven in the absence of degradation so yeah I'll talk about that, it is true, yeah right, so we are not corrective for this so this might not reflect perfectly the amount of protein but it's close enough and these two give us the I'll tell you, it's a bit slightly more complicated than that but that's a very good intuition it's kind of the right idea and I'll show you actually it's one of the things I want to talk about in most of the rest of the lecture and so the early cell cycle seemed to happen in the absence of significantly degradation however if you do an experiment in which you prevent the degradation from happening the cell cycle stops so this seems a little counter-intuitive and this part of Arel and Tenten-Sue came up with the idea probably what is happening is that activity is only oscillating around the nuclei and because when you do a Western or because when we are always looking on the surface and the nuclei are inside it's very difficult to develop optical methods now we like shit maybe one cool but if you look with confocal microscopy you can go deep inside the embryo look where the nuclei are so we look at the cortex where the cytoplasm is and we can see it oscillates there that doesn't mean that it may not be oscillating just around the nuclei inside and that will show why this requires that you have the degradation but yet you really don't see a global oscillation however that also poses a and we actually could measure that we have done some measurement and about how the APC activity spreads as a function of distance and probably for sake of time and we can just go through this quickly you can believe me that we can measure this but essentially this activity drops very very fast if you have an assay for activity of the APC over distance of about six or eight micron it's almost it's gone down at least three, four, four what that means is that because the nuclei are about 50 or micron away from the cortex essentially it is reasonable to imagine that the degradation is really only happening around the nuclei so what is the problem with this model and why we were a little unhappy in thinking about this was that actually a lot happens during cell division and one of the things that happens into cell division is a lot of the process that are happening in nuclei are coupled with what happens in the cytoplasm so the cytoplasm is essentially what all cells together and so if you look at a cell it's really not just like a bag of water it has all this structure that really give it its integrity and its mechanical properties and one of the major component here is the active network is this a protein that forms filament collecting and here you have a stain and you can really, really see what beautiful structure it forms and this structure it could also be controlled by cross linkers but also by some molecular motors called myosin which can like contract this mesh you have this polymer mesh that can be contracted by sliding this filament across by myosin so what we did was already done before we repeated some experiments in which we are now actually looking at myosin level so this contra tile is a machine that can contract the active network and we look at it at the surface so remember the nuclei are inside I gave you some good argument of which probably protein degradation only happens around here yet I'll now show you that there are cortical events that appear to be coupled to the cell cycle so how does the cortex in which cycling level and maybe CDK1 activity is not oscillating know what is going on inside and to show you that really there is oscillation you can look at myosin and what you'll see is like this it gets brighter and then it gets darker and then it gets brighter and I'll show you some quantification later but you really can see that there are oscillation and they are really on the right time scale for the cell cycle they show you how we see that this is actually coupled to the cell cycle what is the role of this contraction why do we care well there is an extra thing that the embryo needs to do the embryo wants to get the nuclei everywhere right so you start with a nucleus here and then you want this nuclei to spread to fill the embryo how do you spread them why don't I just divide and stay low around here and the idea is that this contraction I just showed you what they do is that essentially they create some vortices as they contract you push all this water around and they generate some vortices they push the nuclei out and this is what drives spreading nuclear spreading and and actually this coupling this coupling between cell cycle and morphogenesis is really a graph it's a phenomenon is observed also in starfish embryo frog embryos and this movie I just showing it really because it's beautiful but you can really it is a great paper from Bill Bement and George Bondasso in which they show that there is a coupling between all these really wave-like activities and we talk about waves on Friday that happens on the cortex with the cell cycle so we think this is a general phenomenon of embryogenesis of serving frog eggs so we wanted to do a little better here and what Victoria sitting in the back there came up with the idea was that she really wanted to be able to measure CDK1 activity so we don't want we really want to do the same biochemistry that people will do on an enzymatic assay but we want to do it in a living embryo so for that we use this technique called FRET and we actually this sensor had already been developed in mammalian cells so we were lucky enough for we had to do it was adapted into embryo so how does this work? so here now you need to know another fact about fluorescent protein is that if they come up with different spectra and if there is enough energy overlap what could happen is that if you excite a blue molecule this molecule instead of emitting a photon and giving you fluorescence if there is a molecule that has an overlapping spectra next to it because this is the excitement of a deep hole moment and this also has a deep hole moment you can have energy transfer so that this molecule like this excited molecule transfer energy to this other molecule so what happens is that once you illuminate it with blue light and they're very close you don't get a blue photon out from these but this molecule transfer excitation to this other molecule and you get actually a yellow photon back so if you makes a molecule in which these CFP or right cyan fluorescent protein and yellow fluorescent protein and their distance is a function of CDK1 activity you could actually measure how much activity is there how do you make it that their distance a function of CDK1 activity and the idea is you pick up a peptide which is phosphorylated by CDK1 so what will happen is that this peptide could be in two states and phosphorylated or modified by the enzyme you're interested in if this peptide is modified it undergoes a conformational change then this two molecule can be brought closer and essentially by measuring the ratio of how much YFP to CFP fluorescence you have you can kind of get a proxy of how much activity is there and this worked beautifully it's actually really exciting when we saw this happen so you can really look at this and what you see and now we can really measure CDK1 activity in a live embryo with two second resolution great spatial resolution so now we are in business if you want to do more quantitative studies because you really now have access to the quantity you really care about with great precision and in great quantitative terms and so what we usually do we just take the ratio of the two molecules and you can see it goes up and down if you knock down CDK1 through genetics or RNAi method so the activity disappears if you just mutate this site also the activity and the oscillation disappear at a time where CDK1 activity we know it's very low this is essentially activity is very low and our signal to noise or this amplitude compared to the noise level here is great so we can now really measure the things right so I'll show you on Friday that there is a great so what is happening if you have seen the movie and pay really attention there are ways my tossing starts at the pole and there is waves that travel this is all like through essentially an excitable system and we have measured all of these but we'll talk about it on Friday there isn't enough time to cover everything today so now what we decide to do was to look at these early cycles again and it had been very well characterized how the nuclei spread and this seems to be a very stereotypical process so what happens the nuclei start usually if you look at the anterior posterior axis so this is the anterior of the eggs we will form the head and this is the tail if you want and if you look around here about 30% closer to the anterior that's where usually the nuclei are and then the spread out and you can see actually this is the the black is the average of how much the spread out and the white is from this paper is the standard deviation so it's a very reproducible process so we decided to see how is this process correlated and how does the cortex talk about with the nuclei we can really only measure activity or mainly measure activity on the cortex and what we found was really interesting was that if we look at let's say cycle three where most of the nuclei are here and we look in a region where nuclei are what is happening is that the activity actually goes up and down you can clearly see an oscillation but if you look far away where there are no nuclei the activity is actually stays high and then as the nuclei expand you start getting actually about cycle five as nuclei invade this other region you start seeing that the activity starts high but then eventually it goes low so it doesn't start low and goes high and then low just starts high and then goes low and then eventually by the time you're filled out all the embryo everybody's oscillating so what we think is happening is that essentially the nuclei are providing a signal that is calling, it's talking to the cortex but this signal cannot spread throughout the embryo it's kind of localized we know that it's not the APC there is something else and this signal is really very well synchronized with the myosin so if we measure myosin and this contractile myosin we can really see that CDK1 activity and myosin are very much, this two oscillators that are very coupled and out of phase so somehow it is true that the cell cycle is talking to the myosin activity somehow has to do with nuclei so what we think is happening and this we still have to prove is that when in everything I've told you I missed an essential component I've been telling you that CDK1 is the master regulator and that's what you need to care about but now if you think about what mitosis really is what mitosis is you need to regulate a bunch of machinery and you regulate it by putting a phosphate on through CDK1 if there wasn't any enzyme that would take this phosphate off once you've entered mitosis there is no way you get out even if you get rid of CDK1 because the phosphate will stay on the reason why the phosphate doesn't stay on is because there are other enzymes which are called phosphatases that push it back up so is it possible that what is you really not only have an oscillation in the CDK1 you also have an oscillator in the phosphatase and that's what's spreading the signal that's what our current model is we haven't really finished proving it but what I'll show you is that we think we know what the phosphatase is so we could inject an inhibitor a product called inhibitor 2 which blocks a product in phosphatase 1 and if you do that what you'll see is that the embryo or the nucleon embryo will enter the cell cycle thinking that everything is fine and is normal they'll get into mitosis but they'll never defosphorylate the sensor so our view now is that you only do great cyclins around the nuclei but that generates the activity increase the activity of the phosphatase and this is known that Pp1 could be down-regulated by CDK1 and the cell cycle regulated but this is we think that this is the activity that spreads and that gives you the coupling so yes you could actually I'm doing this for sake of simplicity if I do this same plot we see the exact same phenomenon and this is the reason so if you look at this oscillation here this will be cycle 2 and 3 they're really not as good as later on but what you see is that if you were to look at the region where the nuclei you will have a nice oscillation but because we are averaging with also the region where the nuclei are and the activity is not really oscillating it looks a bit more messy so this view that as CDK1 activity all the oscillates where the nuclei are so that's myosin so we think that you are talking to each other we don't really know exactly the mechanism right so this is what I'm trying to say so these early cycles this signal is a little messy because out of laziness and out of try to keep the message a little simpler but you guys are really perspective is that here you do not every but not all the cytoplasm is oscillating and I'm averaging over everybody so what is happening here is that this region is oscillating by this season and when I average this region together with this region the signal get messy if I showed you only for this region where the nuclei are you'll see this beautiful exact relationship so it's a good point but the correlation is there it's just the way I'm plotting the data doesn't make it obvious okay so I made it prepared a little more than what I might be able to cover so now this is what happens when the nuclei are inside then they all come to the surface and now what happens as they come to the surface and as he was pointing out it looks like the cell cycle is getting longer right so what is happening with that and again now you'll see this beautiful spatial organization and I promise I'll tell you all about how that works on Friday and I know it's hard to miss and not to be fascinated by those waves of mitosis but what you should also see is that if you have been paying attention at the pace or you look at the clock this cell cycle now will be a little longer and there will be one last cell cycle so this cell cycle was about 12 minutes and now there will be another one that will actually be 18 minutes the early cycles are only eight minutes and progressively you get to 18 minutes and we actually think that this really is this amazing speed of the early fly embryo doing cell cycle all the eight minutes is the reason why they don't destroy cycling fully and re-synthesize it is that there won't be enough time if you really want to divide every eight minutes you're much better off controlling the cycle through signaling because it's much easier to reverse signal you don't need to wait to remake products every two seconds yeah yeah so you need the fast imaging they are dividing so this is exactly what you see if you plot CDK1 activity now not only a function of time but space and time what you'll see is that this spread out so these last two cell cycles are longer why are they longer? I told you that how do you define the embryo? what? how do you define the embryo? oh just the middle of the embryo I mean the thing is that if you think of an embryo and the way you image it you are really only imaging around here so we don't really know where the anterior is and the posterior is so rather than have arbitrary coordinates we took the middle of what we image it's just some coordinates, right? so now you could think that maybe they are accumulating CDK1 activity is lower that's why the cell cycle is longer and what we found was that actually if you look at the activity as they enter mitosis that's the same for all the cell cycle what is changing and the reason why the cell cycle are getting longer is that what you do is that you accumulate slower in S phase and then still rapidly mitosis so what is happening? why are these S phase accumulating CDK1 faster and these one are not and the reason for that is that as you make more and more nuclei DNA content is increasing as you make more and more DNA content it becomes very difficult for the embryo to replicate all these DNA so quickly because you just need so much machinery that needs to be dedicated to make new DNA from DNA and so what happens is that the embryo can keep up with it but you don't want to enter mitosis before you have done replicating your DNA that will be disastrous so cells have developed what are called checkpoint or DNA replication checkpoint these are mechanisms that prevent a cell to enter mitosis before you have done replicating the DNA if you have any problems store replication for DNA damage it activates a signaling cascade that talk to this protein check 1 and check 2 and they stop the cell cycle and probably you're not surprised to hear that what check 1 does it controls CDK1 so if I'm now correct that the mechanism goes through check 1 and this was already known also from the work of other people in the fly embryo if it was true if I have a genetic manipulation that we do have to delete this check 1 you should lose this biphasic activation of CDK1 and this is exactly what you observe so if you look at check 1, check 2 mute and what you see is that they keep going up and down fast they never become biphasic and the cell cycle doesn't lengthen so essentially now the embryo doesn't know that there is DNA damage and they hasn't finished replicating his DNA so it's just gonna keep trying on dividing so this is the reason why they get longer and the cell cycle get longer there's too much DNA and I don't know that we can skip this for now for sake of simplicity and what we could also get rid of these other feedbacks and what we can show is that these other feedbacks in the system are what controls the rate of mitosis but not of S phase I could synthesize this in a simple model in which we think that essentially the balance of cycling CDK1 to phosphatase it will control the early cell cycles the activity of the checkpoint is what controls why the cell cycle get longer at the end and then these feedback are really required to make sure that you always do mitosis really, really fast but they are not really timing the cell cycle so now there's one last thing I will quickly talk about before we are done and this has to do, how do they count to 13? Why do they divide 13 times and they stop? And you had a question everyone asked that oscillation I was showing you is mainly outside of the nucleus you can measure if you compare activity inside and outside the nucleus they are actually very similar so we think that the cytoplasmic shadowing of this activity is very fast so essentially nuclear and cytoplasm are essentially equilibrated so how does the embryo count to 13 and how could you think about a mechanism in which you do 13 division and you stop well they may have a clock and it's not that they are counting they just be like after two hours and 20 minutes stop dividing another mechanism they may count the number of cell divisions they somehow know how many times they divide but another idea is maybe they know how much DNA they've got relative to cytoplasm and this is actually the way they do it and this was shown in a very beautiful experiment by Newport and Kirchner and this is with Xenopus egg what you do here they took actually a single layer and they constricted the egg so you could go in with the air and if you are very very careful you can constrict it now you have generated two alps and what will happen is that this cell will divide and eventually at one time you'll get lucky enough that the chromosome plate aligns in such a way that one nucleus end up here so now the model I've told you before make very different prediction if you had a clock these alps and that alps will stop dividing at the same time because they will have the same clock similarly if they count in number of divisions somehow they will count all together and they will also stop at the same time but if they are measuring how much DNA they have versus cytoplasm these alps will always have more cell than that so these alps should get that earlier and stop earlier and this will stop later and that's exactly what they observed so somehow the egg knows how much DNAs got relative to cytoplasm the same is true in Drosophila so Drosophila if you look at diploid embryo or you can make through genetics an diploid embryo this embryo or instead of doing 30 division that's 14 because every nucleus has half of DNA content so to get to total amount of DNA has to divide one more time you can test that even more carefully with genetics in flies and the way you can do it when you usually have myosis as I show you you generate chromosome but what you usually get you get one chromosome from your father and one from your mother and you end up with two copies and you're fine but in fly like all some old fly geneticists devise this chromosome in which instead of having two you only have one all the genetic content is fused into one big chromosome and now you either get this chromosome or you get nothing so half of the time you get one from your mother and none from your father or the other way around and the flies fine but about quarter of the time you get too much or you get too little and if you do a plot of how many cell division embryo does as a function of DNA content it looks that this idea holds so if you have 50% of what the wild type is that you do 14 division and you stop a cycle 15 but as you get about 70% of a wild type DNA content a cycle 14 now you see there's a sharp transition so now you do 13 division something funny here but this sort of holds so what is the idea that people have very quickly about this and some people have test biochemical way the idea is that essentially there is some factor in the cytoplasm that can bind DNA that is important for cell cycle processes or transcriptional processes and it's tight rated out as you get more and more DNA the idea is very simple it will work like this when you have a little bit of DNA a lot of these factors there's a lot as you start getting more DNA and eventually you get so much DNA that they get tight rated out a lot of different people and work on finding these factors and there's a lot of different molecules that have come out some people have shown that nucleotide may be important so really machinery required for DNA replication at some point they seem to have such an abundance of nuclei it gets harder to transcribe your to replicate your DNA some other people found enzyme important in DNA replication instance so this maybe suggest there's a lot of components that are right at the right balance and they may be getting tight rated out similarly the DNA replication checkpoint gets activated which is required so if you get rid of the DNA replication checkpoint not only you lose that biophasic behavior as you can see here the embryo do 2x recel cycle so you need the DNA replication checkpoint to count to 13 otherwise they'll go up to 15 and this DNA replication checkpoint or a proxy for how much activity of the DNA replication checkpoint we have is how much of this lobe is this lobe or how much this lobe is reduced a function of DNA content and what we could show is that it is so if we generate use this we are chromosome I told you before to generate the embryo of different DNA content you can show that how much you inhibit CDK1 in S phase which is a proxy of checkpoint activity it's really a function of how much DNA is in the embryo however there is also some exception to this rule and is there not all chromosomes are the same so if you do it with chromosome 3 versus chromosome 2 you get a slightly different answer if you do with chromosome X versus chromosome Y this is from a beautiful paper from Shelby and Eric Wieschaus you'll and probably Eric will talk about some of this work later in the school you also get a different answer and Shelby did a really really cool experiment in which he looked at how many genes are transcribed in this region and what he realized was that essentially you have that Y chromosome is so-called heterochromatic this is really a crumbled up chromosome there's very few genes on it so you don't get much transcription out of this chromosome but you get a lot more transcription out of the X chromosome it was if instead of plotting this as a function of how much DNA you got in the embryo I plot it as a function of how many genes are transcribing what would happen and the curve that before split now falls on a perfect monotonic line which really made it suggest that there is some feedback from transcription so he thinks that transcription and as other experiments also supporting this transcription essentially as you start transcribing you are trying to transcribe the same DNA you are trying to replicate this tumor machinery bump into each other and that is seen as a stress from the cell and triggers the DNA replication checkpoint which is important for triggering the MBT the converse though is also true, right? you because transcription stops during mitosis if you want to accumulate transcript zygotically you need the cell cycle to be longer so this is just a way and this is actually from Ernangers here when he was a postdoc in Thomas Gregorlam and should Thomas wish you along more of these tomorrow or Wednesday but there are ways to visualize a marine vivo in which you can force your marinate to encode certain secondary structure or stem loops then then combine GFP and to make a long story short what will happen is that a lot of GFP is recruited at the low side that are transcribing so if you image an embryo what you'll see you'll see bright foci of GFP when a gene is on and all I wanted to show you is this movie and what you'll see in this movie is that transcription is on and that shuts off completely mitosis the foci disappear so mitosis is usually incompatible because you get all this chromosome condensation and so many structural changes usually incompatible from transcription so as you start activating transcription as you get close to the MBT you want your cell cycle to get longer otherwise you just don't have enough time to accumulate transcripts because you destroyed them before you have made it so there's all feedback between what is going on trying to activate transcription and lengthening the cell cycle and before the last couple of slides going back to the cell cycle what is happening molecularly at the level of CDK1 regulation that is telling the embryo when to stop and what we show is that this is again the post-translation of the inhibitor and the activator I told you before so this is what we do before I call CDC25 this is what activates CDK1 and we want is more repressed and what we could show was that essentially the stability of this CDC25 molecule of which twine is the major one is highly regulated so what is happening is whatever the NC ratio is doing and this decision is being made eventually it talks to this protein called twine so what will happen is that you get this last division and then you'll see that this protein will disappear very, very quickly so there is a mechanism essentially as the embryo knows that it's done 13 division it gets rid of this activator and you can see that it's gone and you can show that that is important because if you find a mutant in which that activator is not degraded then what will happen is that the embryo will get confused and will not know that it needs to so now this embryo is doing its normal divisions and now it gets to cycle as done 13 division it will want to stop but because it can get rid of this protein you get a big chunk of the embryo that does an extra division and this is bad this is throwing this nuclei out of schedule with those ones and this embryo will die so you really need to degrade this is like really whatever the NC ratio is doing and I will say that we really don't know yet how it's sensed is talking to the cell cycle machinery to this protein twine and we know that that needs transcription and we actually the way we know that that is true is that again using another and this is the last thing I want to show it's like another really weird flight genetic tools this time instead of people fusing two two chromosome into one what they did they took the two left arm of a chromosome and fuse together and the two right arm and fuse together when you cross these flies you will end up with one flight that only get left arm or only get right arm and so you can do this type of genetics and what happens if you do this for chromosome 3 fruit fly embryo only have four chromosomes so this is relatively easy to do and if you have flies that lack the left arm of chromosome 3 again now you find so they are unable to activate transcription they have normal wild type content of DNA so they should be in term of the NC ratio be fine but they don't have the enzyme that degrade twine and instead of stopping dividing they will do an extra division so this really puts the regulation of this phosphate and so you'll see now probably about now that it's they're all entering into cell division so so this is how the NC ratio somehow talks to gene that are on this chromosome to tell the cell cycle to stop so to conclude I think I try to make an argument that early embryogenesis and one of the major feature of early embryonic development is rapid proliferation and now this rapid proliferation must be coupled to site of skeletal rearrangement and now there may be specialized mechanism to do this because one of the issues and the interesting biology of the embryo is that it's big and dealing with such a really huge cells requires some signaling mechanism that often we don't see in somatic cells that have a small size and I think it's revealing a lot of really interesting special temporal organization and principle by which you can deal with like large cells or large tissues and finally somehow this maternal to zygotic transition or this decision of the embryo when to activate the DNA replication checkpoint and finally stop dividing at the MBT is controlled by they somehow measure how much cytoplasm to DNA they've got and that eventually talks to the DNA replication checkpoint machinery so a lot of the work I covered has been done by other people but almost all the original work from my lab was done by Victoria she's an HHMI fellow graduate student in my lab and she's sitting right there so you can talk to her also and this pretty much